Patent Application
20230023311
In remote sensing/aerial imaging industry, imagery may be used to capture views of a geographic area in order to measure objects and/or structures within the images. These are generally referred to as “geo-referenced images” and come in two basic categories:
Vertical Imagery—images captured with a camera pointed vertically downward thus generally capturing the tops of structures; and,
Oblique Imagery— images captured with a camera aimed at an angle capturing the sides, as well as, tops of structures.
Most vertical imagery may be processed in order to fit a mathematically rectangular projection or map. This process is known as ortho-rectification and attempts to create an appearance as if the sensor were directly above each pixel in the image. The resulting image is known as an ortho-photo. Since the images are mathematically projected, they may be combined into a seamless mosaic resulting in a composite image known as an ortho-mosaic. The term ‘ortho image’ is used to denote a geo-referenced image that is either an ortho-photo image or an ortho-mosaic image.
Because they are captured looking straight down, an ortho-photo or othro-mosaic contains a view of world which many are not accustomed. As a result, there may be difficulty in distinguishing between two different properties as the only portions of the structures visible in the ortho-mosaic are roof tops. An oblique image, in contrast, is captured at an angle showing sides of objects and structures.
Aerial imagery may be used in identification of dimensions of buildings or structures. Assessors generally rely on dimensions of a building to assess a value to that building. In many cases, dimensions of the building may be determined by measuring a building's footprint provided by the base of that building.
Traditional ortho-rectified imagery, however, does not reveal the base of the building, but instead reveals the edge of the roof. For example, ortho-rectified imagery may only show edges or eaves of a roof and not the base of a building. For commercial buildings with no overhangs, this may not create an issue as the edge of the roof may be the same footprint of the building.
In contrast, residential homes and other similar buildings may have eaves extending from the roof line beyond walls of a building. As such, when extracting building “footprints” from ortho-rectified imagery, a “hat-print” is created, not a footprint. That is, the resulting outline may be representative of an edge of eaves projected down to the ground adding additional square footage to the dimensions. Relatively standard eaves may even add one to two feet to the dimensions of a house in all directions. Thus, a 25′ by 30′ house may increase in measurement to 28′ by 33′, thereby increasing measured square footage from 750 to 924 square feet. This creates a 23% error in the measured size of that house.
To be able to see under the edges of the roof, imagery may need to be captured at an angle. For example, imagery may be captured using oblique aerial imagery. These oblique aerial images, however, may only see two sides, or at most three sides of a building. In order to generate a footprint of a building using oblique aerial imagery, each side of the building may need to be found and then fitted together to form the entire footprint.
To assist those of ordinary skill in the relevant art in making and using the subject matter hereof, reference is made to the appended drawings, which are not intended to be drawn to scale, and in which like reference numerals are intended to refer to similar elements for consistency. For purposes of clarity, not every component may be labeled in every drawing.
Before explaining at least one embodiment of the disclosure in detail, it is to be understood that the disclosure is not limited in its application to the details of construction, experiments, exemplary data, and/or the arrangement of the components set forth in the following description or illustrated in the drawings.
The disclosure is capable of other embodiments or of being practiced or carried out in various ways. Also, it is to be understood that the phraseology and terminology employed herein is for purpose of description and should not be regarded as limiting.
As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Further, unless expressly stated to the contrary, “or” refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by anyone of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).
In addition, use of the “a” or “an” is employed to describe elements and components of the embodiments herein. This is done merely for convenience and to give a general sense of the inventive concept. This description should be read to include one or more and the singular also includes the plural unless it is obvious that it is meant otherwise.
Further, use of the term “plurality” is meant to convey “more than one” unless expressly stated to the contrary.
Finally, as used herein any reference to “one embodiment” or “an embodiment” means that a particular element, feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.
