Patent Application
20250237614
This application claims priority to and the benefit of Korean Patent Application No. 10-2024-0008855, filed on Jan. 19, 2024, and Korean Patent Application No. 10-2024-0191866, filed on Dec. 19, 2024, the disclosures of which are incorporated herein by reference in their entirety.
The present invention relates to an apparatus and method for measuring a defect in a molded product surface that are capable of measuring a defect in a molded product surface, which is generated during a molding process in a production line in a factory, according to objective criteria.
In many small and medium-sized factories in Korea, numerous molding processes including plastic molding, sheet metal molding, shape molding, and the like are performed,
Among molded products, there are cases in which surfaces are formed unevenly after molding.
In a case in which there are defects in surfaces after molding, there are many cases in which all other parts that have been input during subsequent processes are damaged, resulting in losses.
However, currently, the quality inspection to determine whether a molded product surface is normal or not is visually performed by workers while light is emitted on the molded product surface. That is, since the workers subjectively determine whether the molded product surface is defective, it is causing great damage in subsequent processes.
Therefore, there is a need for a method that enables a worker to measure a defect in a surface of a molded product (e.g., an inner box of a refrigerator) according to objective criteria instead of his/her experience or subjective determination.
The present invention is directed to providing an apparatus and method for measuring a defect in a molded product surface that are capable of measuring a defect in a molded product surface, which is generated during a molding process in a production line in a factory, according to objective criteria.
According to an aspect of the present invention, there is provided a molded product surface defect measuring apparatus, which includes a sensor module configured to detect a signal for performing a door closing/opening event or a lighting-on/off event of the molded product surface defect measuring apparatus, a lighting module configured to emit light required for measuring a defect in a surface of a molded product, a camera module configured to photograph the surface of the molded product, and a processor configured to control the lighting module and the camera module to perform the measuring of the defect in the surface of the molded product.
The product may be input into the molded product surface defect measuring apparatus by a robot, a machine, or a facility.
When the product is completely input into the molded product surface defect measuring apparatus and a door of the molded product surface defect measuring apparatus is closed, the molded product surface defect measuring apparatus may receive a door closing event signal from the sensor module and control the lighting module.
When the product is completely input into the molded product surface defect measuring apparatus, the processor may control the lighting module and the camera module to emit light on the surface of the molded product, photograph the surface of the molded product using one or more cameras, extract data, and at the same time, determine whether the surface of the molded product is normal or abnormal through data analysis.
The processor may perform a lighting-on function of the lighting module to support the photographing of the camera module by emitting light on an inner surface of the product, activate a surface defect inspection function, and operate the camera module to adjust a focus of each camera.
In order to adjust the focus of each camera, the processor may receive a predetermined button of a keyboard from a user as an input or activate the surface defect inspection function, and at the same time, automatically operate a camera focus adjustment function.
When adjustment of a focus of each camera of the camera module is completed and a scratch is detected from the product, the processor may control the scratch to be more clearly photographed by adjusting an angle of each camera or adjusting light brightness through a lighting brightness adjustment function for distinguishing a sharpness of the scratch.
The lighting brightness adjustment function for distinguishing the sharpness of the scratch may be a function for adjusting the light brightness in order to ensure that a camera image of the scratch is captured well, and may be a function in which the processor uses a dimming function of lighting to find an optimal illuminance value by increasing the light brightness by a predetermined unit brightness starting from a predetermined default brightness up to a maximum brightness.
The processor may collect image data through the camera module, then perform normal data learning to generate an image data model that extracts abnormal data and compare a plurality of pieces of image data newly input to the image data model through an image data model comparison function to determine whether the molded product surface is normal or abnormal.
The processor may display an abnormality on a monitoring device so that a worker performs a product check and determination process when it is determined that the molded product surface is abnormal through the image data model comparison operation, and terminates the inspection process when it is determined that the molded product surface is normal, wherein when displaying the abnormality on the monitoring device, the processor displays a location of the defect with a predetermined color in a region in which the abnormality has occurred so that the worker visually recognizes rapidly and easily where and what type of defect has occurred in the product.
