JP2020017048A - Information processing apparatus, method for controlling the same, and program - Google Patents

Abstract

To easily acquire positional information on the earth on an object to be detected included in an aerial image.SOLUTION: An object detection device 100 comprises: an object detection unit 104 that detects an object to be detected included in an aerial image of an image size including a predetermined number of map tiles dividing and showing an image of a map, and outputs in-image coordinates indicating the position of the object to be detected in the aerial image; and a latitude and longitude conversion unit 105 that converts the in-image coordinates output by the object detection unit 104 into the latitude and longitude indicating the position on the earth of the object to be detected on the basis of tile coordinates specifying the map tiles.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an information processing device, a control method thereof, and a program.

  Conventionally, a technique for detecting a detection target included in an aerial image has been known. For example, Patent Literature 1 below discloses a method of performing image analysis and detecting a road included in an aerial image.

JP 2006-107462 A

  When detecting the detection target included in the aerial image, even if the analysis result reveals the position information of the detection target in the aerial image, it is difficult to grasp the position of the detection target on the earth. In particular, when comparing the positions of the detection targets detected from a plurality of aerial images with each other, even if the position information itself in each aerial image is compared, the mutual positional relationship is not known.

  Therefore, an object of the present invention is to provide an information processing apparatus capable of easily acquiring position information on the earth of a detection target included in an aerial image, a control method thereof, and a program.

  The information processing apparatus according to the present invention detects a detection target included in an aerial image of an image size including a predetermined number of map tiles that divide a map image and indicates a position of the detection target in the aerial image. Object detecting means for outputting coordinates in the image, and converting the coordinates in the image output by the object detecting means into latitude and longitude indicating the position on the earth of the detected object based on the tile coordinates for specifying the map tile. And longitude and latitude conversion means.

  The control method of the information processing apparatus according to the present invention detects a detection target included in an aerial image of an image size including a predetermined number of map tiles that divide a map image and displays the map tile. An object detection step of outputting coordinates in the image indicating the position, and the coordinates in the image output in the object detection step, based on the tile coordinates specifying the map tile, the latitude and the position indicating the position of the detection object on the earth. And a latitude-longitude conversion step of converting into longitude.

  A program according to the present invention detects an object to be detected in an aerial image of an image size including an image size including a predetermined number of map tiles that divide a map image and displays the information processing device. An object detecting means for outputting coordinates in the image indicating the position, and the coordinates in the image output by the object detecting means, based on the tile coordinates specifying the map tile, the latitude and the earth indicating the position on the earth of the detection object. This is a program for functioning as latitude / longitude conversion means for converting to longitude.

  In the information processing apparatus, the control method, and the program according to the present invention, the coordinates in the image of the detection target are converted into the latitude and longitude indicating the position of the detection target on the earth based on the tile coordinates specifying the map tile. You. This makes it possible to easily acquire the position information of the detection target on the earth.

  ADVANTAGE OF THE INVENTION According to this invention, the position information on the earth of a detection target object can be easily acquired.

