Neural network based physical condition evaluation of electronic devices, and associated systems and methods

Information

  • Patent Grant
  • 11798250
  • Patent Number
    11,798,250
  • Date Filed
    Tuesday, February 18, 2020
    4 years ago
  • Date Issued
    Tuesday, October 24, 2023
    7 months ago
Abstract
Systems and methods for evaluating the physical and/or cosmetic condition of electronic devices using machine learning techniques are disclosed. In one example aspect, an example system includes a kiosk that comprises an inspection plate configured to hold an electronic device, one or more light sources arranged above the inspection plate configured to direct one or more light beams towards the electronic device, and one or more cameras configured to capture at least one image of a first side of the electronic device. The system also includes one or more processors in communication with the one or more cameras configured to extract a set of features of the electronic device and determine, via a first neural network, a condition of the electronic device based on the set of features.
Description
TECHNICAL FIELD

The present technology is generally directed to evaluating the condition of mobile phones and/or other electronic devices, such as evaluating the presence, quantity, and/or distribution of surface scratches or cracks in such devices, based on machine learning techniques.


BACKGROUND

Consumer electronic devices, such as mobile phones, laptop computers, notebooks, tablets, MP3 players, etc., are ubiquitous. Currently, there are over 6 billion mobile devices in use in the world; and the number of these devices is growing rapidly, with more than 1.8 billion mobile phones being sold in 2013 alone. There are now more mobile devices in use than there are people on the planet. Part of the reason for the rapid growth in the number of mobile phones and other electronic devices is the rapid pace at which these devices evolve, and the increased usage of such devices in third world countries.


As a result of the rapid pace of development, a relatively high percentage of electronic devices are replaced every year as consumers continually upgrade their mobile phones and other electronic devices to obtain the latest features or a better operating plan. According to the U.S. Environmental Protection Agency, the U.S. alone disposes of over 370 million mobile phones, PDAs, tablets, and other electronic devices every year. Millions of other outdated or broken mobile phones and other electronic devices are simply tossed into junk drawers or otherwise kept until a suitable disposal solution arises.


Although many electronic device retailers and cell carrier stores now offer mobile phone trade-in or buyback programs, many old mobile phones still end up in landfills or are improperly disassembled and disposed of in developing countries. Unfortunately, however, mobile phones and similar devices typically contain substances that can be harmful to the environment, such as arsenic, lithium, cadmium, copper, lead, mercury, and zinc. If not properly disposed of, these toxic substances can seep into groundwater from decomposing landfills and contaminate the soil with potentiality harmful consequences for humans and the environment.


As an alternative to retailer trade-in or buyback programs, consumers can now recycle and/or sell their used mobile phones using self-service kiosks located in malls, retail stores, or other publicly accessible areas. Such kiosks are operated by ecoATM, LLC, the assignee of the present application, and are disclosed in, for example, U.S. Pat. Nos. 8,463,646, 8,423,404, 8,239,262, 8,200,533, 8,195,511, and 7,881,965, which are commonly owned by ecoATM, LLC and are incorporated herein by reference in their entireties.


It is often necessary to visually evaluate the physical and/or cosmetic condition of an electronic device. For example, pricing the electronic device, assessing the electronic device for possible repair, and evaluating the electronic device for warranty coverage all can require identification of scratches, cracks, water damage, or other cosmetic defects in the device's screen and/or in non-screen portions of the device. Individualized manual inspection of devices can be slow, cumbersome, and can yield inconsistent results among devices. There remains a need for more efficient technologies for evaluating the physical and/or cosmetic condition of electronic devices.





BRIEF DESCRIPTION OF THE DRAWINGS


FIG. 1A is a schematic illustration of a representative operating environment having elements configured in accordance with some embodiments of the present technology.



FIGS. 1B-1E are a series of isometric views of the kiosk shown in FIG. 1A with the housing removed to illustrate selected internal components configured in accordance with some embodiments of the present technology.



FIG. 2 is a flowchart illustrating a method for evaluating the cosmetic condition of electronic devices in accordance with some embodiments of the present technology.



FIG. 3 illustrates an example neural network that can be implemented in accordance with some embodiments of the present technology.



FIG. 4A illustrates examples of pre-processed images showing a front side of a smartphone in accordance with some embodiments of the present technology.



FIG. 4B illustrates other examples of pre-processed images showing a front side of a smartphone in accordance with some embodiments of the present technology.



FIG. 5 is a flowchart illustrating a method for training a neural network for evaluating the cosmetic condition of electronic devices in accordance with some embodiments of the present technology.



FIG. 6 is a block diagram illustrating an example architecture for a computer system that can be utilized to implement various portions of the present technology.



FIG. 7 is a flowchart representation of a method for evaluating a physical condition of an electronic device in accordance with some embodiments of the present technology.



FIG. 8 illustrates an example architecture of a system for examining consumer devices and providing offer prices in accordance with some embodiments of the present technology.



FIG. 9A illustrates a side view of an example arrangement of light sources 901a,b in the upper chamber in accordance with one or more embodiments of the present technology.



FIG. 9B illustrates an example arrangement of two sets of light sources in accordance with some embodiments of the present technology.



FIG. 10 illustrates an example of evaluating an electronic device using another mobile device in accordance with some embodiments of the present technology.



FIG. 11 illustrates an example architecture of training a neural network in accordance with some embodiments of the present technology.





DETAILED DESCRIPTION

The present disclosure describes various embodiments of systems and methods for evaluating the cosmetic and/or physical condition of mobile phones and/or other electronic devices using machine learning techniques. As described in greater detail below, in some embodiments these systems and methods can be implemented by a consumer operated kiosk to evaluate whether, for example, a display screen of a mobile phone is cracked or otherwise damaged.


Efficiently and consistently evaluating the cosmetic condition of electronic devices can be challenging. For example, manual identification of defects as shown in images of electronic devices can be costly, tedious, and subject to variability among different inspectors or even the same inspector. The manual process can also be inaccurate in many cases. For example, when the screen of the device is on, a human inspector can not be able to differentiate cosmetic defects from the background image shown on the device. As another example, a screen protector or case attached to the device can make manual inspection difficult. In this regard, certain feature- or rule-based automatic pattern recognition methods can not provide satisfactory and consistent evaluation results either. Additionally, the evaluation of cosmetic condition can not be limited to the identification of a pre-defined set of defects (e.g., scratches, cracks, dents, water damage, and/or bad pixels). Rather, the evaluation can correspond to a comprehensive, overall “look and feel” of an electronic device, such as identifying whether a device is a counterfeit product. Therefore, predefined feature- or rule-based methods can be inefficient and/or insufficient to handle various cosmetic evaluation scenarios.


Aspects of the present technology use machine learning techniques (artificial neural networks (ANNs) in particular) to perform cosmetic condition evaluation based on images of electronic devices, without predetermined feature(s) or rule(s). Among other things, the use of ANN(s) as described herein contributes to various advantages and improvements (e.g., in computational efficiency, detection accuracy, system robustness, etc.) in processing images of electronic devices. As those skilled in the art would appreciate, ANNs are computing systems that “learn” (i.e., progressively improve performance on) tasks by considering examples, generally without task-specific programming. For example, in image recognition, ANN can learn to identify images that contain cats by analyzing example images that have been manually labeled as “cat” or “no cat” and using the results to identify cats in other images.


An ANN is typically based on a collection of connected units or nodes called artificial neurons. Each connection between artificial neurons can transmit a signal from one artificial neuron to another. The artificial neuron that receives the signal can process it and then signal artificial neurons connected to it. Typically, in ANN implementations, the signal at a connection between artificial neurons is a real number, and the output of each artificial neuron is calculated by a non-linear function of the sum of its inputs. Artificial neurons and connections typically have a weight that adjusts as learning proceeds. The weight increases or decreases the strength of the signal at a connection. Artificial neurons can have a threshold such that only if the aggregate signal crosses that threshold is the signal sent. Typically, artificial neurons are organized in layers. Different layers can perform different kinds of transformations on their inputs. Signals travel from the first (input) to the last (output) layer, possibly after traversing the layers multiple times.


In some embodiments, one or more ANNs used by the present technology includes convolutional neural network(s) (CNN or ConvNet). Typically, CNNs use a variation of multilayer perceptrons designed to require minimal pre-processing. CNNs can also be shift invariant or space invariant artificial neural networks (SIANN), based on their shared-weights architecture and translation invariance characteristics. Illustratively, CNNs were inspired by biological processes in that the connectivity pattern between neurons resembles the organization of the animal visual cortex. Individual cortical neurons respond to stimuli only in a restricted region of the visual field known as the receptive field. The receptive fields of different neurons partially overlap such that they cover the entire visual field.



FIGS. 1A-E illustrate details about a kiosk model in accordance with some embodiments of the present technology. FIG. 1A illustrates an example kiosk 100 for recycling, selling, and/or other processing of mobile phones and other consumer electronic devices in accordance with some embodiments of the present technology. In some embodiments, at least some portions of the technology described herein can be carried out using a kiosk that includes an imaging device therein. For example, the kiosk can process and evaluate images received from the imaging device. The kiosk can include, for example, a processing component (e.g., including one or more physical processors) and memory storing instructions that, when executed by the processing component, perform at least some operations described herein. The term “processing” is used herein for ease of reference to generally refer to all manner of services and operations that can be performed or facilitated by the kiosk 100 on, with, or otherwise in relation to an electronic device. Such services and operations can include, for example, selling, reselling, recycling, donating, exchanging, identifying, evaluating, pricing, auctioning, decommissioning, transferring data from or to, reconfiguring, refurbishing, etc., mobile phones and other electronic devices. Although many embodiments of the present technology are described herein in the context of mobile phones, aspects of the present technology are not limited to mobile phones and can generally apply to other consumer electronic devices. Such devices include, as non-limiting examples, all manner of mobile phones; smartphones; handheld devices; personal digital assistants (PDAs); MP3 or other digital music players; tablet, notebook, ultrabook, and laptop computers; e-readers all types; GPS devices; set-top boxes; universal remote controls; wearable computers; etc. In some embodiments, it is contemplated that the kiosk 100 can facilitate selling and/or otherwise processing larger consumer electronic devices, such as desktop computers, TVs, game consoles, etc., as well smaller electronic devices such as Google® Glass™, smartwatches (e.g., the Apple Watch™, Android Wear™ devices such as the Moto 360®, or the Pebble Steel™ watch), etc. The kiosk 100 and various features thereof can be at least generally similar in structure and function to the systems, methods and corresponding features described in the following patents and patent applications, which are incorporated herein by reference in their entireties: U.S. Pat. Nos. 10,127,647, 10,055,798; 10,032,140; 9,904,911; 9,881,284; 8,200,533; 8,195,511; 8,463,646; 8,423,404; 8,239,262; 8,200,533; 8,195,511; and 7,881,965; U.S. patent application Ser. Nos. 12/573,089; 12/727,624; 13/113,497; 12/785,465; 13/017,560; 13/438,924; 13/753,539; 13/658,825; 13/733,984; 13/705,252; 13/487,299; 13/492,835; 13/562,292; 13/658,828; 13/693,032; 13/792,030; 13/794,814; 13/794,816; 13/862,395; 13/913,408; U.S. patent application Ser. No. 14/498,763, titled “METHODS AND SYSTEMS FOR PRICING AND PERFORMING OTHER PROCESSES ASSOCIATED WITH RECYCLING MOBILE PHONES AND OTHER ELECTRONIC DEVICES,” filed by the applicant on Sep. 26, 2014; U.S. patent application Ser. No. 14/500,739, titled “MAINTAINING SETS OF CABLE COMPONENTS USED FOR WIRED ANALYSIS, CHARGING, OR OTHER INTERACTION WITH PORTABLE ELECTRONIC DEVICES,” filed by the applicant on Sep. 29, 2014; U.S. patent application Ser. No. 14/873,158, titled “WIRELESS-ENABLED KIOSK FOR RECYCLING CONSUMER DEVICES,” filed by the applicant on Oct. 1, 2015; U.S. patent application Ser. No. 14/506,449, titled “SYSTEM FOR ELECTRICALLY TESTING MOBILE DEVICES AT A CONSUMER-OPERATED KIOSK, AND ASSOCIATED DEVICES AND METHODS,” filed by the applicant on Oct. 3, 2014; U.S. patent application Ser. No. 14/925,357, titled “SYSTEMS AND METHODS FOR RECYCLING CONSUMER ELECTRONIC DEVICES,” filed by the applicant on Oct. 28, 2015; U.S. patent application Ser. No. 14/925,375, titled “METHODS AND SYSTEMS FOR FACILITATING PROCESSES ASSOCIATED WITH INSURANCE SERVICES AND/OR OTHER SERVICES FOR ELECTRONIC DEVICES,” filed by the applicant on Oct. 28, 2015; U.S. patent application Ser. No. 14/934,134, titled “METHODS AND SYSTEMS FOR EVALUATING AND RECYCLING ELECTRONIC DEVICES,” filed by the applicant on Nov. 5, 2015; U.S. patent application Ser. No. 14/964,963, titled “METHODS AND SYSTEMS FOR PROVIDING INFORMATION REGARDING COUPONS/PROMOTIONS AT KIOSKS FOR RECYCLING MOBILE PHONES AND OTHER ELECTRONIC DEVICES,” filed by the applicant on Dec. 10, 2015; U.S. patent application Ser. No. 14/568,051, titled “METHODS AND SYSTEMS FOR IDENTIFYING MOBILE PHONES AND OTHER ELECTRONIC DEVICES,” filed by the applicant on Dec. 11, 2014; U.S. patent application Ser. No. 14/966,346, titled “SYSTEMS AND METHODS FOR RECYCLING CONSUMER ELECTRONIC DEVICES,” filed by the applicant on Dec. 11, 2015; U.S. patent application Ser. No. 14/598,469, titled “METHODS AND SYSTEMS FOR DYNAMIC PRICING AND PERFORMING OTHER PROCESSES ASSOCIATED WITH RECYCLING MOBILE PHONES AND OTHER ELECTRONIC DEVICES,” filed by the applicant on Jan. 16, 2015; U.S. patent application Ser. No. 14/660,768, titled “SYSTEMS AND METHODS FOR INSPECTING MOBILE DEVICES AND OTHER CONSUMER ELECTRONIC DEVICES WITH A LASER,” filed by the applicant on Mar. 17, 2015; U.S. patent application Ser. No. 14/663,331, titled “DEVICE RECYCLING SYSTEMS WITH FACIAL RECOGNITION,” filed by the applicant on Mar. 19, 2015; U.S. provisional application No. 62/169,072, titled “METHODS AND SYSTEMS FOR VISUALLY EVALUATING ELECTRONIC DEVICES,” filed by the applicant on Jun. 1, 2015; U.S. provisional application No. 62/202,330, titled “METHODS AND SYSTEMS FOR INSPECTING MOBILE DEVICES AND OTHER CONSUMER ELECTRONIC DEVICES WITH ROBOTIC ACTUATION,” filed by the applicant on Aug. 7, 2015; and U.S. patent application Ser. No. 15/057,707, titled “METHODS AND SYSTEMS FOR INTERACTIONS WITH A SYSTEM FOR PURCHASING MOBILE PHONES AND OTHER ELECTRONIC DEVICES,” filed by the applicant on Mar. 1, 2016; U.S. patent application Ser. No. 15/176,975, titled “METHODS AND SYSTEMS FOR DETECTING SCREEN COVERS ON ELECTRONIC DEVICES,” filed by the applicant on Jun. 8, 2016. In some embodiments, the kiosk 100 can share many or all of the features of the kiosks disclosed and described in U.S. patent application Ser. No. 16/719,699, entitled “SYSTEMS AND METHODS FOR VENDING AND/OR PURCHASING MOBILE PHONES AND OTHER ELECTRONIC DEVICES,” filed on Dec. 18, 2019, U.S. patent application Ser. No. 16/788,169, entitled KIOSK FOR EVALUATING AND PURCHASING USED ELECTRONIC DEVICES, filed on Feb. 11, 2020, U.S. patent application Ser. No. 16/788,153, entitled “CONNECTOR CARRIER FOR ELECTRONIC DEVICE KIOSK,” filed on Feb. 11, 2020, and U.S. Provisional Application No. 62/950,075, entitled “SYSTEMS AND METHODS FOR VENDING AND/OR PURCHASING MOBILE PHONES AND OTHER ELECTRONIC DEVICES,” filed on Dec. 18, 2019. All the patents and patent applications listed in the preceding sentences and any other patents or patent applications identified herein are incorporated herein by reference in their entireties.