In the lighting module, in order to prevent a light scattering phenomenon of lighting from being mis-recognized as the defect in the molded product surface when the defect in the surface of the molded product is measured, a bulb-type light and a directional light may be used together.
According to another aspect of the present invention, there is provided a molded product surface defect measuring method, which includes detecting by, a processor of a molded product surface defect measuring apparatus, a signal for performing a door closing/opening event or a lighting-on/off event of the molded product surface defect measuring apparatus using a sensor module, and controlling a lighting module and a camera module to perform measurement of a defect in a surface of a molded product when the signal is detected using the sensor module.
In the controlling of the lighting module, when the product is completely input into the molded product surface defect measuring apparatus and a door of the molded product surface defect measuring apparatus is closed, the processor may receive a door closing event signal from the sensor module and control the lighting module.
When the product is completely input into the molded product surface defect measuring apparatus, the processor may control the lighting module and the camera module to emit light on the surface of the molded product, photograph the surface of the molded product using one or more cameras, extract data, and at the same time, determine whether the surface of the molded product is normal or abnormal through data analysis
In the controlling of the lighting module and the camera module, the processor may perform a lighting-on function of the lighting module to support the photographing of the camera module by emitting light on an inner surface of the product, activate a surface defect inspection function, and operate the camera module to adjust a focus of each camera.
In order to adjust the focus of each camera, the processor may receive a predetermined button of a keyboard from a user as an input or activate the surface defect inspection function, and at the same time, automatically operate a camera focus adjustment function.
When adjustment of a focus of each camera of the camera module is completed and a scratch is detected from the product, the processor may control the scratch to be more clearly photographed by adjusting an angle of each camera or adjusting light brightness through a lighting brightness adjustment function for distinguishing a sharpness of the scratch.
The lighting brightness adjustment function for distinguishing the sharpness of the scratch may be a function for adjusting the light brightness in order to ensure that a camera image of the scratch is captured well, and may be a function in which the processor uses a dimming function of lighting to find an optimal illuminance value by increasing the light brightness by a predetermined unit brightness starting from a predetermined default brightness up to a maximum brightness.
In the controlling of the lighting module and the camera module to perform the measurement of the defect in the surface of the molded product, the processor may collect image data through the camera module, then perform normal data learning to generate an image data model that extracts abnormal data, and compare a plurality of pieces of image data newly input to the image data model through an image data model comparison function to determine whether the molded product surface is normal or abnormal.
In the controlling of the lighting module and the camera module to perform the measurement of the defect in the surface of the molded product, the processor may display an abnormality on a monitoring device so that a worker performs a product check and determination process when it is determined that the molded product surface is abnormal through the image data model comparison operation, and terminates the inspection process when it is determined that the molded product surface is normal, wherein when displaying the abnormality on the monitoring device, the processor displays a location of the defect with a predetermined color in a region in which the abnormality has occurred so that the worker visually recognizes rapidly and easily where and what type of defect has occurred in the product.
The above and other objects, features and advantages of the present invention will become more apparent to those of ordinary skill in the art by describing exemplary embodiments thereof in detail with reference to the accompanying drawings, in which:
Hereinafter, examples of an apparatus and method for measuring a defect in a molded product surface according to embodiments of the present invention will be described.
In this process, thicknesses of lines, sizes of components, and the like shown in the accompanying drawings may be exaggerated for clarity and convenience of description. Further, some terms which will be described below are defined in consideration of functions in the present invention and meanings may vary depending on, for example, a user or operator's intentions or customs. Therefore, the meanings of these terms should be interpreted based on the scope throughout this specification.