It is a block diagram showing the functional composition of the object detecting device concerning one embodiment of the present invention. FIG. 2 is a diagram illustrating a hardware configuration of the object detection device in FIG. 1. FIG. 2 is a block diagram illustrating a detailed functional configuration of an image acquisition unit in FIG. 1. It is a figure showing typically an example of an aerial image acquired by an object detecting device of this embodiment. It is a conceptual diagram for explaining a map tile. It is a conceptual diagram for explaining the tile coordinate which specifies a map tile. It is a conceptual diagram for explaining a panel. It is a conceptual diagram for explaining a detection range. It is a conceptual diagram for explaining the method of acquiring an aerial image by an image acquisition unit. FIG. 3 is a conceptual diagram illustrating a plurality of aerial images acquired by an image acquisition unit. FIG. 4 is a diagram illustrating an example of an aerial image database recorded by an image management unit. FIG. 3 is a conceptual diagram for explaining coordinates of an object to be detected in an image. It is a figure which shows the conversion formula which converts a tile coordinate into a panel coordinate. It is a flow chart which shows an example of the flow of processing by an object detecting device. 15 is a flowchart illustrating details of the process of step S1403 in FIG. 15 is a flowchart illustrating details of the process of step S1406 in FIG. It is a figure which shows the conversion formula from latitude and longitude to world coordinates. It is a figure which shows the conversion formula from world coordinates to pixel coordinates. It is a figure which shows the conversion formula from a pixel coordinate to a tile coordinate. It is a figure which shows the conversion formula from a tile coordinate to a pixel coordinate. It is a figure which shows the conversion formula from a pixel coordinate to a world coordinate. It is a figure which shows the conversion formula from a world coordinate to a latitude and a longitude. It is a figure showing an aerial image acquired without overlapping. It is a figure showing a plurality of aerial images which have an overlap part. It is a conceptual diagram for explaining the conventional method of converting the coordinates in an image into latitude and longitude. It is a flowchart which shows the flow of the process of converting the conventional coordinate in an image into latitude and longitude.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In the description, the same elements or elements having the same functions will be denoted by the same reference symbols, without redundant description.

  The information processing apparatus according to the present embodiment is an object detection apparatus that acquires a plurality of aerial images to be analyzed and detects a detection object included in each acquired aerial image. The detailed configuration of the object detection device will be described later with reference to FIGS. First, an outline of the object detection device will be described.

  FIG. 4 is a diagram schematically illustrating an example of an aerial image acquired by the object detection device according to the present embodiment. As shown in FIG. 4, the aerial image 400 is an image in which a map image is displayed at a predetermined display magnification using an aerial photograph taken by an imaging device or the like. Here, the map is a planar map represented by the Mercator projection.

  The detection target is an object that can be recognized as an object in the aerial image 400 and is a detection target. The detection target is, for example, a car, a road, a building, or the like, and may be any object that can be detected in the aerial image 400 by image analysis or the like.

  The object detection device according to the present embodiment acquires an aerial image 400 of an image size including a predetermined number of map tiles in accordance with the map tiles. The image size is set in advance by, for example, a system administrator or a user. In the present embodiment, the image size is set to 2048 pixels × 2048 pixels. This image size is, for example, the upper limit of the image size of an aerial image that can be acquired by a Web service or the like or the image size that can be analyzed at high speed. Hereinafter, the upper limit of the image size is also referred to as “upper limit image size”.

  Here, the concept of the map tile will be described. The map tile is a square tile that divides and displays an image of a map of the entire earth. FIG. 5 is a conceptual diagram for explaining a map tile. As shown in FIG. 5, the map tile 501 is a square image forming an entire map image 500 (hereinafter, simply referred to as “entire map image”) showing an image of a map of the entire earth. The length of one side of the map tile 501 is, for example, 256 pixels.

  The number of map tiles 501 constituting the whole map image 500 changes according to the zoom level for specifying the display magnification for displaying the whole map image 500. FIG. 5A shows a case where the zoom level is 0, and FIG. 5B shows a case where the zoom level is 1.

  As shown in FIG. 5A, when the zoom level is 0, the entire map image 500 is composed of one map tile 501. That is, when the zoom level is 0, a map is represented such that the entire world can be accommodated in one map tile 501.

The coordinates indicating the position on the whole map image 500 when the zoom level is 0 are called world coordinates.
The world coordinates are represented by real numbers from 0 to 255. The world coordinates at the upper left corner 500a of the whole map image 500 are shown as (0,0), and the world coordinates at the lower right corner 500b of the whole map image 500 are shown as (255,255).

  World coordinates correspond to latitude and longitude indicating the actual location on the earth. The world coordinates (0, 0) at the upper left corner 500a correspond to the actual latitude and longitude (85.05, -180) at the position of the upper left corner 500a, and the world coordinates (255, 255) at the lower right corner 500b are: This corresponds to the actual latitude and longitude (−85.05, 180) at the position of the lower right corner 500b.