Referring now to the drawings, and in particular to
In some embodiments, system 10 may be distributed, and include a host system 12, communicating with one or more user devices 14 via a network 16. Network 16 may be the Internet and/or other network. Host system 12 may include one or more servers 18 configured to communicate with network 16 via one or more gateways 20. If network 16 is the Internet, a primary user interface of system 10 may be delivered through a series of web pages. It should be noted that the primary user interface of system 10 may be replaced by another type of interface, such as a Windows-based application (e.g., deploying system 10 in a stand-alone environment such as a kiosk).
Network 16 may be almost any type of network. For example, network 16 may be an Internet and/or Internet 2 network. In one embodiment, network 16 exists in an Internet environment (e.g.,) TCP/IP-based network. It is conceivable that in the near future, embodiments of the present disclosure may use more advanced networking topologies.
Servers 18 may be networked with a LAN 30. Gateway 20 may be an entity responsible for providing access between the LAN 30 and network 16. In some embodiments, gateway 20 may also be used as a security means to protect LAN 30 from attack from external networks such as network 16.
In some embodiments, LAN 30 may be based on a TCP/IP network (e.g., the Internet), or may be based on another underlying network transport technology. For example, LAN 30 may include an Ethernet network with TCP/IP because of the availability and acceptance of underlying technologies, but other embodiments may use other types of networks such as Fibre Channel, SCSI, Gigabit Ethernet, and/or the like
Host system 12 may include one or more servers 18. Configuration of the server hardware may be dependent on requirements and needs of the particular embodiment of system 10. For example, host system 12 may include multiple servers 18 with load balancing to increase stability and availability. Servers 18 may include database servers and application/web servers. In one embodiment, database servers may be separated from application/web servers for availability and/or to provide database servers with increased hardware and/or storage.
User devices 14 may be any number and/or type of devices. For example, user device 14 may involve a user 32, using a computer 34 with a display 36, keyboard 38, and mouse 40. Display 36 may be a single monitor or multiple adjacent monitors. In some embodiments, user device 14 may include a type of software called a “browser” 42 to render code content (e.g., HTML/XHTML) generated when requesting resources from a source (e.g., host system 12). Additionally, in some embodiments, system 10 may be designed to be compatible with major Web Browser vendors (e.g., Microsoft Internet Explorer, Google Chrome, Mozilla Firefox, and Opera). Other embodiments may focus on one particular browser depending upon the common user base using system 10.
User devices 14 may be implemented as a portable device such as a laptop computer 50 (or handheld computer); a cellular telephone 52 with a micro or embedded Web Browser; a Portable Digital Assistant 54 (PDA) capable of wireless network access; a pen-based and/or tablet computer 56. In some embodiments, user device 14 may be a cable box 60 and/or other similar device for viewing through a display 62 and/or television. Current embodiments of system 10 may also be modified to use any of these or future developed devices.
System 10 may be designed to provide flexibility in deployment. Depending upon the requirements of the particular embodiment, instructions may be designed to work in almost any environment such as a desktop application, a web application, and/or even simply as a series of web services designed to communicate with an external application.
Hardware and system software may be designed with two key concerns; flexibility and scalability. Although some specifics for software and hardware components may be mentioned herein, it will be understood that a wide array of different components may be substituted, such as using different database vendors and/or even replacing the databases with XML-based document stores.
When system 10 is used to execute the logic of the processes described herein, such computer(s) and/or execution may be conducted at a same geographic location or multiple different geographic locations. Furthermore, the execution of logic may be conducted continuously or at multiple discrete times. Further, the execution of the logic can be implemented on one or more of the servers 18 of the host system 12, the user devices 14 and combinations thereof.
In general, system 10 may be capable of displaying and navigating geo-referenced imagery, such as aerial oblique imagery or aerial orthogonal imagery. The geo-referenced imagery may be represented by a single pixel map, and/or by a series of tiled pixel maps that when aggregated recreate the image pixel map. The geo-referenced imagery can be stored in a non-transitory memory in one or more electronic files that can be rendered into a picture. The electronic files can be any suitable format, such as JPEG, BMP, TIFF or the like.