In the illustrated embodiment, the kiosk 100 is a floor-standing self-service kiosk configured for use by a user 101 (e.g., a consumer, customer, etc.) to recycle, sell, and/or perform other operations with a mobile phone or other consumer electronic device. In other embodiments, the kiosk 100 can be configured for use on a countertop or a similar raised surface. Although the kiosk 100 is configured for use by consumers, in various embodiments the kiosk 100 and/or various portions thereof can also be used by other operators, such as a retail clerk or kiosk assistant to facilitate the selling or other processing of mobile phones and other electronic devices.


In the illustrated embodiment, the kiosk 100 includes a housing 102 that is approximately the size of a conventional vending machine. The housing 102 can be of conventional manufacture from, for example, sheet metal, plastic panels, etc. A plurality of user interface devices is provided on a front portion of the housing 102 for providing instructions and other information to users, and/or for receiving user inputs and other information from users. For example, the kiosk 100 can include a display screen 104 (e.g., a liquid crystal display (LCD) or light emitting diode (LED) display screen, a projected display (such as a heads-up display or a head-mounted device), etc.) for providing information, prompts, and so on, to users. The display screen 104 can include a touch screen for receiving user input and responses to displayed prompts. In some embodiments, the kiosk 100 can include a separate keyboard or keypad for this purpose. The kiosk 100 can also include an ID reader or scanner 112 (e.g., a driver's license scanner), a fingerprint reader 114, and one or more cameras 116a-c (e.g., digital still and/or video cameras, identified individually as cameras). The kiosk 100 can additionally include output devices, such as a label printer having an outlet 110, and a cash dispenser having an outlet 118. Although not identified in FIGS. 1A-1E, the kiosk 100 can further include a speaker and/or a headphone jack for audibly communicating information to users, one or more lights for visually communicating signals or other information to users, a handset or microphone for receiving verbal input from the user, a card reader (e.g., a credit/debit card reader, loyalty card reader, etc.), a receipt or voucher printer and dispenser, as well as other user input and output devices. The input devices can include a touchpad, pointing device such as a mouse, joystick, pen, game pad, motion sensor, scanner, eye direction monitoring system, etc. Additionally, the kiosk 100 can also include a bar code reader, QR code reader, bag/package dispenser, a digital signature pad, etc. In the illustrated embodiment, the kiosk 100 additionally includes a header 120 having a display screen 122 for displaying marketing advertisements and/or other video or graphical information to attract users to the kiosk. In addition to the user interface devices described above, the front portion of the housing 102 also includes an access panel or door 106 located directly beneath the display screen 104. The access door can be configured to automatically retract so that the user 101 can place an electronic device (e.g., a mobile phone) in an inspection area 108 for automatic inspection, evaluation, and/or other processing by the kiosk 100.


A sidewall portion of the housing 102 can include a number of conveniences to help users recycle or otherwise process their mobile phones. For example, in the illustrated embodiment the kiosk 100 includes an accessory bin 128 that is configured to receive mobile device accessories that the user wishes to recycle or otherwise dispose of. Additionally, the kiosk 100 can provide a free charging station 126 with a plurality of electrical connectors 124 for charging a wide variety of mobile phones and other consumer electronic devices.



FIGS. 1B-1E illustrate a series of isometric views of the kiosk 100 with the housing 102 removed to illustrate selected internal components configured in accordance with some embodiments of the present technology. Referring first to FIG. 1B, in the illustrated embodiment the kiosk 100 includes a connector carrier 140 and an inspection plate 144 operably disposed behind the access door 106 as shown in FIG. 1A. In the illustrated embodiment, the connector carrier 140 is a rotatable carrousel that is configured to rotate about a generally horizontal axis and carries a plurality of electrical connectors 142 (e.g., approximately 25 connectors) distributed around an outer periphery thereof. In other embodiments, other types of connector carrying devices (including both fixed and movable arrangements) can be used. In some embodiments, the connectors 142 includes a plurality of interchangeable USB connectors configured to provide power and/or exchange data with a variety of different mobile phones and/or other electronic devices. In operation, the connector carrier 140 is configured to automatically rotate about its axis to position an appropriate one of the connectors 142 adjacent to an electronic device, such as a mobile phone 150, that has been placed on the inspection plate 144 for recycling. The connector 142 can then be manually and/or automatically withdrawn from the connector carrier 140 and connected to a port on the mobile phone 150 for electrical analysis. Such analysis can include, for example, an evaluation of the make, model, configuration, condition, etc.


In the illustrated embodiment, the inspection plate 144 is configured to translate back and forth (on, e.g., parallel mounting tracks) to move an electronic device, such as the mobile phone 150, between a first position directly behind the access door 106 and a second position between an upper chamber 130 and an opposing lower chamber 132. Moreover, in this embodiment the inspection plate 144 is transparent, or at least partially transparent (e.g., formed of glass, Plexiglas, etc.) to enable the mobile phone 150 to be photographed and/or otherwise optically evaluated from all, or at least most viewing angles (e.g., top, bottom, sides, etc.) using an imaging device 190 (e.g., one or more cameras) mounted to or otherwise associated with the upper and lower chambers 130 and 132. When the mobile phone 150 is in the second position, the upper chamber 130 can translate downwardly to generally enclose the mobile phone 150 between the upper chamber 130 and the lower chamber 132. The upper chamber 130 is operably coupled to a gate 138 that moves up and down in unison with the upper chamber 130.


In some embodiments, the imaging device 190 can include one or more cameras disposed within both the upper chamber 130 and the lower chamber 132 to capture images of top and bottom surfaces of the mobile device 150 in order to detect cracks and/or scratches in the screen. The upper chamber 130 and/or the lower chamber 132 can include one or more light sources (e.g., spotlights) to allow the imaging device 190 to capture high quality images that demonstrate cosmetic defects on the mobile device 150.


In some embodiments, the one or more light sources are arranged in the upper chamber 130 and/or the lower chamber 132. FIG. 9A illustrates a side view of an example arrangement of light sources 901a,b in the upper chamber in accordance with one or more embodiments of the present technology. The light beams 911a,b from the light sources 901a,b form small angles (e.g., equal to or smaller than 60 degrees) with respect to the display of the mobile phone 150 to avoid direct reflection of the lights from the highly reflective display of the mobile phone 150. The relative positions between the one or more light sources 901a,b and the one or more cameras 921a,b of the imaging device 190 can be adjusted to ensure that the reflected light beams 913a,b from the mobile phone 150 can reach the cameras 921a,b. In some embodiments, kiosks can perform self-calibration to adjust the angles of the light sources to ensure that the correct angles are formed. In some embodiments, technicians can be dispatched periodically or upon request to perform calibrations of the kiosks.


In some embodiments, the one or more light sources includes two sets of light sources that are arranged orthogonal to each other. Because the cracks and/or scratches on the mobile device 150 can run in different directions (e.g., both horizontally and/or vertically), having two sets of orthogonally arranged light sources allows the cameras to capture various combinations of the cracks and/or scratches. For example, a first angle between light beams from one set of lights and the top side of the inspection plate 144 can be between 30 to 60 degrees (e.g., preferably 45 degrees) while a second angle between light beams from a second set of lights and the left side of the inspection plate 144 can be between 30 to 60 degrees (e.g., preferably 45 degrees). The two sets of lights are positioned orthogonal to each other. FIG. 9B illustrates an example arrangement of two sets of light sources in accordance with some embodiments of the present technology. A first set of light sources 931a,b is arranged orthogonally with a second set of light sources 941a,b. Light beams 951a,b from the first set of light sources 931a,b are about 45 degrees from either side of the inspection plate 144 (e.g., X axis and/or Y axis). Similarly, light beams 961a,b from the second set of light sources 941a,b are about 45 degrees from either side of the inspection plate 144 (e.g., X axis or/or y axis). Such arrangement can help reduce or eliminate imaging noise or shadows from other components of the kiosk 100 that are arranged along the sides of the inspection plate. In some embodiments, additional sets of light sources can be arranged within the upper and/or lower chamber to reveal damage that can not be visible from orthogonal arrangements of the light sources.


In some embodiments, the light beams from the one or more light sources can be collimated to produce more defined shadows of the cracks and/or scratches. In some embodiments, the one or more light sources support a wide range of brightness so that multiple sets of images can be taken at different light intensities with exposure times. For example, different devices can have different background colors (e.g., a white phone or a black phone), which can affect the processing of the captured images. Having at least two sets of images taken at different camera exposures, different light intensities, and/or different white balance settings can allow more accurate processing of cosmetic features of the device.


Because the mobile phone 150 is positioned on the transparent plate 144, light beams from light sources disposed in the lower chamber 132 undergo additional reflections within the transparent plate 144 before reaching the mobile phone 150, thereby impacting the quality of the captured images. Therefore, in some embodiments, all cameras of the imaging device 190 and the light sources are disposed within the upper chamber 130 only. The kiosk 100 can include a flipping mechanism 148 (e.g., a robot arm) to flip the mobile phone 150 so that images of both the top and bottom surfaces of the mobile phone 150 can be captured without any reflections between the cameras and the mobile phone 150.


Furthermore, to improve quality of the captured images, the color of the upper chamber 130 and the lower chamber 132 can be a middle gray, such as the 18% gray for calibrating light meters. A proper color of the chambers provides enough contract for glints over the display and shadows of hairline cracks of the mobile phone 150.


The images captured by the kiosk 100 can be transmitted to a qualified human operator to examine the quality of images as a measure of ensuring input quality to the computer-implemented visual analysis. Alternatively, the captured images can be transmitted to another neural network model to automatically determine the quality of the images and to provide feedback to the kiosks. If an operator or the neural network model determines that images captured by a particular kiosk routinely demonstrate certain defects (e.g., images are too dark, images are overexposed, etc.), technicians can be dispatched to re-calibrate the kiosk to ensure that uniform input images are obtained at different kiosks.


In some embodiments, the upper chamber 130 and/or the lower chamber 132 can also include one or more magnification tools, scanners (e.g., bar code scanners, infrared scanners, etc.), or other imaging components (not shown) and an arrangement of mirrors (also not shown) to view, photograph, and/or otherwise visually evaluate the mobile phone 150 from multiple perspectives. In some embodiments, one or more of the cameras and/or other imaging components discussed above can be movable to facilitate device evaluation. For example, as noted above with respect to FIG. 1A, the imaging device 190 can be affixed to a moveable mechanical component, such as an arm, which in turn can be moved using a belt drive, rack and pinion system, or other suitable drive system coupled to an electronic controller (e.g., the computing device). The inspection area 108 can also include weight scales, heat detectors, UV or infrared readers/detectors, and the like, for further evaluation of electronic devices placed therein. For example, information from the weight scales, UV, or infrared readers/detections can provide accurate information to facilitate the determination of the model of the mobile phone 150. The kiosk 100 can further include an angled binning plate 136 for directing electronic devices from the transparent plate 144 into a collection bin 134 positioned in a lower portion of the kiosk 100.


The kiosk 100 can be used in a number of different ways to efficiently facilitate the recycling, selling, and/or other processing of mobile phones and other consumer electronic devices. Referring to FIGS. 1A-1E together, in one embodiment, a user 101 wishing to sell a used mobile phone, such as the mobile phone 150, approaches the kiosk 100 and identifies the type of device (e.g., a mobile phone, a tablet, etc.) the user wishes to sell in response to prompts on the display screen 104. Next, the user can be prompted to remove any cases, stickers, or other accessories from the device so that it can be accurately evaluated. Additionally, the kiosk 100 can print and dispense a unique identification label (e.g., a small adhesive-backed sticker with a quick response code (“QR code”), barcode, or other machine-readable indicia, etc.) from the label outlet 110 for the user to adhere to the back of the mobile phone 150. After this is done, the door 106 retracts and opens allowing the user to place the mobile phone 150 onto the transparent plate 144 in the inspection area 108 as shown in FIG. 1B. The door 106 then closes and the transparent plate 144 moves the mobile phone 150 under the upper chamber 130 as shown in FIG. 1C. The upper chamber 130 then moves downwardly to generally enclose the mobile phone 150 between the upper and lower chambers 130 and 132, and the cameras and/or other imaging components in the upper and lower chambers 130 and 132 perform a visual inspection of the mobile phone 150. In some embodiments, the visual inspection of the mobile phone 150 includes performing at least a part of method 200 (as shown in FIG. 2), at least a part of method 500 (as shown in FIG. 5), and/or at least a part of method 600 (as shown in FIG. 6) to evaluate the physical and/or cosmetic condition of the mobile phone 150. In some embodiments, the visual inspection includes a computer-implemented visual analysis (e.g., a three-dimensional (3D) analysis) performed by a processing device within the kiosk to confirm the identification of the mobile phone 150 (e.g., make, model, and/or sub-model) and/or to evaluate or assess the condition and/or function of the mobile phone 150 and/or its various components and systems. For example, the visual analysis can include computer-implemented evaluation (e.g., a digital comparison) of images of the mobile phone 150 taken from top, side, and/or end view perspectives to determine length, width, and/or height (thickness) dimensions of the mobile phone 150. The visual analysis can further include a computer-implemented inspection of a display screen and/or other surface of the mobile phone 150 to check for, for example, cracks in the glass and/or other damage or defects in the LCD (e.g., defective pixels, etc.).