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that the embodiments can be easily performed by those skilled in the art. However, embodiments of the present invention may be implemented in several different forms and are not limited to embodiments described herein. In addition, parts irrelevant to description are omitted in the drawings in order to clearly explain embodiments of the present invention. Similar parts are denoted by similar reference numerals throughout this specification.
Throughout this specification, when a certain part “includes” a certain component, it means that another component may be further included rather than other components being excluded unless otherwise stated.
Implementations described herein may be implemented, for example, as a method, a process, a device, a software program, a data stream, or a signal. Although discussed only in the context of a single form of implementation (e.g., only as a method), implementations of the features discussed may also be implemented in other forms (e.g., devices or programs). The device may be implemented as appropriate hardware, software, firmware, etc. The method may be implemented in a device such as a processor, which is generally a processing device including a computer, microprocessor, integrated circuit, or programmable logic device.
Generally, in a molding process, when there is the quality of the surface of a portion that a customer directly touches after molding, the quality of the surface is linked to the image of the entire brand of the product and affects the overall brand image. Therefore, the quality of the surface that the consumer directly touches should be thoroughly managed.
However, when a worker visually inspects a molded product surface, the worker may miss a defect in the molded product surface, and because the worker has to directly see and determine hundreds of products (e.g., refrigerators) produced each day, physical and mental fatigue can reduce overall work efficiency.
Therefore, the present invention provides a method in which the existing method of subjectively determining the quality of a molded product surface by a worker can be improved so that a defect in the molded product surface can be objectively determined regardless of the worker. The present invention provides a technology in which light is emitted on a surface of a molded product, the surface of the molded product can be photographed using one or more cameras, and at the same time, whether the surface of the molded product is normal or abnormal can be determined through data analysis.
As illustrated in
The sensor module 110 detects a signal for performing a door closing/opening event of a product (e.g., a refrigerator), a lighting-on/off event, or the like.
For example, the sensor module 110 may include a proximity sensor, a magnetic sensor, a pressure sensor, etc.
The lighting module 120 includes at least one light that emits light required for inspection (i.e., defect measurement) of a surface of the molded product (e.g., the refrigerator).
The camera module 130 includes at least one camera for photographing the surface of the molded product (e.g., the refrigerator).
The processor 140 controls the lighting module 120 and the camera module 130 on the basis of the signal detected through the sensor module 110 to perform inspection (i.e., the defect measurement) on the surface of the molded product (e.g., the refrigerator).
The molded product surface defect measuring apparatus 100 may input the product (e.g., the refrigerator) into the molded product surface defect measuring apparatus 100 by a robot 200 (e.g., a machine or a facility).
The robot 200 may be controlled through communication with the processor 140 or may be controlled by a separate external robot controller (not illustrated).
The robot 200 may open a door (not illustrated) of the molded product surface defect measuring apparatus 100 and input the product (e.g., the refrigerator) thereinto. In this case, a person may intervene, and when the input of the product is completed, the robot 200 may close the door (not illustrated) of the molded product surface defect measuring apparatus 100.
When the product is input into the molded product surface defect measuring apparatus 100, the processor 140 controls the lighting module 120 and the camera module 130 to emit light on the surface of the molded product, photograph the surface of the molded product using at least one camera, extract data, and at the same time, determine whether the surface of the molded product is normal or abnormal through data analysis.
Referring to
In this case, the product may be manually input by a person or may be input using a machine or a facility during a production process.
The processor 140 receives a door closing event signal through the sensor module 110.
When the door closing event signal is received (Yes in S103), the processor 140 controls the lighting module 120.
The processor 140 performs a lighting-on function of the lighting module 120 to support the photographing of the camera module 130 by emitting light on an inner surface of the product (S104).
The processor 140 activates a surface defect inspection program (function) (S105).
The processor 140 operates a plurality of cameras (i.e., n cameras) (S106) and then adjusts a focus of each camera (S107).
For example, since a proximity distance to a molded product surface is different depending on an orientation of each camera, it is necessary to adjust the focus of each camera.