  The world coordinates and the latitude and longitude correspond one-to-one, and can be mutually converted. FIG. 17 is a diagram showing a conversion formula from latitude and longitude to world coordinates. The latitude and longitude can be converted into world coordinates by the conversion formula shown in FIG. FIG. 22 is a diagram showing a conversion formula from world coordinates to latitude and longitude. The world coordinates can be converted into latitude and longitude by the conversion formula shown in FIG. The units of latitude and longitude in the conversion formulas of FIGS. 17 and 22 are radians.

When the zoom level becomes larger than 0, the number of map tiles 501 constituting the entire map image 500 increases. For example, as shown in FIG. 5B, when the zoom level is 1, the whole map image 500 is composed of 2 × 2 map tiles 501. When the zoom level is n, the number of map tiles 501 constituting the entire map image 500 is ²ⁿ × ²ⁿ .

Pixel coordinates (hereinafter, also simply referred to as “pixel coordinates”) indicating positions on the whole map image 500 take integer values, and the maximum value changes depending on the zoom level. When the zoom level is n, the pixel coordinates are shown as a value obtained by multiplying the world coordinates by ²ⁿ , and the world coordinates and the pixel coordinates can be converted to each other.

  FIG. 18 is a diagram showing a conversion formula from world coordinates to pixel coordinates. The world coordinates can be converted to pixel coordinates by the conversion formula shown in FIG. Note that the floor function in the conversion formula in FIG. 18 is a function that rounds down decimal places. FIG. 21 is a diagram showing a conversion formula from pixel coordinates to world coordinates. The pixel coordinates can be converted to world coordinates by the conversion formula shown in FIG.

  FIG. 6 is a conceptual diagram for explaining tile coordinates for specifying the map tile 501. FIG. 6 shows an entire map image 500 composed of 4 × 4 map tiles 501 when the zoom level is 2. In each map tile 501, tile coordinates TP (0,0), (1,0), (2,0), (3,0), (0,1), (0) (1,1), (2,1), (3,1), (0,2), (1,2), (2,2), (3,2), (0,3), (1,2) 3), (2, 3) and (3, 3). By specifying the tile coordinates TP, the map tile 501 is uniquely specified.

  The tile coordinates TP and the pixel coordinates are mutually convertible. FIG. 19 is a diagram showing a conversion formula from pixel coordinates to tile coordinates. The pixel coordinates can be converted to the tile coordinates TP by the conversion formula shown in FIG. Note that the floor function in the conversion formula of FIG. 19 is a function that rounds down decimal places. FIG. 20 is a diagram showing a conversion formula from the tile coordinates TP to the pixel coordinates. The tile coordinates TP can be converted to pixel coordinates by the conversion formula shown in FIG.

(Overall configuration of object detection device)
Next, with reference to FIG. 1, the overall configuration of the target object detection device according to the present embodiment will be described in detail. FIG. 1 is a block diagram illustrating a functional configuration of the object detection device according to the present embodiment.

  As shown in FIG. 1, the object detection device 100 functionally includes an input receiving unit 101 (input receiving unit), a panel managing unit 102 (panel managing unit), and an image obtaining unit 103 (image obtaining unit). ), An image management unit 104, an object detection unit 105 (object detection unit), a latitude / longitude conversion unit 106 (latitude / longitude conversion unit), an overlap determination unit 107 (overlap determination unit), and a detection target identification. Unit 108 (detection target object specifying means).

  The input receiving unit 101 receives inputs of the zoom level and the detection range information and outputs the information to the image obtaining unit 103. The zoom level is information indicating the display magnification of the acquired aerial image 400. The zoom level can take an integer value from 0 to 20. The input receiving unit 101 receives, for example, a numerical value “20” as a zoom level according to a user input.

  The detection range information is information indicating a detection range on a map for detecting a detection target. FIG. 8 is a conceptual diagram for explaining a detection range. As shown in FIG. 8, the detection range is, for example, a rectangular detection range 800.