Referring next to FIG. 1D, after the visual analysis is performed and the device has been identified, the upper chamber 130 returns to its upper position and the transparent plate 144 returns the mobile phone 150 to its initial position near the door 106. The display screen 104 can also provide an estimated price, or an estimated range of prices, that the kiosk 100 can offer the user for the mobile phone 150 based on the visual analysis, and/or based on user input (e.g., input regarding the type, condition, etc., of the phone 150). If the user indicates (via, e.g., input via the touch screen) that they wish to proceed with the transaction, the connector carrier 140 automatically rotates an appropriate one of the connectors 142 into position adjacent the transparent plate 144, and door 106 is again opened. The user can then be instructed (via, e.g., the display screen 104) to withdraw the selected connector 142 (and its associated wire) from the carrousel 140, plug the connector 142 into the corresponding port (e.g., a USB port) on the mobile phone 150, and reposition the mobile phone 150 in the inspection area on the transparent plate 144. After doing so, the door 106 once again closes and the kiosk 100 (e.g., the kiosk CPU) performs an electrical inspection of the device via the connector 142 to further evaluate the condition of the phone, as well as specific component and operating parameters, such as the memory, carrier, etc. In some embodiments, the electrical inspection can include a determination of phone manufacturer information (e.g., a vendor identification number or VID) and product information (e.g., a product identification number or PID). In some embodiments, the kiosk 100 can perform the electrical analysis using one or more of the methods and/or systems described in detail in the commonly owned patents and patent applications identified herein and incorporated by reference in their entireties.


After the visual and electronic analysis of the mobile phone 150, the user 101 is presented with a phone purchase price via the display screen 104. If the user declines the price (via, e.g., the touch screen), a retraction mechanism (not shown) automatically disconnects the connector 142 from the mobile phone 150, the door 106 opens, and the user can reach in and retrieve the mobile phone 150. If the user accepts the price, the door 106 remains closed and the user can be prompted to place his or her identification (e.g., a driver's license) in the ID scanner 112 and provide a thumbprint via the fingerprint reader 114. As a fraud prevention measure, the kiosk 100 can be configured to transmit an image of the driver's license to a remote computer screen, and an operator at the remote computer can visually compare the picture (and/or other information) on the driver's license to an image of the person standing in front of the kiosk 100 as viewed by one or more of the cameras 116a-c as shown in FIG. 1A to confirm that the person attempting to sell the phone 150 is in fact the person identified by the driver's license. In some embodiments, one or more of the cameras 116a-c can be movable to facilitate viewing of kiosk users, as well as other individuals in the proximity of the kiosk 100. Additionally, the person's fingerprint can be checked against records of known fraud perpetrators. If either of these checks indicate that the person selling the phone presents a fraud risk, the transaction can be declined and the mobile phone 150 returned. After the user's identity has been verified, the transparent plate 144 moves back toward the upper and lower chambers 130 and 132. As shown in FIG. 1E, when the upper chamber 130 is in the lower position, the gate 138 permits the transparent plate 144 to slide underneath but not electronic devices carried thereon. As a result, the gate 138 knocks the mobile phone 150 off of the transparent plate 144, onto the binning plate 136 and into the bin 134. The kiosk can then provide payment of the purchase price to the user. In some embodiments, payment can be made in the form of cash dispensed from the cash outlet 118. In other embodiments, the user can receive remuneration for the mobile phone 150 in various other useful ways. For example, the user can be paid via a redeemable cash voucher, a coupon, an e-certificate, a prepaid card, a wired or wireless monetary deposit to an electronic account (e.g., a bank account, credit account, loyalty account, online commerce account, mobile wallet etc.), Bitcoin, etc.


As those of ordinary skill in the art will appreciate, the foregoing routines are but some examples of ways in which the kiosk 100 can be used to recycle or otherwise process consumer electronic devices such as mobile phones. Although the foregoing example is described in the context of mobile phones, it should be understood that the kiosk 100 and various embodiments thereof can also be used in a similar manner for recycling virtually any consumer electronic device, such as MP3 players, tablet computers, PDAs, and other portable devices, as well as other relatively non-portable electronic devices, such as desktop computers, printers, devices for implementing games, entertainment or other digital media on CDs, DVDs, Blu-ray, etc. Moreover, although the foregoing example is described in the context of use by a consumer, the kiosk 100 in various embodiments thereof can similarly be used by others, such as a store clerk, to assist consumers in recycling, selling, exchanging, etc., their electronic devices.



FIG. 8 illustrates an example architecture of a system 800 for examining consumer devices and providing offer prices in accordance with some embodiments of the present technology. The system 800 includes a capturing module 801 that captures information about consumer devices. The capturing module 801 can be implemented on a kiosk as described in connection with FIGS. 1A-1E. The capturing module 801 can capture device information 811 such as the device identifier (ID) of a consumer device, the time and/or the location that the consumer device is examined. The capturing module 801 can also capture images 813 of various surfaces of the device that demonstrate various features, such as cosmetic defect(s) of the consumer device, which can indicate the condition of the device. For example, images can be captured to show sides of the device, location or existence of buttons on the device, light emitted from the screen to indicate the LCD panel health. In some embodiments, images can be captured while the device is moving so as to capture the nature and extent of the damage. The images can also show depth of scratches and/or cracks to facilitate an estimation of the impact to underlying electronics. In some embodiments, the entire system 800 can be implemented on the kiosk 100.


The input information captured by the capturing module 801 is transmitted to a price prediction model 803 that is configured to determine candidate price for the input consumer device. The price prediction model 805 can extract features (e.g., scratches, hairline cracks, water damage marks) from the input information and determine the candidate price based on the number of cosmetic defects on the device. Alternatively and/or additionally, the capturing module 801 can extract features from the input information and transmit the extracted features to the price prediction model 803 to determine the candidate price based on the number of cosmetic defects on the device.


The system 800 also includes a pricing policy model 805 that accepts input from both the capturing module 801 and the price prediction model 805. The pricing policy model 805 can leverage various sub-models to generate a final offer price. The sub-models can include at least a sub-model to predict resale value, a sub-model to predict incoming volume of the consumer device, a sub-model to predict processing costs associated with the device, and/or other sub-models to facilitate the prediction process. Additional features that can affect the final offer price include the location of kiosk, the time at which the device was examined, the age of the device, the predicted repair costs, volume of devices in similar conditions, risk of counterfeit or fraud, the anticipated demand of the device, predicted resale channels, other electrical information retrieved from the device. These sub-models can locate centrally with the pricing policy model. The sub-models can also be distributed across different locations in a network as part of a cloud-based computing service. Each of the models and/or sub-models can be implemented using a neural network, such as CNN and/or ConvNet. As compared to human operators, the neural networks can produce more consistent analysis results across different geographical locations and are much more scalable when a large number of consumer devices need to be evaluated.


Upon customer's acceptance or rejection of the final offer price, the relevant data for this consumer device can be fed back to the price prediction model for further training and improvement of the model. As mentioned above, the capturing module 801 can be deployed in a kiosk while the other parts of the system are situated in a distributed manner in remote server(s). In some embodiments, the entire system can be deployed in a kiosk as described in detail in connection with FIGS. 1A-1E.


In some embodiments, instead of finding a kiosk to perform evaluation of a used consumer device (as discussed in connection with FIGS. 1A-E), the customer can download and install a software implementation of the capturing module 801 on another device (e.g., another mobile phone, tablet, wearable device, and so on). The software implementation of the capturing module 801 can provide a user interface to the customer to specify device information 811 (e.g., device ID, brand, model, etc.) and to capture images 813 of the target consumer device. FIG. 10 illustrates an example of evaluating an electronic device 1005 using another mobile device 1003 in accordance with some embodiments of the present technology. A customer 1001 can download a software application on his current mobile device 1003 (also referred to as the capturing device). The software application is configured to control one or more of a light source (e.g., a flash light) and/or camera(s) of the mobile device 1003 to capture at least one image of a target electronic device 1005. The customer 1001 can also be prompted to provide additional information about the target device 1005, such as device manufacturer, model, purchase date, general condition(s), device features, etc., via a user interface.


Referring back to FIG. 8, the input data (e.g., the captured images and/or additional device information provided by the customer) can be transmitted over a network to remote server(s) that host the price prediction model 803 and the pricing policy model 805 to determine the condition of the target device and/or a final offer price. Once the final offer price is determined, the capturing module 801 can display the final offer price of the target device on a user interface of the capturing device, and the customer can determine whether to accept or reject the offer price. Upon customer's acceptance or rejection of the final offer price, the relevant data for this consumer device can be fed back to the price prediction model 803 for further training of the model. If the customer accepts the offer price, the capturing module 801 can provide further instructions to package and mail the device to corresponding recycling and processing center(s).


To ensure image quality of the captured images, in some embodiments, the capturing module 801 can control the light source(s) of the capturing device to produce various light conditions. The capturing model 801 can further provide a set of predetermined settings or templates to guide the customer to take images of the target consumer device. Each setting or template can specify at least a desired angle to hold the capturing device with respect to the used consumer device, a desired exposure level, a desired light intensity, a desired white balance level, brightness, contrast, and/or other parameters. The predetermined templates help users to capture uniform input data to allow the system to generate consistent analysis results.


In some cases, network bandwidth limit can cause delays when a large amount of input date (e.g., a large set of images) needs to be transmitted to the remote server(s). To address such problems, some of the computation logic (e.g., pre-processing of the captured image) can be deployed locally on the capturing device. For example, a neural network that performs feature extraction to extract cosmetic defects (e.g., scratches, cracks, water marks, etc.) can be deployed on the capturing device as a part of the capturing module. Once the features are extracted, only the extracted features and information about the device (e.g., device ID, model, release date) are transmitted over the network to the prediction and policy models, thereby reducing bandwidth requirements for transmitting the relevant data.


In some embodiments, pre-processing of the images also includes operations, such as filtering, scrubbing, normalization, or the like, to generate preliminary features as input to feed into the neural network(s). As discussed above, pre-processing the captured images can alleviate network bandwidth limit for transmitting data in some embodiments. Pre-processing of the images can also be particularly useful for capturing modules that are deployed on customers' own devices because, unlike the kiosks, customers generally do not have accurate control of the cameras and positions of the devices. For example, pre-processing can adopt object detection algorithms to remove images that fail to include any consumer devices. Pre-processing of the images can also generate uniform inputs that are suitable for visual analysis by the neural networks so as to produce consistent results. For example, based on image segmentation techniques, an image of an electronic device can be cropped to show one side (e.g., front, back, top, bottom, or the like) of the electronic device. For the same device, cropped images showing different sides can be combined into a single image.



FIG. 2 is a flowchart illustrating a method 200 for evaluating the cosmetic condition of electronic devices, in accordance with some embodiments of the present technology. With reference to FIG. 2, the method includes feeding one or more images of an electronic device to a pre-processing module 210. In some embodiments, the image(s) can be obtained by the various camera(s) and/or other imaging component(s) of the kiosk 100 as described with reference to FIGS. 1A-1E or a capturing device owned by the customer. As describe above, the image(s) can be pre-processed to generate preliminary features. In some embodiments, the pre-processing can be performed by the processing component of the kiosk 100 or by the capturing device. In other embodiments, the image(s) can be transmitted to a remote system or device (e.g., a cloud-based computing service), and at least some or all of the pre-processing operations can be performed remotely. Illustratively, an image of an electronic device can be cropped to show one side (e.g., front, back, top, bottom, or the like) of the electronic device. Alternatively or in addition, the images can be taken under natural and/or controlled lighting. Still further, the images can be taken while the device is powered on or off. For the same device, cropped images showing different sides, images taken under different lighting, images taken while the device is on or off, and/or images of the device taken with other controlled/uncontrolled conditions can be combined into a single image.


The pre-processing can further include resizing an image (either an original image, combined image, or otherwise processed image) to a predefined size. The image is resized to provide a uniform input to the cosmetic evaluation neural network. The predefined size for neural network input can be determined in a manner that generally does not affect ability to detect cosmetic defects. For example, the predefined size must be sufficiently large so that damages or defects shown in an original image still appear in the resized image. Illustratively, each image can be resized to 299×299 pixels. In some embodiments, if the image is a color image, the present technology can separate out the red, green, and blue color spaces and convert the image into a three-dimensional integer matrix.


In some embodiments, if the image is a color image, the present technology can separate out the various color spaces (e.g., the red, green, and blue color spaces) and convert the image into a multi-dimensional (e.g., three-dimensional) integer matrix. For example, as used in standard RGB encoding, each value in the matrix is an integer ranging from 0-255. In some embodiments, the matrix can be rescaled by dividing by 255 to create a decimal value between 0 and 1 for each matrix entry.



FIGS. 4A and 4B illustrate examples of pre-processed images 400a-h for inputting into neural network(s) in accordance with some embodiments of the present technology. FIG. 4A illustrates a combined image showing the front side of a smartphone 402 under three different scenarios: lighting of a first white balance setting with the screen turned on 400a, lighting of a second white balance setting with the screen turned on 400b, the screen turned off 400c, and the back side of the smartphone 400d. The images do not show obvious scratches or hairline cracks, thus the smartphone 402 can be considered as in “cosmetically good” condition. FIG. 4B illustrates a combined image showing the front side of a smartphone 404 under three different scenarios: lighting of the first white balance setting with the screen turned on 400e, lighting of a second white balance setting with the screen turned on 400f, the screen turned off 400g, and the back side of the smartphone 400h. This combined image shows scratches on the screen of the smartphone 404, thus the smartphone 404 can be considered as in “cosmetically bad” condition.


Referring back to FIG. 2, the method 200 includes feeding the preliminary features 212 (e.g., original image, pre-processed image, or three-dimensional matrix depending on whether or how pre-processing is performed) into the neural network(s) 220. The neural network(s) can include the price prediction model and pricing policy model as shown in FIG. 1F. The method 200 further includes obtaining output 222 from the neural network(s) 220.