System 10 will be described by way of example utilizing aerial geo-referenced images as the geo-referenced imagery shown on the display 36 of the computer 34. However, it should be understood that system 10 may use other types of geo-referenced images, such as architectural images.
Referring to
By using oblique imagery to extract a height measurement, a volume of a structure having footprint 102 may also be calculated and/or averaged. For example, story heights of a structure with footprint 102 may be used in determining the number of floors of that structure, and thus, additional square footage of an upper living area may also be added into calculations.
Further, if roof lines of a structure have been determined, (e.g., by systems and methods as described in U.S. patent application Ser. No. 12/909,692, which is hereby incorporated by reference in its entirety), then footprint 102 may be extruded upwards until footprint 102 meets roof lines of that structure and a full three-dimensional model of that structure may be generated. The three-dimensional model may then become the basis of a virtual property model containing not only information about roof, living area, and/or side walls, but additional information from multiple data sources may be attached to the data record providing additional uses including, providing bids for remodeling and/or additional capabilities as described in European Application No. 99123877.5 filed on Feb. 12, 1999, and Publication No. EP1010966 filed on Feb. 12, 1999.
Information provided by footprint 102 may be used within industries including, but not limited to, assessment industries, pipeline industries, roofing industries, and the like. For example, footprint 102 may be used within the pipeline industry to determine potential impact of a pipeline leak rupture through classification of high consequence areas.
Alternatively, geo-referenced images 300 may be obtained using oblique terrestrial imagery. For example, in some embodiments, images may be obtained using oblique terrestrial imagery if images are capable of being measured upon (metric) and/or determined to reveal base of structure 106.
In some embodiments, oblique geo-referenced images 300 may be provided using techniques described in U.S. Pat. No. 7,424,133, which is hereby incorporated by reference in its entirety. Geo-referenced images 300 may be oblique images having stored geo-referenced parameters. Geo-referenced images 300 and parameters when combined with a tessellated ground plane may provide a determination of pixel location in real world coordinates such as latitude/longitude in geo-referenced image 300 using solely that image. Further measurements may then be made by calculating differences in pixel location between points of measurements. For example, for distance measurements, calculations may be determined using circle routes and/or across a terrain by tracing a route along a tessellated ground plane.
Referring to
Location of structure 106 may be provided from a variety of sources. In some embodiments, geo-referenced images 300 of structures 106 or any additional images of structures 106 may be selected by user 32 as indicated by reference number 250 in
In some embodiments, system 10 may use an application (e.g., software application) to evaluate and select geo-referenced image 300 containing structure 106, as well as corresponding geo-referenced images from other cardinal directions for structure 106 as indicated by reference number 253 in
Once targeted location and structure 106 is selected, system 10 may select geo-referenced oblique images 300 to show all facets of structure 106 as indicated by reference number 257.
Referring to
Any suitable edge detection algorithm capable of finding the edges 108, 110 and 112 of structure 106 may be used to locate edges within geo-referenced image 300b. For example, a LoG algorithm may be used to provide definition of edges 108, 110 and 112 of structure 106. System 10 stores each edge 108, 110 and 112 in a table in an electronic format with geo-referencing information, image information, and/or confidence value as indicated by a reference numeral 249 of
In some embodiments, a color clustering algorithm (e.g., YaW algorithm) may be used in addition to, or in lieu of, an edge detection algorithm. Color clustering algorithms may provide spectral content of each pixel in geo-referenced image 300b and grouping of adjacent pixels having similar spectral content. Further, by providing multiple algorithms, one or more algorithms may aid in eliminating misclassifications by another algorithm.
System 10 may assign a value indicative of a level of confidence an algorithm achieved when determining whether edges 108, 110 and 112 are present in geo-referenced image 300b as indicated by reference number 259 in
Confidence values may provide for classification of detected edges as: (1) good edges, (2) likely edges, (3) unlikely edges, and/or (4) discarded edges as indicated by reference numeral 261 in
In using multiple algorithms, if a detected edge scores a high confidence value in one algorithm and a low confidence value in a separate algorithm, the detected edge may be considered a likely edge. If a detected edge scores a high confidence value in one algorithm but is non-existent in another algorithm, the detected edge may be considered an unlikely edge. If a detected edge scores a low confidence value in multiple algorithms, the detected edge may be considered an unlikely edge.