  The detection range information is the latitude and longitude of the upper left corner 800a of the detection range 800 and the latitude and longitude of the lower right corner 800b of the detection range 800. The input receiving unit 101 receives the latitude and the longitude of the upper left corner 800a of the detection range 800 and the latitude and the longitude of the lower right corner 800b of the detection range 800 by a user input.

  The input receiving unit 101 may output the received zoom level to the panel management unit 102.

  The panel management unit 102 manages a panel serving as a reference for acquiring the aerial image 400 by the image acquisition unit 103. When the aerial image 400 has an image size including a predetermined number of map tiles 501, the panel is a section that divides the map tiles 501 into groups each corresponding to the predetermined number.

  In the present embodiment, the image size of the aerial image 400 is the upper limit image size of 2048 pixels × 2048 pixels and includes 8 × 8 map tiles 501. The panel management unit 102 manages sections that collectively divide the map tiles 501 by the number corresponding to the number as a panel. More specifically, the panel management unit 102 manages, as a panel, a section that divides the map tiles 501 into groups at least for every number of sheets equal to or less than this number (that is, a section having a size equal to or less than the upper limit image size).

  The panels managed by the panel management unit 102 are preset by, for example, a system administrator or a user, depending on how the image acquisition unit 103 wants to acquire the plurality of aerial images 400. For example, when an aerial image 400 having an image size including a predetermined number of map tiles 501 is to be acquired adjacent to each other, a section for grouping the map tiles 501 by a predetermined number is set as a panel. You. In other words, a section having the same size as the upper limit image size is set as a panel.

  Also, for example, when it is desired to obtain an aerial image 400 having an image size including a predetermined number of map tiles 501 so as to have an overlapping portion, the map tiles 501 are collectively divided into smaller numbers than the predetermined number. The section to be set is set as a panel. In other words, a section smaller than the upper limit image size is set as a panel.

  The number of map tiles 501 that are collectively divided into panels may be changed according to the width of the overlapping portion of the acquired aerial image 400. For example, if the width of the overlapping portion is to be the width of one map tile 501, the number of map tiles 501 to be collectively classified as a panel is set to a number smaller by one each in the vertical and horizontal directions than the predetermined number. When the width of the overlapping portion is desired to be larger than the width of one map tile 501, the number of map tiles 501 to be collectively classified as a panel is made smaller than a predetermined number.

  FIG. 7 is a conceptual diagram for explaining the panel according to the present embodiment. In the present embodiment, as illustrated in FIG. 7, the panel management unit 102 manages a section that divides the map tiles 501 into groups of 7 × 7 as a panel 700. As described above, in the present embodiment, when acquiring an aerial image 400 having an image size including 8 × 8 map tiles 501, the number of map tiles 501 to be collectively classified as the panel 700 is set to be smaller than 8 × 8. Are also reduced one by one vertically and horizontally. That is, the panel management unit 102 manages a section smaller than the upper limit image size by a width of one map tile 501 as the panel 700.

  In FIG. 7, the panel coordinates PP are shown on a map tile 501 (a portion shaded in gray) located at the upper left corner in each panel 700. The panel coordinates PP are coordinates that specify each panel 700. By specifying the panel coordinates PP, the panel 700 is uniquely specified.

  The panel coordinates PP are also tile coordinates TP that specify the map tiles 501 (portions shown in gray) located at the upper left corner among the map tiles 501 included in each panel 700. The panel coordinates PP are tile coordinates TP that are divisible by the number of map tiles 501 included in the panel 700 (7 in the present embodiment).

  The correspondence between the panel coordinates PP and the tile coordinates TP is associated by the conversion formula shown in FIG. The panel 700 is associated with the map tile 501 by this conversion formula. In the conversion formula of FIG. 13, n is the number of map tiles 501 included in the panel 700 (the number of map tiles 501 that are collectively divided by one panel 700), and is 7 in the present embodiment. The panel management unit 102 stores the conversion formula of FIG.

  The image acquisition unit 103 acquires an aerial image 400 of a predetermined area based on the information received by the input reception unit 101 and the information managed by the panel management unit 102. As described above, the image size of the aerial image 400 acquired by the image acquiring unit 103 is preset to, for example, 2048 pixels × 2048 pixels.