In some embodiments, the output of the neural network(s) includes an integer 0 or 1. Zero can represent “cosmetically good” (e.g., non-cracked, without significant scratches, or the like), and 1 can represent “cosmetically bad” (e.g., cracked, with significant scratches, or the like). In these embodiments, rescaling the inputs to a range between 0 and 1 can help the network train more consistently, as the inputs and outputs are more closely aligned. In some embodiments, instead of a binary value, the output of the neural network(s) can be a score of a range of values that indicate the severity of the damages on the consumer device. The output of the neural network(s) can also include at least one of a cosmetic rating or a category, a type of defect(s) detected, an orientation of defect(s) detected, a location of defect(s) detected, a size of defect(s) detected, associated confidence level(s), or other cosmetic evaluation indication. In some embodiments, the output of the neural network(s) can further include a brand, model, and/or type of the electronic device shown in the input image. Experimental results have demonstrated that the accuracy of neural network(s) in determining cosmetic defects can achieve 91%, which exceeds average human capacity (accuracy around 89.9%).


As discussed above, the neural networks 220 can be implemented as a part of the processing component of the kiosk 100 as described above with reference to FIGS. 1A-E or a user device. In other embodiments, at least some portion of the neural networks 220 can be implemented on a remote system or device (e.g., a cloud-based computing service). In these cases, the complete set of input date (e.g., images of the electronic device 202), the preliminary features 212, and/or certain intermediate data (e.g., the input/output between neural network layers) can be transmitted to the remote system or device for processing.



FIG. 3 illustrates an example neural network 300 that can be implemented in accordance with some embodiments of the present technology. The example neural network 300 can be a CNN or a modified CNN. The example neural network 300 can include two main types of network layers, namely, the convolution layer and the pooling layer. A convolution layer can be used to extract various features from the input to convolution layer. In particular, different kernel sizes can be applied in convolution layers for feature extraction to account for the fact that scratches and/or hairline cracks have various sizes. A pooling layer can be utilized to compress the features that are input to the pooling layer, thereby reducing the number of training parameters for the neural network and easing the degree of model over-fitting. The example neural network 300 can include multiple cascaded convolution and pooling layers that are connected with one another in various structural arrangements (e.g., serial connection). In some embodiments, the final layers of the network can include a layer of dense fully connected nodes, a dropout layer to mitigate overfitting, and/or one or more sigmoid activations to derive the final classification. In some embodiments, a sigmoid activation can be used for binary prediction (e.g., outputting values 0 and 1 indicating whether the condition of the device is acceptable). In some embodiments, other types of activation (e.g., a softmax activation) can be used so that the neural network can output different categories of predictions (e.g., “Fraud-Do Not Buy”, “Fake”, etc).



FIG. 11 illustrates an example architecture 1100 of training a neural network in accordance with some embodiments of the present technology. As shown in FIG. 11, the neural networks can be trained using pre-collected images 1101 which have been labeled by inspectors 1103 (e.g., human inspectors, electronic labeling systems, etc.). In some embodiments, images in the training set are each associated with a cosmetic evaluation indication (e.g., “cosmetically good” or “cosmetically bad”) agreed on by at least a threshold number of inspectors (e.g., two human inspectors). Therefore, the training set includes representative images of electronic devices in a particular cosmetic status that a threshold number of inspectors have agreed are, and the cosmetic status can be reasonably determined by visual inspection without requiring presence of the device phone on site.


The training set can include images that have been pre-processed the same way as would image(s) that contribute to the input of the machine learning system 1105 (e.g., neural network(s)) once it is deployed. The training set can include equal-sized or substantially equal-sized (e.g., within 5%, 10%, or 15% difference in size) subsets of images associated with each distinct cosmetic evaluation indication. For example, for approximately 700,000 images used in training, about 350,000 are associated with a “cosmetically good” indication and the other 350,000 are associated with a “cosmetically bad” indication. Dividing the training set in this manner can prevent or mitigate certain “random guess” effects of trained neural network(s), where an output can be biased to favor those reflected by a larger portion of the training set. In some embodiments, at least some of the images in the training set can be mirrored, rotated, or subject to other positional processing to generate additional images for inclusion in the training set.


The trained neural network 1105 can be validated using other pre-collected images which have been labeled by human inspectors 1103. Similar to the training set, a validation set can include subsets of images associated with each distinct cosmetic evaluation indication. In contrast with the training set, the relative sizes of the subsets are more consistent or otherwise reflect the real-world statistics of electronic devices that have previously been evaluated. Illustratively, approximately 300,000 images are used for validating the trained neural network.


In some embodiments, the machine learning system 1105 (e.g., neural network(s)) is deployed after successful validation (e.g., the false positive and/or false negative rate of the network's output over the validation set does not exceed predefined threshold(s)). Additional data, such as a portion of the captured images 1107 to the deployed network and/or associated outputs that have been verified by human inspectors, can be collected for further training of the neural network. In some embodiments, for each round of further training, layers closer (e.g., within a threshold number) to the input layer can be frozen while parameters of layers closer to the output can be adjusted. Doing so can help preserve concrete, basic aspects (e.g., representing small fractions of cracks in different orientations) already learned by the network while allowing the network to adjust parameters directed to more generalized, higher level features, which can efficiently adapt to newer models of devices, different lightings, and/or other changed scenarios. For example, the concrete, basic features learned when training on cracks for an iPhone 8 can still be applicable for detecting cracks on a Galaxy 9, even if the phones are different in size, shape, color, etc. In some embodiments, as shown in FIG. 11, a portion of the captured images can be directed to human inspectors 1103 to perform manual evaluation and/or generate more training data for the machine learning system 1105.



FIG. 5 is a flowchart illustrating a method 500 for training a neural network for evaluating the cosmetic condition of electronic devices in accordance with some embodiments of the present technology. In various embodiments, the method 500 can be performed by a remote system or device associated with the kiosk 100 as described with reference to FIGS. 1A-1E. With reference to FIG. 5, at block 510, the method 500 includes creating a training set including equally sized or similarly sized subsets of images associated with each distinct cosmetic evaluation indication.


The training set can include pre-collected images (e.g., those obtained by the kiosk 100) which have been labeled by human inspectors. In some embodiments, images in the training set are each associated with a cosmetic evaluation indication (e.g., “cosmetically good” or “cosmetically bad”) agreed on by at least two human inspectors. The images in the training set can be pre-processed the same way as would image(s) that contribute to the input of the neural network once it is deployed. The training set can include equal-sized or substantially equal-sized (e.g., within 5%, 10%, or 15% difference in size) subsets of images associated with each distinct cosmetic evaluation indication. In some embodiments, at least some of the images in the training set can be mirrored, rotated, or subject to other positional processing to generate additional images for inclusion in the training set.


In addition, the training set can include information about the devices (e.g., brand, model, release date) so that the model can be trained to identify damages that are specific to a particular set of devices.


At block 520, the method 500 includes training at least a portion of the neural network based on the training set. Illustratively, the training set provides “ground-truth” samples of network input and associated output (e.g., sample image(s) of an electronic device and associated cosmetic evaluation indication), and the components of the neural network can be trained in various ways as deemed proper by those skilled in the art. The parameters of the neural network can be learned through a sufficiently large number of training samples in the training set.


At block 530, the method 500 includes creating a validation set, including subsets of images associated with each distinct cosmetic evaluation indication, that are generally consistent in relative size as reflected in real world statistics. Similar to the training set, a validation set can include subsets of images associated with each distinct cosmetic evaluation indication. In contrast with the training set, the relative sizes of the subsets can be more consistent or otherwise reflect the real-world statistics of electronic devices that have previously been evaluated.


At block 540, the method 500 includes validating the trained neural network, and if successful, deploy the neural network. As described above, in some embodiments, each class of output is equally (or substantially equally) represented during training, but the ratio among the output classes is more consistent with field statistics during validation. Such arrangements can be a basis for determining that the trained network is not generally classifying every input in a particular direction (e.g., a particular cosmetic evaluation indication), and can still effectively extract cosmetic condition(s) that is less represented in the dataset.


The neural network can be deployed (e.g., to be executed on the kiosk 100 or as a part of the capturing module on a customer's device) after successful validation (e.g., the false positive and/or false negative rate of the network's output over the validation set does not exceed predefined threshold(s)). In some embodiments, the method 500 includes collecting additional data (e.g., inputs to the deployed network and associated outputs that have been verified by human inspectors) for further training of the neural network. This can be achieved by looping back to block 510 of the method. In some embodiments, for each round of further training, layers closer (e.g., within a threshold number) to the input layer can be frozen while parameters of layers closer to the output can be adjusted. Doing so can help preserve concrete, basic aspects (e.g., representing small fractions of cracks in different orientations) already learned by the network while allowing the network to adjust parameters directed to more generalized, higher level features, which can efficiently adapt to newer models of devices, different lightings, and/or other changed scenarios.



FIG. 6 is a block diagram illustrating an example of the architecture for a computer system 600 that can be utilized to implement various portions of the present technology. In FIG. 6, the computer system 600 includes one or more processors 605 and memory 610 connected via an interconnect 625. The interconnect 625 can represent any one or more separate physical buses, point to point connections, or both, connected by appropriate bridges, adapters, or controllers. The interconnect 625, therefore, can include, for example, a system bus, a Peripheral Component Interconnect (PCI) bus, a HyperTransport or industry standard architecture (ISA) bus, a small computer system interface (SCSI) bus, a universal serial bus (USB), IIC (I2C) bus, or an Institute of Electrical and Electronics Engineers (IEEE) standard 674 bus, sometimes referred to as “Firewire.”


The processor(s) 605 can include central processing units (CPUs) to control the overall operation of, for example, the host computer. In certain embodiments, the processor(s) 605 accomplish this by executing software or firmware stored in memory 610. The processor(s) 605 can be, or can include, one or more programmable general-purpose or special-purpose microprocessors, digital signal processors (DSPs), programmable controllers, application specific integrated circuits (ASICs), programmable logic devices (PLDs), or the like, or a combination of such devices.


The memory 610 can be or include the main memory of the computer system. The memory 610 represents any suitable form of random access memory (RAM), read-only memory (ROM), flash memory, or the like, or a combination of such devices. In use, the memory 610 can contain, among other things, a set of machine instructions which, when executed by processor(s) 605, causes the processor(s) 605 to perform operations to implement embodiments of the present technology. In some embodiments, the memory 610 can contain an operating system (OS) 630 that manages computer hardware and software resources and provides common services for computer programs.


Also connected to the processor(s) 605 through the interconnect 625 is a (optional) network adapter 615. The network adapter 615 provides the computer system 600 with the ability to communicate with remote devices, such as the storage clients, and/or other storage servers, and can be, for example, an Ethernet adapter or Fiber Channel adapter.


The techniques described herein can be implemented by, for example, programmable circuitry (e.g., one or more microprocessors) programmed with software and/or firmware, or entirely in special-purpose hardwired circuitry, or in a combination of such forms. Special-purpose hardwired circuitry can be in the form of, for example, one or more application-specific integrated circuits (ASICs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), etc. Systems implemented using the disclosed techniques can be deployed either centrally (e.g., the kiosks) or in a distributed manner (e.g., client device and remote servers) according to network resources, bandwidth cost, desired performance, etc.


Software or firmware for use in implementing the techniques introduced here can be stored on a machine-readable storage medium and can be executed by one or more general-purpose or special-purpose programmable microprocessors. A “machine-readable storage medium,” as the term is used herein, includes any mechanism that can store information in a form accessible by a machine (a machine can be, for example, a computer, network device, cellular phone, personal digital assistant (PDA), manufacturing tool, any device with one or more processors, etc.). For example, a machine-accessible storage medium includes recordable/non-recordable media (e.g., read-only memory (ROM); random access memory (RAM); magnetic disk storage media; optical storage media; flash memory devices; etc.). The term “logic,” as used herein, can include, for example, programmable circuitry programmed with specific software and/or firmware, special-purpose hardwired circuitry, or a combination thereof.



FIG. 7 is a flowchart representation of a method 700 for evaluating a condition of an electronic device in accordance with some embodiments of the present technology. The method 700 includes, at operation 710, capturing, by at least one camera of a kiosk, at least one image of a first side of the electronic device, wherein the kiosk includes multiple light sources. The method 700 includes, at operation 720, extracting, by a neural network, a set of features of the electronic device based on the at least one image of the electronic device. The method 700 also includes, at operation 830, determining a condition of the electronic device based on the set of features.


In some embodiments, the method includes capturing, via the at least one camera, at least one image of a second side of the electronic device that is different from the first side based on at least one lighting condition generated by the multiple light sources. Using different settings of the light sources and/or cameras to create different lighting conditions can facilitate the imaging of the scratches and/or hairline cracks. To image the second side of the electronic device, the method can include, prior to capturing the at least one image of the second side of the electronic device, flipping the electronic device such that the light beams are directed towards the second side of the electronic device. In some embodiments, images are pre-processed as described in connection with FIGS. 4A-B. The method includes processing multiple images of multiple sides of the electronic device such that the multiple images have a uniform size and combining the multiple images into a single image to be provided to the neural network.


The angle of the light beams and the arrangement of the light sources can affect the final captured images, as discussed in connection with FIGS. 9A-B. In some embodiments, the method includes adjusting one of the multiple light sources such that an angle between a light beam from the light source and the first side of the electronic device is equal to or smaller than 60 degrees.


In some embodiments, the method includes determining a model of the electronic device in part based on the at least one image and identifying a cosmetic defect on the electronic device that is specific to the model. In some embodiments, the method includes determining, via a second neural network, an price for the electronic device in part based on the initial estimated price. The final offer price can be determined further based on at least (1) a predicted resale value of the electronic device, (2) a predicted incoming volume of a model of the electronic device, or (3) a predicted processing cost of the electronic device. In some embodiments, the method includes receiving an input from a user indicating an acceptance or a rejection of the final price and training the neural network in part based on the at least one image and the input from the user.


Some examples of the disclosed techniques are further described below.


Example 1

A system for evaluating a condition of an electronic device, comprising: a kiosk that includes an inspection plate configured to hold the electronic device, one or more light sources arranged above the inspection plate configured to direct one or more light beams towards the electronic device; and one or more cameras configured to capture at least one image of a first side of the electronic device based on at least one lighting condition generated by the one or more light sources. The system also includes one or more processors in communication with the one or more cameras, the one or more processors configured to extract a set of features of the electronic device based on the at least one image of the electronic device; and determine, via a first neural network, a condition of the electronic device based on the set of features.


Example 2

The system of example 1, wherein the one or more light sources comprises a first subset of light sources and a second subset of light sources, light beams of the first subset of light sources and light beams of the second subset of light sources arranged to be orthogonal to each other.