In downstream processing, generally, good edges may initially be used. If there is insufficient data to create footprint data using solely good edges, likely edges may be considered. If, after all good edges and likely edges are considered, there is still insufficient data to create footprint data, the process may indicate a failure. If the process indicates a failure, an operator may manually provide additional data regarding the edges 108, 110 and/or 112 of structure 106.
Referring to
Edge detection of roof 132 may provide a bound box area 134 to be applied to oblique geo-referenced images 300 and/or in validation of edges 108, 110 and 112 as indicated by reference numbers 263-264 of
Referring to
Vertical edges 108 may be provided in a list and sorted by relative position from a first point to a second point along structure 106a. For example, vertical edges 108 may be sorted by relative position from left (point A) to right (point B) along structure 106a. Similarly, horizontal edges 110 may be provided in a list and sorted by relative position from a first point to a second point. For example, horizontal edges 110 may be sorted by relative position from bottom (point C) of structure to top (point D) of structure 106a. It should be noted that direction of sorting is arbitrary, fixed or combinations thereof.
Referring to reference numbers 269-270 of
Referring to
Referring to
An angle of intersection Θ may be determined between horizontal edge 110 and vertical edges 108 at vertices 114a and 114b. Angle of intersection Θ may be approximately a 90° angle or another angle. If angle of intersection Θ is not approximately a 90° angle, another horizontal edge 110 extending between vertices 114a and 114b may be located and/or used.
Referring to
In providing for extended edges, edge 110 of roof 132 may first be determined. To provide edge 110 of roof 132, a listing of vertical edges 108 may be searched to determine the longest vertical edge. The longest vertical edge typically connects to horizontal edge 110 of roof 132.
With location and position of edge 110 of roof 132 determined, a listing of horizontal edges 110 may then be searched finding at least one collinear horizontal edge to edge 110 of roof 132 (e.g., horizontal edges 110a and 110b). An extension 116 may be made between one horizontal edge segment (e.g., horizontal edge 110a) to connect to another horizontal edge segment (e.g., horizontal edge 110b) and/or extend a horizontal edge segment to connect to vertical edges 108. This method assumes that collinear horizontal edges 110a and 110b are detected from horizontally extending objects located within the geo-referenced image(s) 300, such as one or more windows, doors, fascia, or similar structures. Collinear horizontal edges 110a and 110b may then be extended between vertical edges 108 to provide line segment 104 as indicated by the dashed lines in
Additionally, angle of intersection Θ may also be identified on collinear horizontal edges 110a and/or 110b. If angle of intersection Θ is not approximately 90 degrees, another collinear horizontal edge 110 may be identified and used.
Once suitable horizontal edge 110 is determined to extend from first vertex 114a to second vertex 114b and/or include angle of intersection Θ of approximately 90 degrees, e.g., within + or − 5 degrees. A relative length L extending along horizontal edge 110 may be determined. Measurements may be made at base 105 of structure 106. In some embodiments, having measurements at base 105 of structure 106 may reduce error from miscalculations based on eaves/roof of structure 106, and/or the like.
To determine relative length L, vertical edges 108 may be traced downward the same number of pixels until whichever vertical edge 108 ends first. By adjusting vertical pixel location, at least two alterations may be made. First, a line connecting both vertices 114 may be substantially parallel to horizontal line 110 of roof 132. Second, location of position closest to base 105 of structure 106 may be determined.