  The image acquisition unit 103 acquires the aerial image 400 from an internal or external storage device in which the aerial image 400 is stored. The image acquiring unit 103 acquires the aerial image 400 using, for example, a predetermined Web service that can distribute map information. The image acquisition unit 103 outputs the acquired aerial image 400 to the image management unit 104.

  FIG. 9 is a conceptual diagram for describing a method of acquiring an aerial image 400 by the image acquiring unit 103. As illustrated in FIG. 9, the image acquisition unit 103 acquires an aerial image 400 having an image size including a predetermined number of map tiles 501 in accordance with the map tiles 501.

  Specifically, the aerial image 400 is acquired such that the outer edge of the outermost map tile 501 included in the aerial image 400 and the outer edge of the aerial image 400 overlap. Then, the image acquiring unit 103 acquires the aerial image 400 such that each of the predetermined number of map tiles 501 is entirely included in the aerial image 400 without missing any part. That is, an aerial image 400 including a predetermined number of map tiles 501 is acquired.

  The predetermined number is determined to be a number that matches the image size of the acquired aerial image 400. In the present embodiment, the image size of the aerial image 400 to be acquired is 2048 pixels × 2048 pixels and includes 8 × 8 map tiles 501. That is, the image obtaining unit 103 obtains an aerial image 400 having an image size including 8 × 8 map tiles 501.

  The image acquisition unit 103 acquires a plurality of aerial images 400 corresponding to the panel 700 managed by the panel management unit 102. In the present embodiment, an aerial image 400 having an image size (the size indicated by a rectangle 702 in FIG. 7) including 8 × 8 map tiles 501 is associated with a panel 700 including 7 × 7 map tiles 501. get.

  FIG. 10 is a conceptual diagram showing a plurality of aerial images 400 acquired by the image acquisition unit 103. As shown in FIG. 10, the image acquisition unit 103 acquires a plurality of aerial images 400 having overlapping portions 1001 with each other. By acquiring an aerial image 400 having an image size including 8 × 8 map tiles 501 in correspondence with the panel 700 including 7 × 7 map tiles 501, the width of the overlapping portion 1001 is reduced to one map. The width is equivalent to the tile 501. The image acquisition unit 103 outputs the acquired aerial image 400 to the image management unit 104.

  The image management unit 104 saves the aerial image 400 acquired by the image acquisition unit 103 and records it as a database as shown in FIG. FIG. 11 is a diagram illustrating an example of a database of the aerial image 400 recorded by the image management unit 104. The database of FIG. 11 stores a panel ID 1101, panel coordinates PP, and an aerial image file name 1102.

  The panel ID 1101 stored in the database of FIG. 11 is an identifier indicating the panel 700 associated with the acquired aerial image 400. Panel coordinates PP are coordinates specifying panel 700. The aerial image file name 1102 is the name of a file indicating the acquired aerial image 400, and is a file name created when the aerial image 400 is stored.

  As shown in FIG. 11, the image management unit 104 records the aerial image 400 acquired by the image acquisition unit 103 in association with the panel coordinates PP. Thus, the aerial image 400 and the panel 700 are managed so as to correspond one-to-one.

  The target object detection unit 105 reads a plurality of aerial images 400 recorded in the image management unit 104, and detects a detection target included in each of the read aerial images 400. An existing technique can be used for the detection method of the detection target, for example, image analysis may be used, or a learned model such as a neural network may be used.

  For each of the plurality of read aerial images 400, the target object detection unit 105 detects, as a detection result of detecting the detection target object, coordinates indicating the position of the detection target object in the aerial image 400 (hereinafter, also referred to as “in-image coordinates”) Output). The coordinates in the image are coordinates indicating the position of the detection target in the aerial image 400. The reference position P2 of the upper left corner (one of the four corners) of the aerial image 400 is set as the origin, and the right side and the lower side are forward directions. And

  FIG. 12 is a conceptual diagram for explaining the coordinates in the image of the detection target. As shown in FIG. 12, the coordinates in the image are rectangular information (vector V0, vector V1) indicating the area of the detection target, and the positions P0, P0 of the rectangular object area 1201 that is the area of the detection target. This is a vector indicating P1 and having the reference position P2 as the origin.