Example 3

The system of example 1 or 2, wherein the kiosk further includes: an upper chamber positioned above the inspection plate, wherein the one or more light sources are arranged within the upper chamber; a lower chamber positioned below the inspection plate, and a second set of light sources positioned within the lower chamber configured to direct light beams towards the electronic device through the inspection plate.


Example 4

The system of one or more of examples 1 to 3, wherein the kiosk further includes: a flipping mechanism configured to flip the electronic device to allow the one or more cameras to capture at least another image of a second side of the electronic device.


Example 5

The system of one or more of examples 1 to 4, wherein at least one of the one or more light sources is configured to produce a collimated light beam.


Example 6

The system of one or more of examples 1 to 5, wherein an angle between a light beam from one of the one or more light sources and the first side of the electronic device is equal to or smaller than 60 degrees.


Example 7

The system of one or more of examples 1 to 6, wherein the one or more cameras are configured to capture multiple images corresponding to multiple sides of the electronic devices under different lighting conditions, and wherein the one or more processors are configured to process and combine the multiple images into a single input image.


Example 8

The system of one or more of examples 1 to 7, wherein the first neural network is configured to output an indicator indicating the condition of the electronic device.


Example 9

The system of one or more of examples 1 to 8, wherein the one or more processors are further configured to determine an estimated price for the electronic device based on the condition.


In some embodiments, the kiosk is configured to provide information about the electronic device, and wherein the one or more processors are configured to invoke a second neural network to determine a final price for the electronic device based on the estimated price and the information about the electronic device.


Example 10

The system of one or more of examples 1 to 9, wherein the condition comprises a physical condition or a cosmetic condition.


Example 11

The system for evaluating a condition of an electronic device, comprising: a capturing device that comprises at least one light source and at least one camera, wherein the at least one camera is configured to capture multiple images of the electronic devices based on one or more predefined settings, each of the one or more predefined settings specifying at least one of: (1) an angle at which the capturing device is positioned with respect to the electronic device, (2) a light intensity of the at least one light source, (3) an exposure setting of the at least one camera, or (4) a white balance setting of the at least one camera. The system also includes one or more processors in communication with the capturing device, the one or more processors configured to process the multiple images to generate a single input image; extract a set of features of the electronic device based on the at least one image of the electronic device; and determine, via a first neural network, a condition of the electronic device.


In some embodiments, the capturing device is configured to provide information about the electronic device, and wherein the one or more processors are further configured to invoke a second neural network to determine a price for the electronic device based on the condition and the information about the electronic device.


Example 12

The system of example 11, wherein the condition comprises a physical condition or a cosmetic condition.


Example 13

A computer-implemented method for evaluating a condition of an electronic device, comprising: capturing, by at least one camera of a kiosk, at least one image of a first side of the electronic device, wherein the kiosk includes multiple light sources; extracting, by a neural network, a set of features of the electronic device based on the at least one image of the electronic device; and determining a condition of the electronic device based on the set of features.


Example 14

The method of example 13, comprising: capturing, via the at least one camera, at least one image of a second side of the electronic device that is different from the first side based on at least one lighting condition generated by the multiple light sources.


Example 15

The method of example 14, comprising, prior to capturing the at least one image of the second side of the electronic device: flipping the electronic device such that the light beams are directed towards the second side of the electronic device.


Example 16

The method of one or more of examples 13 to 14, comprising: processing multiple images of multiple sides of the electronic device such that the multiple images have a uniform size; and combining the multiple images into a single image to be provided to the neural network.


Example 17

The method of one or more of examples 13 to 16, comprising: adjusting one of the multiple light sources such that an angle between a light beam from the light source and the first side of the electronic device is equal to or smaller than 60 degrees.


Example 18

The method of one or more of examples 13 to 17, comprising: determining a model of the electronic device in part based on the at least one image; and identifying a cosmetic defect on the electronic device that is specific to the model.


In some embodiments, the method comprises determining, via a second neural network, an offer price for the electronic device in part based on the condition, wherein the offer price is determined further based on at least (1) a predicted resale value of the electronic device, (2) a predicted incoming volume of a model of the electronic device, or (3) a predicted processing cost of the electronic device.


Example 19

The method of one or more of examples 13 to 18, comprising: receiving an input from a user indicating an acceptance or a rejection of the offer price; and training the neural network in part based on the at least one image and the input from the user.


Example 20

The method of one or more of examples 13 to 19, wherein the condition comprises a physical condition or a cosmetic condition.


Some embodiments of the disclosure have other aspects, elements, features, and/or steps in addition to or in place of what is described above. These potential additions and replacements are described throughout the rest of the specification. Reference in this specification to “various embodiments,” “certain embodiments,” or “some embodiments” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosure. These embodiments, even alternative embodiments (e.g., referenced as “other embodiments”) are not mutually exclusive of other embodiments. Moreover, various features are described which can be exhibited by some embodiments and not by others. Similarly, various requirements are described which can be requirements for some embodiments but not other embodiments. As used herein, the phrase “and/or” as in “A and/or B” refers to A alone, B alone, and both A and B.


In other instances, well-known structures, materials, operations, and/or systems often associated with smartphones and other handheld devices, consumer electronic devices, computer hardware, software, and network systems, etc., are not shown or described in detail in the following disclosure to avoid unnecessarily obscuring the description of the various embodiments of the technology. Those of ordinary skill in the art will recognize, however, that the present technology can be practiced without one or more of the details set forth herein, or with other structures, methods, components, and so forth. The terminology used below should be interpreted in its broadest reasonable manner, even though it is being used in conjunction with a detailed description of certain examples of embodiments of the technology. Indeed, certain terms can even be emphasized below; however, any terminology intended to be interpreted in any restricted manner will be specifically defined as such in this Detailed Description section.


The accompanying figures depict embodiments of the present technology and are not intended to be limiting of the scope of the present technology. The sizes of various depicted elements are not necessarily drawn to scale, and these various elements can be arbitrarily enlarged to improve legibility. Component details can be abstracted in the figures to exclude details such as the position of components and certain precise connections between such components when such details are unnecessary for a complete understanding of how to make and use the invention.


In the figures, identical reference numbers can identify identical, or at least generally similar, elements. To facilitate the discussion of any particular element, the most significant digit or digits of any reference number can refer to the figure in which that element is first introduced.

Claims
  • 1. A system for evaluating a condition of an electronic device, comprising: a kiosk that includes: an inspection plate configured to hold the electronic device;one or more light sources arranged above the inspection plate and configured to direct one or more light beams towards the electronic device;one or more cameras configured to capture at least one image of a first side of the electronic device based on at least one lighting condition generated by the one or more light sources; andone or more processors in communication with the one or more cameras, the one or more processors configured to: extract a set of features of the electronic device based on the at least one image of the electronic device;apply a trained first neural network on the set of features to compute a score within a range of values and indicative of a severity of damage to the electronic device, wherein the first neural network is trained using a set of images associated with distinct cosmetic evaluation indications; andwherein an output of the first neural network comprises a brand and/or model of the electronic device;determine the condition of the electronic device based on the score and the brand and/or model; anddetermine, via a second neural network, an offer price for the electronic device based on the determined condition, wherein the second neural network is different from the first neural network.
  • 2. The system of claim 1, wherein the one or more light sources comprises a first subset of light sources and a second subset of light sources, light beams of the first subset of light sources and light beams of the second subset of light sources arranged to be orthogonal to each other.
  • 3. The system of claim 1, wherein the kiosk further includes: an upper chamber positioned above the inspection plate, wherein the one or more light sources are arranged within the upper chamber;a lower chamber positioned below the inspection plate; anda second set of light sources positioned within the lower chamber configured to direct light beams towards the electronic device through the inspection plate.
  • 4. The system of claim 1, wherein the kiosk further includes: a flipping mechanism configured to flip the electronic device to allow the one or more cameras to capture at least another image of a second side of the electronic device.
  • 5. The system of claim 1, wherein at least one of the one or more light sources are configured to produce a collimated light beam.
  • 6. The system of claim 1, wherein an angle between a light beam from one of the one or more light sources and the first side of the electronic device is equal to or smaller than 60 degrees.
  • 7. The system of claim 1, wherein the one or more cameras are configured to capture multiple images corresponding to multiple sides of the electronic device under different lighting conditions, and wherein the one or more processors are configured to process and combine the multiple images into a single input image.
  • 8. The system of claim 1, wherein the first neural network is configured to output an indicator indicating the condition of the electronic device.
  • 9. The system of claim 1, wherein the one or more processors are further configured to determine an estimated price for the electronic device based on the condition.
  • 10. The system of claim 1, wherein the condition comprises a physical condition or a cosmetic condition.
  • 11. A system for evaluating a condition of an electronic device, comprising: a capturing device that comprises at least one light source and at least one camera, wherein the at least one camera is configured to capture multiple images of the electronic device based on one or more predefined settings, each of the one or more predefined settings specifying at least one of: (1) an angle at which the capturing device is positioned with respect to the electronic device, (2) a light intensity of the at least one light source, (3) an exposure setting of the at least one camera, or (4) a white balance setting of the at least one camera; andone or more processors in communication with the capturing device, the one or more processors configured to: process the multiple images to generate a single input image;extract a set of features of the electronic device based on the single input image;apply a trained first neural network on the set of features to compute a score within a range of values and indicative of a severity of damage to the electronic device, wherein the first neural network is trained using a set of images associated with distinct cosmetic evaluation indications; andwherein an output of the first neural network comprises a brand and/or model of the electronic device;determine the condition of the electronic device based on the score and the brand and/or model; anddetermine, via a second neural network, an offer price for the electronic device based on the determined condition, wherein the second neural network is different from the first neural network.
  • 12. The system of claim 11, wherein the condition comprises a physical condition or a cosmetic condition.
  • 13. A computer-implemented method for evaluating a condition of an electronic device, comprising: capturing, by at least one camera of a kiosk, at least one image of a first side of the electronic device, wherein the kiosk includes multiple light sources;extracting a set of features of the electronic device based on the at least one image of the electronic device;applying a trained first neural network on the set of features to compute a score within a range of values and indicative of a severity of damage to the electronic device, wherein the first neural network is trained using a set of images associated with distinct cosmetic evaluation indications; andwherein an output of the first neural network comprises a brand and/or model of the electronic device;determining the condition of the electronic device based on the score and the brand and/or model; anddetermine, via a second neural network, an offer price for the electronic device based on the determined condition, wherein the second neural network is different from the first neural network.
  • 14. The method of claim 13, comprising: capturing, via the at least one camera, at least one image of a second side of the electronic device that is different from the first side based on at least one lighting condition generated by the multiple light sources.
  • 15. The method of claim 14, comprising, prior to capturing the at least one image of the second side of the electronic device: flipping the electronic device such that light beams of the multiple light sources are directed towards the second side of the electronic device.
  • 16. The method of claim 13, comprising: processing multiple images of multiple sides of the electronic device such that the multiple images have a uniform size; andcombining the multiple images into a single image to be provided to the first neural network.
  • 17. The method of claim 13, comprising: adjusting one of the multiple light sources such that an angle between a light beam from the one of the multiple light sources and the first side of the electronic device is equal to or smaller than 60 degrees.
  • 18. The method of claim 13, comprising: identifying a cosmetic defect on the electronic device that is specific to the brand and/or model.
  • 19. The method of claim 13, comprising: receiving an input from a user indicating an acceptance or a rejection of the offer price; andtraining the second neural network in part based on the at least one image and the input from the user.
  • 20. The method of claim 13, wherein the condition comprises a physical condition or a cosmetic condition.
CROSS-REFERENCE TO RELATED APPLICATIONS

This patent application claims priority to and the benefit of U.S. Provisional Patent Application No. 62/807,165, entitled “NEURAL NETWORK BASED PHYSICAL CONDITION EVALUATION OF ELECTRONIC DEVICES, AND ASSOCIATED SYSTEMS AND METHODS,” filed Feb. 18, 2019. The entire contents of the above-mentioned patent application are incorporated herein by reference in their entirety.