Having vertices 114a and 114b at the substantially similar vertical pixel location, geo-referencing information associated with images and vertical pixel locations may be used to calculate a relative length L of line segment 104. In particular, the location of each vertice 114 can be determined using a single image with geo-referencing information and a tessellated ground plane as described in detail in U.S. Pat. No. 7,424,133, which is expressly incorporated herein by reference. The geographical locations of the vertices 114a and 114b can also be determined using two geo-referenced images (preferably captured at different times and from different locations) with stereo photogrammetry techniques. Once the geographic locations of the vertices 114a and 114b have been determined, the relative length L can then be determined using a number of approaches including, but not limited to, the Gaussian formula for determining distance between two geographic locations that are shown in the same geo-referenced image or geo-referenced images. The term “relative length” as used herein refers to both of the vertices 114a and 114b being shown in the one or more geo-referenced images used to calculate the geographic locations of the vertices 114a and 114b.
Referring to
In structures 106 with multiple line segments 104, vertical edges 108 may be searched to determine if there are additional vertical edges 108 between the outermost vertical edges. If additional vertical edges 108 are present, two vertical edges 108 between the outermost vertical edges may be selected and the process repeated to provide for additional line segments 104. For example,
Each compass direction may include one set of line segments 104. For example, in
Referring to
A final geographic location may be assigned to footprint 102. In one embodiment, geography of footprint 102 may be determined by averaging location between similar vertices 114 in geo-referenced images. For example, a centroid of a single vertex (e.g., vertex 114a) may be calculated by finding an average of all locations of that vertex across all geo-referenced images. Opposing side dimensions may be averaged to determine proper dimensions (e.g., North side of structure 106 and South side of structure 106 may be averaged, and East side of structure 106 and West side of structure 106 may also be averaged). A single vertex 114 may be held at a constant and other vertices 114 may be adjusted accordingly. The centroid may be found by averaging the adjusted vertices 114. Distance and direction between two centroids may be calculated (e.g., Gaussian algorithm) such that all vertices 114 may be offset centering footprint 102 of structure 106.
For non-simple structures, each set of vertices 114 from each image may be assembled. If only vertices 114 from a single image are used, gaps between line segments 104 may exist. Such gaps may be filled in by identifying line segments 104 using other geo-referenced images (e.g., line segments 104 identified using an oblique image showing North side of a structure may be filled in using an oblique image showing South side of the structure).
Line segments 104 for an entire length of a non-rectangular structure may be averaged with its opposing wall (e.g., North/South, East/West). For example, if a composite wall averages out to be 100′, but it is made of three line segments 104 of 42′, 36′ and 28′ for a total of 106′, each line segment 104 may be compressed by roughly 6%. This may yield segments of 39.6′, 34′, and 26.4′ for a total length of 100′.
Referring to
In various aspects, the set of instructions discussed above can be distributed or used in a variety of manners. For example, one or more computer readable medium storing the set of instructions could be sold and/or distributed through retail locations as a set of one or more CD-ROMs or downloaded from a server. The term “sold” as used herein includes a sale where ownership is transferred, as well as an exchange of funds where a license or rights are granted but ownership has not changed. As another example, the set of instructions could be made available to the processor for execution in a variety of manners, such as by installing the set of instructions onto a local hard drive or memory, or by having the processor access a remote server or memory providing the set of instructions.
The set of instructions can be used in various manners to generate reports/information that can be used in a variety of industries, as set forth below. The information regarding the footprint 102 can be used to determine the size, area, layout, and/or shape of the footprint 102. Information about the footprint 102 can be assessed or compared with a separately created sketch of the footprint 102 to verify the accuracy of the separately created sketch, or vice-versa.
Information about the footprint 102 can also be used to determine encroachment/compliance of the structure 106 with various rules and regulations. For example, boundaries of the footprint 102 can be placed in a correct geographic location on a site plan or survey having one or easements, buildings/structures (such as electrical lines or gas pipes) and/or property boundaries to visually indicate any overlap.
The system 10 may allow interactive user calculations on the footprint 102.