In this case, images captured by the plurality of cameras (i.e., n cameras) may be displayed on a monitoring device (e.g., a user monitor) (not illustrated) through the surface defect inspection program (function), as illustrated in
For example, as illustrated in
The processor 140 controls the plurality of cameras (i.e., n cameras) simultaneously through a camera focus adjustment function to adjust a focus of each camera at its own location.
In this case, the camera focus adjustment function may be configured to instruct a user to adjust the focus of each camera by pressing a predetermined button (e.g., a button) of a keyboard, or to automatically operate the camera focus adjustment function upon activation.
When the adjustment of the focus of the camera is completed, the processor 140 controls an angle or brightness of the corresponding camera to make a scratch more clearly visible through a lighting brightness adjustment function for distinguishing a sharpness of a scratch when the scratch is detected from the product (S108).
Here, the lighting brightness adjustment function for distinguishing the sharpness of the scratch is a function for ensuring that a camera image of a scratch can be captured well, and is a function for finding an optimal illuminance value by increasing the light brightness by a predetermined unit brightness (e.g., 50 lx) starting from a predetermined unit brightness (e.g., 200 lx) up to a maximum brightness (e.g., 3,500 lx) using a dimming function of the lighting to change the light brightness.
Referring to
After the processor 140 initializes the lighting with the default brightness (e.g., 200 lx), when the brightness (current illumination of the lighting) is less than or equal to the maximum brightness (e.g., 3,500 lx), the processor 140 increases the unit brightness (e.g., 50 lx) (S108-3) and operates an image capturing scratch sharpness distinction algorithm.
When the processor 140 performs the lighting brightness adjustment function for distinguishing the sharpness of the scratch, the processor 140 may use a normal/abnormal distinction algorithm that is pre-learned using an image data model comparison function, which is a higher function of the image capturing scratch sharpness distinction algorithm (S108-4).
However, the present invention is not limited to this algorithm.
When a scratch-like shape is found (Y of S108-5), the processor 140 stores a current illumination as an optimal brightness, determines that the product is a defective product containing a scratch, and terminates this process.
More specifically, when the scratch-like shape is not found (N of S108-5) and the current illuminance is greater than or equal to the maximum brightness (e.g., 3,500 lx) (N of S108-3), the processor 140 determines that the product is a normal product without scratches and terminates this process.
Further, when the scratch-like shape is found (Y of S108-5) and the current illuminance is less than the maximum brightness (e.g., 3,500 lx) (Y of S108-6), the processor 140 stores a current illuminance value (i.e., a lx value) in a memory (not illustrated) and terminates this process (S108-7).
However, when the scratch-like shape is not found (N of S108-5), the processor 140 increases the unit brightness (e.g., 50 lx) again and operations S103-3 to S108-5 may be repeated.
Referring to
Here, the concept of a memory (not illustrated) includes a hard disk, a solid-state drive (SSD), a random access memory (RAM), a read-only memory (ROM), a cloud storage, etc.
The processor 140 performs normal data learning to generate an image data model that extracts abnormal data (S111 and S112).
For example, the image data model may be generated by improving the algorithm performance based on an end-to-end autonomous driving model (UniAD) algorithm. However, the present invention is not limited to the method of generating the image data model.
The processor 140 compares a plurality of pieces of image data (i.e., n pieces of image data) newly input to the image data model through the image data model comparison function to determine whether the molded product surface is normal or abnormal (S113).
For reference, according to the test, as illustrated in
When the processor 140 determines that the molded product surface is abnormal through the image data model comparison operation (Y of S114), the processor 140 displays the abnormality on a monitoring device (e.g., a user monitor) (S115), causes a product check and determination function to be performed by a worker (S116), and when the processor 140 determines that the molded product surface is abnormal (N of S114), terminates the inspection process.