US Referenced Citations (519)
Number Name Date Kind
3808439 Renius Apr 1974 A
4248334 Hanley et al. Feb 1981 A
4519522 McElwee May 1985 A
4593820 Antonie et al. Jun 1986 A
4715709 Sekine et al. Dec 1987 A
4821118 Lafreniere Apr 1989 A
4845636 Walker Jul 1989 A
4870357 Young et al. Sep 1989 A
4878736 Hekker et al. Nov 1989 A
4893789 Novorsky Jan 1990 A
4927051 Falk et al. May 1990 A
4951308 Bishop et al. Aug 1990 A
5025344 Maly et al. Jun 1991 A
5027074 Haferstat Jun 1991 A
5077462 Newell et al. Dec 1991 A
5091773 Fouche et al. Feb 1992 A
5105149 Tokura Apr 1992 A
5216502 Katz Jun 1993 A
5280170 Baldwin Jan 1994 A
5319459 Mochizuki et al. Jun 1994 A
5339096 Beaufort et al. Aug 1994 A
5419438 Squyres et al. May 1995 A
5436554 Decker Jul 1995 A
5570920 Crisman et al. Nov 1996 A
5572444 Lentz et al. Nov 1996 A
5610710 Canfield et al. Mar 1997 A
5717780 Mitsumune et al. Feb 1998 A
5747784 Walter et al. May 1998 A
5748084 Isikoff May 1998 A
5775806 Allred Jul 1998 A
5839058 Phillips et al. Nov 1998 A
5871371 Rothenberger et al. Feb 1999 A
5920338 Katz Jul 1999 A
5949901 Nichani et al. Sep 1999 A
5965858 Suzuki et al. Oct 1999 A
5966654 Croughwell et al. Oct 1999 A
5987159 Nichani Nov 1999 A
5988431 Roe Nov 1999 A
6029851 Jenkins et al. Feb 2000 A
6041229 Turner Mar 2000 A
6055512 Dean et al. Apr 2000 A
6100986 Rydningen Aug 2000 A
6170702 Zettler et al. Jan 2001 B1
6181805 Koike et al. Jan 2001 B1
6228008 Pollington et al. May 2001 B1
6234812 Ivers et al. May 2001 B1
6259827 Nichani Jul 2001 B1
6264104 Jenkins et al. Jul 2001 B1
6323782 Stephens et al. Nov 2001 B1
6330354 Companion et al. Dec 2001 B1
6330958 Ruskin et al. Dec 2001 B1
6393095 Robinson May 2002 B1
6462644 Howell et al. Oct 2002 B1
6529837 Kang Mar 2003 B1
6535637 Wootton et al. Mar 2003 B1
6573886 Lehtinen et al. Jun 2003 B1
6587581 Matsuyama et al. Jul 2003 B1
6595684 Casagrande et al. Jul 2003 B1
6597552 Griepentrog et al. Jul 2003 B1
6633377 Weiss et al. Oct 2003 B1
6667800 Larsson et al. Dec 2003 B1
6748296 Banerjee et al. Jun 2004 B2
6754637 Stenz Jun 2004 B1
6758370 Cooke et al. Jul 2004 B2
6798528 Hartman Sep 2004 B1
6822422 Sagawa Nov 2004 B2
6842596 Morii et al. Jan 2005 B2
6854656 Matsumori Feb 2005 B2
6947941 Koon Sep 2005 B1
D512964 Kissinger et al. Dec 2005 S
7069236 Tsunenari Jun 2006 B1
7076449 Tsunenari et al. Jul 2006 B2
7086592 Wagner et al. Aug 2006 B2
7178720 Strubbe et al. Feb 2007 B1
7234609 DeLazzer et al. Jun 2007 B2
7251458 O'Connell Jul 2007 B2
7268345 Schultz Sep 2007 B2
7334729 Brewington Feb 2008 B2
7407392 Cooke et al. Aug 2008 B1
7408674 Moro et al. Aug 2008 B2
7455226 Hammond et al. Nov 2008 B1
7520666 Pevzner et al. Apr 2009 B2
7529687 Phan May 2009 B1
7567344 LeBlanc et al. Jul 2009 B2
7642687 Kageyama et al. Jan 2010 B2
7646193 Suzuki et al. Jan 2010 B2
7649450 Campion et al. Jan 2010 B2
7702108 Amon et al. Apr 2010 B2
7735125 Alvarez et al. Jun 2010 B1
7761331 Low et al. Jul 2010 B2
7783379 Beane et al. Aug 2010 B2
7848833 Li Dec 2010 B2
7881965 Bowles et al. Feb 2011 B2
7890373 Junger Feb 2011 B2
D640199 Wilson Jun 2011 S
8010402 Sharma et al. Aug 2011 B1
8019588 Wohlberg et al. Sep 2011 B1
8025229 Hammond et al. Sep 2011 B2
8031930 Wang et al. Oct 2011 B2
8107243 Guccione et al. Jan 2012 B2
8112325 Foy et al. Feb 2012 B2
8142199 Almouli Mar 2012 B1
8195511 Bowles et al. Jun 2012 B2
8200533 Librizzi et al. Jun 2012 B2
8200736 Shi Jun 2012 B2
8215546 Lin et al. Jul 2012 B2
8239262 Bowles et al. Aug 2012 B2
8254883 Uchida Aug 2012 B2
8266008 Siegel et al. Sep 2012 B1
8340815 Peters et al. Dec 2012 B2
8369987 Claessen Feb 2013 B2
8401914 Kim Mar 2013 B1
8417234 Sanding et al. Apr 2013 B2
8423404 Bowles et al. Apr 2013 B2
8429021 Kraft et al. Apr 2013 B2
8463646 Bowles Jun 2013 B2
8536472 Wu et al. Sep 2013 B2
8543358 Trabona Sep 2013 B2
8566183 Bonar et al. Oct 2013 B1
8606633 Tarbert et al. Dec 2013 B2
8718717 Vaknin et al. May 2014 B2
8755783 Brahami et al. Jun 2014 B2
8806280 Stephenson Aug 2014 B2
8823794 Suzuki et al. Sep 2014 B2
8824136 Interian et al. Sep 2014 B1
8743215 Lee Nov 2014 B1
8922643 Ji et al. Dec 2014 B2
9010627 Prasad et al. Apr 2015 B1
9043026 Lien et al. May 2015 B2
9124056 Lewis, Jr. Sep 2015 B1
9189911 Kavli et al. Nov 2015 B2
9195979 Geller Nov 2015 B2
9256863 Chayon et al. Feb 2016 B2
9317989 Grow et al. Apr 2016 B2
9355515 Brahami et al. May 2016 B2
9367982 Chayun et al. Jun 2016 B2
9378606 Chayun et al. Jun 2016 B2
9390442 Lyle Jul 2016 B2
9497563 Hornung et al. Nov 2016 B2
9582101 Chang et al. Feb 2017 B2
9595238 Won Mar 2017 B2
9621947 Oztaskent Apr 2017 B1
9641997 Vratskides May 2017 B2
9697548 Jaff et al. Jul 2017 B1
9704142 Ahn Jul 2017 B2
9818160 Bowles et al. Nov 2017 B2
9866664 Sinha et al. Jan 2018 B2
9881284 Bowles et al. Jan 2018 B2
9885672 Forutanpour et al. Feb 2018 B2
9904911 Bowles et al. Feb 2018 B2
9911102 Bowles Mar 2018 B2
9934644 Chayun et al. Apr 2018 B2
10032140 Bowles et al. Jul 2018 B2
10043339 Walker et al. Aug 2018 B2
10044843 Sinha et al. Aug 2018 B2
10055798 Bowles et al. Aug 2018 B2
10127647 Forutanpour et al. Nov 2018 B2
10157379 Singh Dec 2018 B2
10157427 Bowles et al. Dec 2018 B2
10269110 Forutanpour et al. Apr 2019 B2
10275813 Fu Apr 2019 B2
10325440 Abdelmalak et al. Jun 2019 B2
10339509 Bordeleau et al. Jul 2019 B2
10401411 Snook et al. Sep 2019 B2
10417615 Bowles et al. Sep 2019 B2
10438174 Bowles et al. Oct 2019 B2
10445708 Hunt et al. Oct 2019 B2
10475002 Silva et al. Nov 2019 B2
10496963 Silva et al. Dec 2019 B2
10572946 Bowles et al. Feb 2020 B2
10679279 Ward Jun 2020 B2
10740891 Chen Aug 2020 B1
10824942 Bhotika Nov 2020 B1
10825082 Librizzi et al. Nov 2020 B2
10846672 Dion et al. Nov 2020 B2
10853873 Bowles et al. Dec 2020 B2
10860990 Bowles et al. Dec 2020 B2
10909673 Forutanpour et al. Feb 2021 B2
10970786 Matheson Apr 2021 B1
11010841 Bowles et al. May 2021 B2
11024111 Abdelmalak et al. Jun 2021 B2
11080662 Bowles et al. Aug 2021 B2
11080672 Bowles Aug 2021 B2
11107046 Bowles Aug 2021 B2
11126973 Silva et al. Sep 2021 B2
11232412 Hunt et al. Jan 2022 B2
11288789 Chen Mar 2022 B1
11315093 Bowles Apr 2022 B2
11379886 Fields Jul 2022 B1
20010025883 Ichihara et al. Oct 2001 A1
20010035425 Rocco et al. Nov 2001 A1
20010039531 Aoki Nov 2001 A1
20020014577 Ulrich et al. Feb 2002 A1
20020035515 Moreno Mar 2002 A1
20020067184 Smith et al. Jun 2002 A1
20020087413 Mahaffy et al. Jul 2002 A1
20020112177 Voltmer Aug 2002 A1
20020129170 Moore et al. Sep 2002 A1
20020157033 Cox Oct 2002 A1
20020162966 Yoder Nov 2002 A1
20020186878 Hoon et al. Dec 2002 A1
20030006277 Maskatiya et al. Jan 2003 A1
20030018897 Bellis, Jr. et al. Jan 2003 A1
20030025476 Treia Feb 2003 A1
20030036866 Nair et al. Feb 2003 A1
20030061150 Kocher et al. Mar 2003 A1
20030100707 Hwang et al. May 2003 A1
20030146898 Kawasaki et al. Aug 2003 A1
20030170529 Sagawa Sep 2003 A1
20030179371 Rangarajan Sep 2003 A1
20030191675 Murashita Oct 2003 A1
20030197782 Ashe Oct 2003 A1
20030204289 Banerjee et al. Oct 2003 A1
20040012825 Tesavis Jan 2004 A1
20040088231 Davis May 2004 A1
20040114153 Andersen et al. Jun 2004 A1
20040141320 Bock et al. Jul 2004 A1
20040150815 Sones et al. Aug 2004 A1
20040156557 Van Der Weij Aug 2004 A1
20040156667 Van Der Weij et al. Aug 2004 A1
20040186744 Lux Sep 2004 A1
20040189812 Gustavsson Sep 2004 A1
20040200902 Ishioroshi Oct 2004 A1
20040205015 DeLaCruz Oct 2004 A1
20040235513 O'Connell Nov 2004 A1
20040242216 Boutsikakis Dec 2004 A1
20040262521 Devitt et al. Dec 2004 A1
20050027622 Walker et al. Feb 2005 A1
20050043897 Meyer Feb 2005 A1
20050109841 Ryan et al. May 2005 A1
20050128551 Yang Jun 2005 A1
20050135917 Kauppila et al. Jun 2005 A1
20050137942 LaFluer Jun 2005 A1
20050139661 Eglen et al. Jun 2005 A1
20050143149 Becker et al. Jun 2005 A1
20050167620 Cho et al. Aug 2005 A1
20050187657 Hashimoto et al. Aug 2005 A1
20050216120 Rosenberg et al. Sep 2005 A1
20050222690 Wang et al. Oct 2005 A1
20050231595 Wang et al. Oct 2005 A1
20050240958 Nguyen et al. Oct 2005 A1
20060167580 Whittier Jan 2006 A1
20060022827 Highham Feb 2006 A1
20060038114 Cofer et al. Feb 2006 A9
20060047573 Mitchell et al. Mar 2006 A1
20060074756 Boykin Apr 2006 A1
20060085158 Cakiner Apr 2006 A1
20060184379 Tan et al. Aug 2006 A1
20060195384 Bauer et al. Aug 2006 A1
20060219776 Finn Oct 2006 A1
20060229108 Cehelnik Oct 2006 A1
20060235747 Hammond et al. Oct 2006 A1
20060217152 Fok et al. Nov 2006 A1
20060258008 Holler et al. Nov 2006 A1
20060261931 Cheng et al. Nov 2006 A1
20060271431 Wehr et al. Nov 2006 A1
20060279307 Wang et al. Dec 2006 A1
20060280356 Yamagashi Dec 2006 A1
20070012665 Nelson Jan 2007 A1
20070013124 Graef et al. Jan 2007 A1
20070032098 Bowles et al. Feb 2007 A1
20070050083 Signorelli Mar 2007 A1
20070057815 Foy et al. Mar 2007 A1
20070129906 Stoecker et al. Jun 2007 A1
20070133844 Waehner et al. Jun 2007 A1
20070150403 Mock et al. Jun 2007 A1
20070140310 Rolton et al. Jul 2007 A1
20070205751 Suzuki et al. Sep 2007 A1
20070258085 Robbins Nov 2007 A1
20070263099 Motta et al. Nov 2007 A1
20070269099 Nishino et al. Nov 2007 A1
20070276911 Bhumkar Nov 2007 A1
20070281734 Mizrachi Dec 2007 A1
20070282999 Tu et al. Dec 2007 A1
20080004828 Mizrachi Jan 2008 A1
20080027581 Saether et al. Jan 2008 A1
20080033596 Fausak et al. Feb 2008 A1
20080109746 Mayer May 2008 A1
20080111989 Dufour et al. May 2008 A1
20080133432 Ramseyer Jun 2008 A1
20080149720 Colville Jun 2008 A1
20080167578 Bryer et al. Jul 2008 A1
20080177598 Davie Jul 2008 A1
20080207198 Juric Aug 2008 A1
20080231113 Guccione et al. Sep 2008 A1
20080255901 Carroll et al. Oct 2008 A1
20080256008 Kwok Oct 2008 A1
20080260235 Cai Oct 2008 A1
20080277467 Carlson Nov 2008 A1
20080281691 Pearson et al. Nov 2008 A1
20080296374 Gonen et al. Dec 2008 A1
20080303915 Omi Dec 2008 A1
20080306701 Zhong et al. Dec 2008 A1
20090051907 Li et al. Feb 2009 A1
20090079388 Reddy Feb 2009 A1
20090078775 Giebel et al. Mar 2009 A1
20090095047 Patel et al. Apr 2009 A1
20090114716 Ramachandran May 2009 A1
20090132813 Schibuk May 2009 A1
20090156199 Steenstra et al. Jun 2009 A1
20090160668 Crowley et al. Jun 2009 A1
20090177319 Garibaldi et al. Jul 2009 A1
20090184865 Valo et al. Jul 2009 A1
20090187491 Bull et al. Jul 2009 A1
20090190142 Taylor et al. Jul 2009 A1
20090207743 Huq et al. Aug 2009 A1
20090244285 Chathukutty Oct 2009 A1
20090247133 Holmen et al. Oct 2009 A1
20090248883 Suryanarayana et al. Oct 2009 A1
20090251815 Wang et al. Oct 2009 A1
20090262341 Konopa et al. Oct 2009 A1
20090265035 Jenkinson et al. Oct 2009 A1
20090299543 Cox et al. Dec 2009 A1
20090312009 Fishel Dec 2009 A1
20090321511 Browne Dec 2009 A1
20090322706 Austin Dec 2009 A1
20100005004 Hudak et al. Jan 2010 A1
20100051695 Yepez et al. Mar 2010 A1
20100063894 Lundy Mar 2010 A1
20100110174 Leconte May 2010 A1
20100115887 Schroeder et al. May 2010 A1
20100147953 Barkan Jun 2010 A1
20100157280 Kusevic et al. Jun 2010 A1
20100161397 Gauthier et al. Jun 2010 A1
20100162359 Casey et al. Jun 2010 A1
20100185506 Wolff Jul 2010 A1