Further, the footprint 102 can be used to calculate total living/usable area of the structure 106. In this regard, areas encompassed by the walls of the structure 106 can be automatically removed from the calculation. For example, if the structure 106 is a rectangle having outer dimensions of 45 feet×75 feet, and the thickness of the walls is one foot, then the livable/usable area would be 43 feet×73 feet=3139 square feet. Logic can be provided for estimating the thickness of the walls based on various matters, such as type of construction, type of wall, type of building and the like. One or more default value(s) for the thickness of the walls may also be stored and used. The system 10 may also have logic for determining/receiving a number of floors to be used in the calculation of the livable/usable area. Information indicative of the size, shape, area, and/or livable/usable area can be incorporated into a report. For example, such information can be incorporated into an electronic display, an electronic report, a hard-copy of the report or the like. For example, such information can be incorporated into the report shown in
The livable/usable area can be provided to property assessors and used to determine tax assessments or the like. Information regarding the footprint 102 can also be used to generate an accurate three-dimensional model of the structure 106, including a three-dimensional model of a roof of the structure 106. The three-dimensional model of the roof of the structure 106 can be determined using any suitable technique, such as disclosed in U.S. Pat. No. 7,509,241, the entire content of which is hereby incorporated by reference.
Information indicative of the footprint 102 can also be used to determine the size (including area) and shapes of one or more walls of the structure 106. This can be accomplished by calculating a height of the roof at one or more locations on the structure 106 and then projecting upward from the footprint 102 to the roof of the structure. The sizes and shapes of one or more walls can be used in reports for estimating materials for reconstruction, modification and/or maintenance of the structure 106. Furthermore, the system 10 may also be used to calculate the cumulative area of the walls of the structure 106. For example, the cumulative areas of the walls of the structure 106 can be used for estimating an amount of siding or paint needed for renovating/repairing the walls of the structure 106.
The volume of the structure 106 may also be estimated and included within a report. For example, the volume of the structure 106 can be used by a heating, ventilation, and air conditioning (HVAC) contractor to estimate the size of a suitable heating, ventilation, and air conditioning system for the structure 106.
The information indicative of the footprint 102 can be used for thermal ratings and/or material estimates in construction and/or reconstruction projects by architects, engineers and/or construction workers; building information modeling applications; sketch verification applications; high consequence area/encroachment compliance; total living area calculations/applications; finished siding reports and/or applications; paint estimate reports and/or applications; roof reports and/or applications; solar reports and applications (including a combination of building footprint and roof three-dimensional model); and accurate three-dimensional models of the structure for building information management applications.
Information indicative of footprint 102, area, and/or volume of the structure 106 can be used by disaster recovery agencies, such as the Federal Emergency Management Agency, or by insurance agencies to estimate the amount of debris created in a disaster and/or the cost of the disaster after one or more buildings has been moved, removed, distorted, or completely disappeared from the building's land parcel. Further, the information can be used to determine what the building footprint should be of a building that may be obscured by debris from an accident or disaster.
Although the foregoing has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be obvious to those skilled in the art that certain changes and modifications may be practiced without departing from the spirit and scope thereof, as described in this specification and as defined in the appended claims below.
The foregoing description provides illustration and description, but is not intended to be exhaustive or to limit the inventive concepts to the precise form disclosed. Modifications and variations are possible in light of the above teachings or may be acquired from practice of the methodologies set forth in the present disclosure.
Even though particular combinations of features are recited in the claims and/or disclosed in the specification, these combinations are not intended to limit the disclosure. In fact, many of these features may be combined in ways not specifically recited in the claims and/or disclosed in the specification. Although each dependent claim listed below may directly depend on only one other claim, the disclosure includes each dependent claim in combination with every other claim in the claim set.
No element, act, or instruction used in the present application should be construed as critical or essential to the invention unless explicitly described as such outside of the preferred embodiment. Further, the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise.
This application is a continuation of U.S. patent application Ser. No. 13/689,006, filed on Nov. 29, 2012, entitled “System for Automatic Structure Footprint Detection from Oblique Imagery”, which claims benefit of U.S. Provisional Application No. 61/564,699, filed on Nov. 29, 2011, all of which are incorporated herein by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61564699 | Nov 2011 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 13689006 | Nov 2012 | US |
| Child | 17831656 | US |