In operation S115 of displaying the abnormality on the monitoring device (e.g., the user monitor), the processor 140 displays a location of the defect with a predetermined color (e.g., red) in a region in which the abnormality (defect) has occurred, as shown in
The product check and determination function performed by the worker refers to a function in which the worker, who has confirmed the location of the defect with the abnormality display function on a screen of a previous worker, to visually check and determine whether an actual defect has occurred in the product.
For example, one inspection device is formed on each of upper and lower portions of the device (or the frame) so that two inspection devices are attached, and eight cameras are attached to each device so that 16 cameras are all attached. Further, a plurality of lights, such as light bulbs, light emitting-diodes (LEDs), stick-shaped lights, etc., that emit light to the inside of each inspection device are attached to control brightness of the inside of the molded product.
However, the present invention is not limited to an external shape of the device (or the frame) of the molded product surface defect measuring apparatus 100 (i.e., the inspector) and the shape and number of the inspection devices attached thereto, and the size and shape of the device of the molded product surface defect measuring apparatus 100 may be modified to correspond to the size and shape of the product to be inspected.
In this case, the inspection devices may be designed with more than 10 axes so that the inspection devices can move up/down, leftward/rightward, or forward/backward, an entire body can move leftward/rightward, and the lights can move forward/backward, and may be adjusted automatically/manually.
For example, in the case of a product (e.g., a refrigerator plastic molded product), scratches in a front region of the device that is in the consumer's line of sight should be measured, and thus a depth of the scratch inspection device is adjusted to allow the locations of the upper/lower and left/right cameras to be moved and changed, and each camera may also be moved and changed to a desired photographing region. Each camera photographs its own designated location.
Here, T stands for a top, L stands for a left, R stands for a right, and B stands for a bottom.
TL denotes a left of a top, TR denotes a right of the top, LT denotes a top of a left, LB denotes a bottom of the left, RT denotes a top of a right, RB denotes a bottom of the right, BL denotes a left of a bottom, and BR denotes a right of a bottom, and thus a total of eight points may be photographed simultaneously.
In order to prevent the light scattering phenomenon of the light, in the present embodiment, a bulb-type light and a directional LED light are also used together.
For example, an internal material of the refrigerator molded product is a plastic glossy mold that reflects light well, and thus, when the light is reflected, the light scattering phenomenon may be mis-recognized as a defect in the molded product surface, and thus the lighting plays an important role.
Therefore, a directional light should be used so that light is not directly emitted onto the camera.
In
Currently, workers' quality inspection is not objective and is subjective in that the workers visually determine whether a product is normal or not while light is emitted.
Accordingly, the subsequent processes cause greater damage to the factory.
Therefore, the present embodiment has an effect of improving productivity by enabling objective, concise, and rapid production quality check by enabling defect detection from images obtained by photographing the inner surface of the refrigerator one time and is advantageous for commercialization because the present embodiment reflects the requirements of the field. Further, the present embodiment has an effect of allowing 16 surfaces to be photographed at the same time for fine scratches that are difficult to visually check and rapidly checking results of image analysis.
The present invention has an effect of enabling measurement of a defect in a molded product surface, which is generated during a molding process in a production line within a factory, according to objective criteria.
While the present invention has been described with reference to embodiments illustrated in the accompanying drawings, the embodiments should be considered in a descriptive sense only, and it should be understood by those skilled in the art that various alterations and other equivalent embodiments may be made.
Therefore, the scope of the present invention should be defined by only the following claims. Further, implementations described herein may be implemented, for example, as a method, a process, a device, a software program, a data stream, or a signal. Although discussed only in the context of a single form of implementation (e.g., only as a method), implementations of the features discussed may also be implemented in other forms (e.g., devices or programs). The device may be implemented as appropriate hardware, software, firmware, etc. The method may be implemented in a device such as a processor, which is generally a processing device including a computer, microprocessor, integrated circuit, or programmable logic device.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10-2024-0008855 | Jan 2024 | KR | national |
| 10-2024-0191866 | Dec 2024 | KR | national |