20100219234 Forbes Sep 2010 A1
20100235198 Fini et al. Sep 2010 A1
20100237854 Kumhyr et al. Sep 2010 A1
20100260271 Kapoor Oct 2010 A1
20100262481 Baker et al. Oct 2010 A1
20100312639 Mastronardi Dec 2010 A1
20110035322 Lively Feb 2011 A1
20110043628 Yun Feb 2011 A1
20110055322 Gregersen Mar 2011 A1
20110060641 Grossman et al. Mar 2011 A1
20110066514 Maraz Mar 2011 A1
20110067520 Ihrke et al. Mar 2011 A1
20110113479 Ganem May 2011 A1
20110296508 Os et al. Dec 2011 A1
20110313840 Mason et al. Dec 2011 A1
20120004761 Madruga Jan 2012 A1
20120016518 Saario et al. Jan 2012 A1
20120022965 Seergy Jan 2012 A1
20120026582 Okabe et al. Feb 2012 A1
20120029985 Wilson et al. Feb 2012 A1
20120030097 Hagan et al. Feb 2012 A1
20120030399 Ben-Harosh Feb 2012 A1
20120054113 Jayaraman et al. Mar 2012 A1
20120063501 Aguren Mar 2012 A1
20120078413 Baker Mar 2012 A1
20120116928 Gventer May 2012 A1
20120116929 Gventer May 2012 A1
20120117001 Gventer et al. May 2012 A1
20120127307 Hassenzahl May 2012 A1
20120146956 Jenkinson Jun 2012 A1
20120235812 De Mello et al. Sep 2012 A1
20120254046 Librizzi et al. Oct 2012 A1
20120280934 Ha Nov 2012 A1
20120301009 Dabic Nov 2012 A1
20120303431 Phillips et al. Nov 2012 A1
20130006713 Haake Jan 2013 A1
20130034305 Jahanshahi et al. Feb 2013 A1
20130041508 Hu et al. Feb 2013 A1
20130046611 Bowles et al. Feb 2013 A1
20130046699 Bowles et al. Feb 2013 A1
20130112440 Alsaif et al. May 2013 A1
20130124426 Bowles et al. May 2013 A1
20130126741 Srivastava et al. May 2013 A1
20130137376 Fitzgerald et al. May 2013 A1
20130144797 Bowles et al. Jun 2013 A1
20130155061 Jahanshahi et al. Jun 2013 A1
20130159119 Henderson et al. Jun 2013 A1
20130173430 Benjamin Jul 2013 A1
20130173434 Hartman Jul 2013 A1
20130181935 McKenzie et al. Jul 2013 A1
20130198089 Bowles Aug 2013 A1
20130198144 Bowles Aug 2013 A1
20130246211 Sullivan Sep 2013 A1
20130246212 Sullivan Sep 2013 A1
20130253700 Carson et al. Sep 2013 A1
20130284805 Kraft et al. Oct 2013 A1
20130290146 West et al. Oct 2013 A1
20130297388 Kyle, Jr. et al. Nov 2013 A1
20140006451 Mullis et al. Jan 2014 A1
20140012643 Behrisch Jan 2014 A1
20140038556 DeSousa Feb 2014 A1
20140046845 Dogin et al. Feb 2014 A1
20140052329 Amirpour Feb 2014 A1
20140067710 Gventer et al. Mar 2014 A1
20140080550 Ino et al. Mar 2014 A1
20140143161 Ahn May 2014 A1
20140150100 Gupta et al. May 2014 A1
20140156883 Bowles Jun 2014 A1
20140214505 Shuster-Arechiga et al. Jul 2014 A1
20140235258 Wang et al. Aug 2014 A1
20140244315 Cahill et al. Aug 2014 A1
20140278244 Humphrey et al. Sep 2014 A1
20140297368 Ferder Oct 2014 A1
20140347473 Wolff et al. Nov 2014 A1
20150006281 Takahashi Jan 2015 A1
20150066677 Bowles et al. Mar 2015 A1
20150088731 Ackerman Mar 2015 A1
20150120485 Nash Apr 2015 A1
20150170237 Powell Jun 2015 A1
20150193797 Gerrity Jul 2015 A1
20150206200 Edmondson et al. Jul 2015 A1
20150278529 Cho et al. Oct 2015 A1
20150293860 Bowles Oct 2015 A9
20150309912 Nguyen et al. Oct 2015 A1
20150324761 Nguyen et al. Nov 2015 A1
20150324870 Nguyen et al. Nov 2015 A1
20150332206 Trew et al. Nov 2015 A1
20150356637 Graffia et al. Dec 2015 A1
20160019607 Burmester et al. Jan 2016 A1
20160019685 Nguyen et al. Jan 2016 A1
20160055392 Nakano Feb 2016 A1
20160078434 Huxham et al. Mar 2016 A1
20160087381 Wong et al. Mar 2016 A1
20160092849 Cirannek et al. Mar 2016 A1
20160125548 Bowles May 2016 A1
20160125612 Seki et al. May 2016 A1
20160171544 Heminger et al. Jun 2016 A1
20160171575 Bowles et al. Jun 2016 A1
20160184990 Song et al. Jun 2016 A1
20160210648 Cirannek et al. Jul 2016 A1
20160269401 Saito et al. Sep 2016 A1
20160269895 Soini et al. Sep 2016 A1
20160275460 Ploetner et al. Sep 2016 A1
20160275518 Bowles et al. Sep 2016 A1
20160301786 Koltsov et al. Oct 2016 A1
20160328684 Bowles et al. Nov 2016 A1
20160379287 Dabiri Dec 2016 A1
20170083886 Silva et al. Mar 2017 A1
20170091823 Adinarayan et al. Mar 2017 A1
20170115235 Ohlsson Apr 2017 A1
20170169401 Beane et al. Jun 2017 A1
20170221110 Sullivan Aug 2017 A1
20170256051 Dwivedi Sep 2017 A1
20170278191 Tassone et al. Sep 2017 A1
20170301010 Bowles et al. Oct 2017 A1
20170301078 Forutanpour Oct 2017 A1
20170343481 Jahanshahi Nov 2017 A1
20170372465 Forutanpour Dec 2017 A1
20180084094 Sinha et al. Mar 2018 A1
20180130136 Bowles May 2018 A1
20180157246 Huang et al. Jun 2018 A1
20180157820 Adams et al. Jun 2018 A1
20180240144 Curtis Aug 2018 A1
20180247280 Bowles Aug 2018 A1
20180260794 Bowles et al. Sep 2018 A1
20180293566 Engles et al. Oct 2018 A1
20180293664 Zhang Oct 2018 A1
20180321163 Casadio Nov 2018 A1
20180322623 Memo Nov 2018 A1
20180342050 Fitzgerald Nov 2018 A1
20180350163 Pofale Dec 2018 A1
20190019147 McCarty Jan 2019 A1
20190066075 Lobo et al. Feb 2019 A1
20190066439 Pinkus Feb 2019 A1
20190073566 Brauer Mar 2019 A1
20190073568 He Mar 2019 A1
20190102874 Goja Apr 2019 A1
20190166278 Hiyama May 2019 A1
20190222748 Weir et al. Jul 2019 A1
20190251777 Abdelmalak et al. Aug 2019 A1
20190272628 Tsou Sep 2019 A1
20190279431 Wurmfeld et al. Sep 2019 A1
20190287141 Bordeleau et al. Sep 2019 A1
20190318465 Nguyen Oct 2019 A1
20190372827 Vasseur Dec 2019 A1
20200020097 Do Jan 2020 A1
20200042969 Ray Feb 2020 A1
20200066067 Herman Feb 2020 A1
20200090137 Bowles et al. Mar 2020 A1
20200104720 Bao Apr 2020 A1
20200151677 Bowles et al. May 2020 A1
20200167748 Dion et al. May 2020 A1
20200175669 Bian Jun 2020 A1
20200202319 Forutanpour et al. Jun 2020 A1
20200202405 Glickman et al. Jun 2020 A1
20200241891 Li et al. Jul 2020 A1
20200258343 Forutanpour et al. Aug 2020 A1
20200259300 Forutanpour et al. Aug 2020 A1
20200393742 Dion et al. Dec 2020 A1
20210012315 Priebatsch Jan 2021 A1
20210035206 Bowles et al. Feb 2021 A1
20210081698 Lindeman Mar 2021 A1
20210081914 Nelms et al. Mar 2021 A1
20210110366 Dion et al. Apr 2021 A1
20210110440 Dion et al. Apr 2021 A1
20210174312 Bowles et al. Jun 2021 A1
20210192484 Forutanpour et al. Jun 2021 A1
20210209512 Gaddam Jul 2021 A1
20210209746 Johnson et al. Jul 2021 A1
20210224867 Bordeleau et al. Jul 2021 A1
20210254966 Hur Aug 2021 A1
20210272208 Leise Sep 2021 A1
20210295494 Forutanpour et al. Sep 2021 A1
20220027879 Bowles et al. Jan 2022 A1
20220050897 Gaddam Feb 2022 A1
20220051212 Forutanpour et al. Feb 2022 A1
20220051300 Forutanpour et al. Feb 2022 A1
20220051301 Forutanpour et al. Feb 2022 A1
20220051507 Forutanpour et al. Feb 2022 A1
20220067798 Forutanpour et al. Mar 2022 A1
20220068076 Forutanpour et al. Mar 2022 A1
20220114854 Forutanpour et al. Apr 2022 A1
20220172178 Forutanpour et al. Jun 2022 A1
20220187802 Wittenberg Jun 2022 A1
20220198407 Beane et al. Jun 2022 A1
20220262189 Dion et al. Aug 2022 A1
20220277281 Dion et al. Sep 2022 A1
20220284406 Hunt et al. Sep 2022 A1
20220292464 Silva et al. Sep 2022 A1
20220318774 Bowles Oct 2022 A1
20230007937 Forutanpour et al. Jan 2023 A1
20230077844 Bowles et al. Mar 2023 A1
20230100849 Bowles et al. Mar 2023 A1
Foreign Referenced Citations (306)
Number Date Country
2760863 Nov 2010 CA
2818533 May 2012 CA
2866147 Sep 2013 CA
3069888 Jan 2019 CA
3069890 Jan 2019 CA
1365479 Aug 2002 CN
1574437 Feb 2005 CN
2708415 Jul 2005 CN
1864088 Nov 2006 CN
1957320 May 2007 CN
2912132 Jun 2007 CN
200965706 Oct 2007 CN
101176124 May 2008 CN
101379488 Mar 2009 CN
102467728 May 2012 CN
202351953 Jul 2012 CN
202394296 Aug 2012 CN
102654927 Sep 2012 CN
102682597 Sep 2012 CN
102930642 Feb 2013 CN
102976004 Mar 2013 CN
103198562 Jul 2013 CN
103226870 Jul 2013 CN
203242065 Oct 2013 CN
103440607 Dec 2013 CN
103514641 Jan 2014 CN
103544772 Jan 2014 CN
203408902 Jan 2014 CN
103662541 Mar 2014 CN
103679147 Mar 2014 CN
203520502 Apr 2014 CN
103824387 May 2014 CN
203588366 May 2014 CN
103843040 Jun 2014 CN
103954626 Jul 2014 CN
302944037 Sep 2014 CN
302944252 Sep 2014 CN
302944253 Sep 2014 CN
303042750 Dec 2014 CN
205129815 Apr 2016 CN
205132514 Apr 2016 CN
205140067 Apr 2016 CN
106022379 Oct 2016 CN
303896361 Oct 2016 CN
106203643 Dec 2016 CN
106293734 Jan 2017 CN
106372638 Feb 2017 CN
304051346 Feb 2017 CN
304139831 May 2017 CN
304169301 Jun 2017 CN
206440635 Aug 2017 CN
107220640 Sep 2017 CN
206466691 Sep 2017 CN
107514978 Dec 2017 CN
206861374 Jan 2018 CN
207037788 Feb 2018 CN
105444678 Mar 2018 CN
304702339 Jun 2018 CN
304702340 Jun 2018 CN
304747709 Jul 2018 CN
304795309 Aug 2018 CN
108596658 Sep 2018 CN
108647588 Oct 2018 CN
207993120 Oct 2018 CN
207993121 Oct 2018 CN
207995226 Oct 2018 CN
304842785 Oct 2018 CN
108764236 Nov 2018 CN
208086545 Nov 2018 CN
208172834 Nov 2018 CN
304958348 Dec 2018 CN
305014434 Jan 2019 CN
305014435 Jan 2019 CN
109831575 May 2019 CN
208819255 May 2019 CN
208819289 May 2019 CN
208819290 May 2019 CN
208969761 Jun 2019 CN
305275610 Jul 2019 CN
110321082 Oct 2019 CN
110333876 Oct 2019 CN
110347341 Oct 2019 CN
110595361 Dec 2019 CN
110653162 Jan 2020 CN
110675399 Jan 2020 CN
110751002 Feb 2020 CN
110788015 Feb 2020 CN
110796646 Feb 2020 CN
110796647 Feb 2020 CN
110796669 Feb 2020 CN
110827244 Feb 2020 CN
110827245 Feb 2020 CN
110827246 Feb 2020 CN
110827247 Feb 2020 CN
110827248 Feb 2020 CN
110827249 Feb 2020 CN
110880028 Mar 2020 CN
110928730 Mar 2020 CN
305638504 Mar 2020 CN
110976302 Apr 2020 CN
111080184 Apr 2020 CN
210348162 Apr 2020 CN
111175318 May 2020 CN
111210473 May 2020 CN
305767220 May 2020 CN
111238430 Jun 2020 CN
111242116 Jun 2020 CN
111262987 Jun 2020 CN
111272067 Jun 2020 CN
111272388 Jun 2020 CN
111272393 Jun 2020 CN
111273704 Jun 2020 CN
111277466 Jun 2020 CN
111277659 Jun 2020 CN
111277695 Jun 2020 CN
111277696 Jun 2020 CN
111290660 Jun 2020 CN
111290949 Jun 2020 CN
111291661 Jun 2020 CN
111292302 Jun 2020 CN
111292374 Jun 2020 CN
111294453 Jun 2020 CN
111294459 Jun 2020 CN
111307429 Jun 2020 CN
111311556 Jun 2020 CN
111311678 Jun 2020 CN
111311687 Jun 2020 CN
111311749 Jun 2020 CN
111314445 Jun 2020 CN
111314535 Jun 2020 CN
111325715 Jun 2020 CN
111325716 Jun 2020 CN
111325717 Jun 2020 CN
111325901 Jun 2020 CN
305818424 Jun 2020 CN
111530774 Aug 2020 CN
211291337 Aug 2020 CN
211402187 Sep 2020 CN
211515235 Sep 2020 CN
211538600 Sep 2020 CN
306113050 Oct 2020 CN
306113051 Oct 2020 CN
306113052 Oct 2020 CN
212023984 Nov 2020 CN
212031269 Nov 2020 CN
306164092 Nov 2020 CN
306164093 Nov 2020 CN
306164094 Nov 2020 CN
306164095 Nov 2020 CN
212314534 Jan 2021 CN
212597202 Feb 2021 CN
306323627 Feb 2021 CN
212681731 Mar 2021 CN
111314537 Apr 2021 CN
213001252 Apr 2021 CN
111327745 Aug 2021 CN
111327746 Aug 2021 CN
111256609 Sep 2021 CN
202702438 Jan 2023 CN
202711369 Jan 2023 CN
10031532 Oct 2001 DE
0116970 Dec 1991 EP
0654003 May 1995 EP
1168253 Jan 2002 EP
1270905 Jan 2003 EP
1703436 Sep 2006 EP
3206194 Aug 2017 EP
2428072 Jan 2018 EP
3047833 Mar 2018 FR
2167553 May 1986 GB
30014296 Aug 2020 HK
7112801 May 1995 JP
H7334583 Dec 1995 JP
H11242005 Sep 1999 JP
2000121564 Apr 2000 JP
2000171409 Jun 2000 JP
2000180371 Jun 2000 JP
3123095 Jan 2001 JP
2002019147 Jan 2002 JP
2002183286 Jun 2002 JP
2002259528 Sep 2002 JP
2002302252 Oct 2002 JP
2002324264 Nov 2002 JP
2002358354 Dec 2002 JP
2003139516 May 2003 JP
2003230229 Aug 2003 JP
2003242243 Aug 2003 JP
2003264007 Sep 2003 JP
2003267509 Sep 2003 JP
2004021569 Jan 2004 JP
2004191496 Jul 2004 JP
2004239850 Aug 2004 JP
2004288143 Oct 2004 JP
2004303102 Oct 2004 JP
2004341681 Dec 2004 JP
2005063203 Mar 2005 JP
2005122059 May 2005 JP
2005308476 Nov 2005 JP
2006127308 May 2006 JP
2006195814 Jul 2006 JP
2006203451 Aug 2006 JP
2006227764 Aug 2006 JP
2006260246 Sep 2006 JP
2007141266 Jun 2007 JP
2007155455 Jun 2007 JP
2007179516 Jul 2007 JP
2007265340 Oct 2007 JP
2008045959 Feb 2008 JP
2008522299 Jun 2008 JP
2008293391 Dec 2008 JP
2007086725 Apr 2009 JP
2009175035 Aug 2009 JP
2009245058 Oct 2009 JP
2009250971 Oct 2009 JP
2009290852 Dec 2009 JP
2010177720 Aug 2010 JP
2010276896 Dec 2010 JP
2011518387 Jun 2011 JP
2012504832 Feb 2012 JP
2012058932 Mar 2012 JP
2013033361 Feb 2013 JP
2013037441 Feb 2013 JP
2017093938 Jun 2017 JP
2019012474 Jan 2019 JP
3223233 Sep 2019 JP
20000064168 Nov 2000 KR
20010074614 Aug 2001 KR
20010097567 Nov 2001 KR
100766860 Oct 2007 KR
20130085255 Jul 2013 KR
101326680 Nov 2013 KR
101329949 Nov 2013 KR
20140037543 Mar 2014 KR
101599251 Mar 2016 KR
20180088062 Aug 2018 KR
20180088063 Aug 2018 KR
1020180086617 Aug 2018 KR
20180117278 Oct 2018 KR
20190026131 Mar 2019 KR
20190107593 Sep 2019 KR
20190107595 Sep 2019 KR
20190107596 Sep 2019 KR
1020190107594 Sep 2019 KR
1020200115308 Oct 2020 KR
20210020717 Feb 2021 KR
1020210059148 May 2021 KR
WO8503790 Aug 1985 WO
WO2001015096 Mar 2001 WO
WO2002005176 Jan 2002 WO
WO0221090 Mar 2002 WO
WO2002025613 Mar 2002 WO
WO2002039357 May 2002 WO
WO2003012717 Feb 2003 WO
WO2003014994 Feb 2003 WO
WO2004021114 Mar 2004 WO
WO2004114490 Dec 2004 WO
WO2005008566 Jan 2005 WO
WO2005101346 Oct 2005 WO
WO2006021825 Mar 2006 WO
WO2006058601 Jun 2006 WO
WO2006080851 Aug 2006 WO
WO2007066166 Jun 2007 WO
WO2009089607 Jul 2009 WO
WO2009128173 Oct 2009 WO
WO2009128176 Oct 2009 WO
WO2009129526 Oct 2009 WO
WO2010040116 Apr 2010 WO
WO2010128267 Nov 2010 WO
WO2010128315 Nov 2010 WO
WO2011131016 Oct 2011 WO
WO2012073126 Jun 2012 WO
WO2013002748 Jan 2013 WO
WO2013074819 May 2013 WO
WO2014075055 May 2014 WO
WO2014141180 Sep 2014 WO
WO2015022409 Feb 2015 WO
WO2015093676 Jun 2015 WO
WO2015108864 Jul 2015 WO
WO2016181224 Nov 2016 WO
WO2015196175 Dec 2016 WO
WO2017034441 Mar 2017 WO
WO2017081527 May 2017 WO
WO2017156046 Sep 2017 WO
WO2018124669 Jul 2018 WO
WO2018133068 Jul 2018 WO
WO2018146374 Aug 2018 WO
WO2019012305 Jan 2019 WO
WO2019012505 Jan 2019 WO
WO2019012506 Jan 2019 WO
WO2019212513 Nov 2019 WO
WO2019212515 Nov 2019 WO
WO2020082991 Apr 2020 WO
WO2020204503 Oct 2020 WO
WO2021019286 Feb 2021 WO
WO2021082918 May 2021 WO
WO2021082919 May 2021 WO
WO2021082920 May 2021 WO
WO2021082921 May 2021 WO
WO2021082922 May 2021 WO
WO2021082923 May 2021 WO
WO2021142009 Jul 2021 WO
WO2021147385 Jul 2021 WO
WO2021147386 Jul 2021 WO
WO2021147387 Jul 2021 WO
WO2021147388 Jul 2021 WO
WO2021172803 Sep 2021 WO
Non-Patent Literature Citations (52)
Entry
U.S. Appl. No. 16/952,340, filed Nov. 20, 2020.
2006 Florida Statutes Title XXXIII, Chapter 538, Sections 538.03 and 538.04, 7 pages.
3GPP Organizational Partners, “3rd Generation Partnership Project; Technical Specification Group Terminals; AT command set for GSM Mobile Equipment (ME),” Global System for Mobile Communications, 1998, 124 pages.
Aftermarket Cellular Accessories, “Cellular Phone Model Identification,” retrieved from http://web/archive.org/web/20060328064957/http://aftermarketcellular.com/ic/identification.html on Mar. 16, 2014, published Mar. 28, 2006, 3 pages.
Altec Lansing User's Guide 2007, 8 pages.
Bhule et al., “Environmental and economic trade-offs in consumer electronic products recycling: a case study of cell phones and computers,” IEEE International Symposium on Electronics and the Environment, Conference Record, 2004.
Bournique, D.: “Mobile Karma Shuts Down as iCloud and Blacklists Challenge Used Phone Buyers”, Prepaid Phone News, Jul. 23, 2014 (Jul. 23, 2014), XP055229747, Retrieved from the Internet <URL:http://www.prepaidphonenews.com/2014/07/mobile-karma-shuts-down-as-icloud-and.html>; accessed Nov. 27, 2017; 2 pages.
Business Wire, “The World's First Office Photography Machine” at CES 2008 Launched by Ortery Technologies, Jan. 7, 2008, 3 pages.
CNET, “Tackling LCD “burn ins”, and dead/stick Pixels”, published Sep. 2, 2009, retrieved from http://www.cnet.com/news/tackling-LCD-burn-ins-and-deadstuck-pixels/.
Cybercom Group Europe AB, “OMSI Forum,” Downloads, 2005, 2 pages.
Cybercom Group Europe AB, “OMSI Provides Easy Service and Maintenance for Mobile Phones,” Press Release, 2005, 1 page.
Cybercom Group Europe AB, “The OMSI 2.0 Interface Supports,” OMSI 2.0 Description, available at least before Oct. 2008, 1 page.
Cybercom Group, “Leading Telecom Organisations Address Device Management Issues,” Press Release, 2007, 1 page.
Evgenii Masunov, Mar. 25, 2010, http://www.appleinsider.ru/news/ipone-obladaet-luchshim-tachskrinom-provereno_robotom.html, 4 pages.
Foster et al., “Automated Visual Inspection: A Tutorial” 1990 Computers Ind. Engng. vol. 18(4): 493-504.
Geekanoids, You Tube Video, “Apple iPhone 3GS Unboxing and Review”, uploaded on Jun. 19, 2009, retrieved from http://www.youtube.com/watch?v=GCEi9QAeDqk on Sep. 2, 2009.
Geyer et al. “The economics of cell phone reuse and recylcing,” The International Journal of Advanced Manufacturing Technology, 47(5): 515-525, 2010.
Graffia et al., “Retail Station for E-Device Identification, Assessment, and Trade-In”, Jun. 6, 2014 (Drawings and Specification) (Year: 2014).
GSM Arena Glossary, “LCD (Liquid Crystal Display”, retrieved from http://www.gsmarena.com/glossary.php3?term=lcd on Apr. 28, 2016, 1 page.
International Numbering Plan, www.numberingplans.com, 2 pages.
Investopedia: What's the difference between weighted average accounting and FIFO/LILO accounting methods? Aug. 19, 2010. Accessed via archive.org [https://web.archive.org/web/20100819200402/http://www.investopedia.com/ask/answers/09/weighted-average-fifo-lilo-accounting.asp].
Kanter, James Max, “Color Crack:Identifying Cracks in Glass,” dated Dec. 9, 2014; retrieved from the internet http://www.jmaxkanter.com/static/papers/color_crack.pdf on Sep. 22, 2017.
Lambert, Emily, “Use It Up, Wear It Out”, Forbes 175.5 (2005): 77-78. Business Source Complete. Web. Jan. 6, 2015, 3 pages.
Littleton Partners with Donations Ink (Jan. 19, 2006) US Fed News Service, Including US State News. Web. Jan. 6, 2015, 1 page.
MobileGazette.com, “2006 in Review: The Good, The Bad and The Ugly”, published Dec. 2006, retrieved from http://www.mobilegazette.com/2006-review-06x12x22.htm on Nov. 11, 2015.
Oliveira, et al., “Automatic crack detection on road imagery using anisotropic diffusion and region linkage,” 18th European Signal Processing Conference (EUSIPCO-2010), Aug. 23, 2010, pp. 274-278.
PC World, “Wipe Your Cell Phone's Memory Before Giving it Away”, published Jan. 2006, retrieved from http://www.washingtonpost.com/wp-dyn/content/article/2006/01/30/AR2006013001144.html on Nov. 10, 2015.
Perng et al., “A Novel Vision System for CRT Panel Auto-Inspection”, Journal of the Chinese Institute of Industrial Engineers, vol. 24, No. 5, pp. 341-350 (2007).
Perng et al., “A Novel Vision System for CRT Panel Auto-Inspection”, Proceedings of the 2005 IEEE International Conference on Mechatronics, Jul. 10-12, 2005, pp. 4.
Phifer, “How to Use your 3G Phone as a Wireless Broad Band Modem,” Computer Weekly News, 2007, 6 pages.
Rawson, Chris, “TUAW: 25 Ways to Check the Hardware on Your iPhone 4”, published Aug. 12, 2010, retrieved at http://www.tuaw.com/2010/08/13/hardware-test-your-iphone-4/ on Feb. 28, 2014.
Rehg et al. “Vision for a Smart Kiosk” IEEE, Computer Society Conference on Computer Vision and Pattern Recognition (1997).
RMS Communications Group, “RMS Communications Group Inc. opens cell phone kiosk at Ocean City Mall in Toms River, N.J.”, retrieved from http://www.prweb.com/releases/2004/11/prweb177351.htm, Nov. 12, 2004, 2 pages.
Rolf Steinhilper “Remanufacturing: The Ultimate Form of Recycling”, Fraunhofer IRBVerlag, 1998, parts 1-3, http://www.reman.org/Publications_main.htm.
Romano “Recycling a Phone at EcoATM is an Easy Route to Feeling Green,” Xconomy, Jan. 22, 2014, pp. 1-3.
Rosebrock, “How to Build a Kick-Ass Mobile Document Scanner in Just 5 Minutes” Pylmage Search, Sep. 2014, 19 pages.
Shotton et al., “Efficiently Combining Contour and Texture Cues for Object Recognition”, Proceedings of the British Machine Vision Conference 2008, (Sep. 1, 2008), pp. 7.1-7.10 * abstract *.
Shue, Jiuh-Biing et al. “Extended consumer responsibility: Syncretic value-oriented pricing strategies for trade-in-for-upgrade programs” Transportation Research Part E: Logistics and Transportation Review 122 (2019) 350-367.
SimplySellular, “Get Cash for your Old Cell Phone”, published Apr. 2, 2010, retrieved from http://simplysellular.com/conditions.php on Jan. 6, 2015, 2 pages.
Sony Ericsson Mobile Communications AB, “P800/P802,” White Paper, 2003, 128 pages.
Sony Ericsson Mobile Communications AB, “T68i/T68ie,” White Paper, 2002, 71 pages.
Tecace Software: “Your phone appraisal-Movaluate—Android Apps on Google Play”, Android Apps on Google Play, Aug. 12, 2013 (Aug. 12, 2013), XP055230264, Retrieved from the Internet <URL:https://play.google.com/store/apps/details?id=com.tecace.android.app.movaluate&hl=en>; accessed Nov. 27, 2017; 2 pages.
Trading devices for dollars, The Economist (US) 405.8813:8 (US), Economist Intelligence Unit N.A. Incorporated, Dec. 1, 2012.
Turner, “5 MP3 Players for Pumping Up Your Workouts,” Mashable.com, Nov. 4, 2010, available online at https://mashable.com/2010/11/04/mp3-players-for-sports/ (Year: 2010).
Waugh, “Phone recycling machine lets you drop in old mobiles—and spits out cash instantly,” Daily Mail Online, Jan. 13, 2012, p. 1-2.
Wikipedia, “Machine Vision” Sep. 19, 2009, 6 pages.
Wiley Encyclopedia of Computer Science and Technology (2009).
Wilson, Doug, “Liquid Crystal Display (LCD) Inspection System”, National Instruments Case Study, available May 10, 2009, retrieved from http://sine.ni.com/cs/app/cod/p/id/cs-345 on Jan. 5, 2015, 2 pages.
Wu, “Overview of Wireless Power and Data Communication” WPC/QI Developers Forum, Oct. 29, 2016, 21 pages.
Yahoo Answers, “What is a Clean ESN?” published Jun. 23, 2009, retrieved from http://web.archive.org/web/20090623215042/http://answers.yahoo.com/question/inde,8020US?qid=20080318061012AANFRco on Apr. 3, 2014.
Zhang, Yiyang, “The design of glass crack detection system based on image preprocessing technology,” 2014 IEEE 7th Joint International Information Technology and Artificial Intelligence Conference, IEEE, Dec. 20, 2014; pp. 39-42.
International Search Report and Written Opinion dated Jun. 4, 2020 in International Application No. PCT/US2020/018681, 12 pages.
Related Publications (1)
Number Date Country
20200265487 A1 Aug 2020 US
Provisional Applications (1)
Number Date Country
62807165 Feb 2019 US