  The reference position P2 is the position of the upper left corner of the aerial image 400 (the position indicated by the panel coordinates PP associated with the aerial image 400), and is a coordinate reference point indicating the origin of the position on the aerial image 400. The position P0 is the position of the upper left corner of the object area 1201, and the position P1 is the position of the lower right corner of the object area 1201.

  The object detection unit 105 calculates rectangular information (vectors V0 and V1) indicating a region of the detection target as coordinates in the image of the detection target. The object detection unit 105 outputs the rectangular information (vectors V0 and V1) to the latitude / longitude conversion unit 106.

  The latitude / longitude conversion unit 106 converts the rectangular information (vectors V0, V1) calculated by the object detection unit 105 into latitude / longitude indicating the position of each detection object on the earth. The latitude / longitude conversion unit 106 converts the rectangular information (vectors V0, V1) based on the tile coordinates TP (in this embodiment, the tile coordinates TP corresponding to the reference position P2) specifying the map tile 501 included in the aerial image 400. Convert to latitude and longitude.

  A specific method for converting the rectangular information (vectors V0 and V1) into latitude and longitude will be described later with reference to the flowchart in FIG. The latitude / longitude conversion unit 106 converts the rectangular information (vectors V0, V1) into latitude and longitude for all the detection targets, and outputs the conversion result to the overlap determination unit 107.

  The overlap determination unit 107 determines whether or not the same detection target has been repeatedly detected based on the conversion result output from the latitude / longitude conversion unit 106. This is because the aerial image 400 is acquired so as to have the overlapping portion 1001 (see FIG. 10) by the image acquiring unit 103 as described above. Because there is. The duplication determination unit 107 determines the presence or absence of such a detection target duplication detection, and outputs the determination result to the detection target specifying unit 108.

  For example, the duplication determination unit 107 determines the presence or absence of duplication detection based on the degree of overlap of rectangles indicated by latitude and longitude output from the latitude and longitude conversion unit 106. Here, the rectangle indicated by the latitude and longitude indicates the first point indicated by the latitude and longitude as a result of converting the vector V0 as the rectangle information, and the latitude and longitude as the result of converting the vector V1 as the rectangle information. It is a rectangle in which the second point is located diagonally.

  The overlap determination unit 107 determines that the overlap has been detected when the degree of overlap of the rectangles is equal to or greater than a predetermined threshold, and determines that no overlap has been detected when the degree of overlap of the rectangles is less than the predetermined threshold. judge.

  The degree of overlap of the rectangles can be determined using, for example, an index IoU (Intersection Over Union) indicating the ratio of overlapping areas. When IоU is equal to or greater than a predetermined threshold, the degree of overlap of rectangles is equal to or greater than a predetermined threshold, and when IоU is less than the predetermined threshold, the degree of overlap of rectangles is less than a predetermined value.

  Further, the duplication determination unit 107 may determine the presence or absence of duplication detection based on, for example, the degree of coincidence of the latitude and longitude output from the latitude / longitude conversion unit 106, without being limited to the degree of rectangular overlap. The overlap determination unit 107 determines that the overlap has been detected when the degree of coincidence between the latitude and the longitude is equal to or greater than a predetermined threshold, and when the degree of coincidence between the latitude and the longitude is less than the predetermined threshold. It is determined that there was no detection.

  The degree of coincidence between the latitude and the longitude is calculated, for example, by calculating the difference between the latitude and the longitude and based on the magnitude of the difference. If the difference is equal to or smaller than the predetermined threshold, the degree of matching is equal to or greater than the predetermined threshold. If the difference is larger than the predetermined threshold, the degree of matching is set to be lower than the predetermined threshold.

  Based on the determination result from the overlap determination unit 107, the detection target specifying unit 108 detects any one of the detection targets that have been redundantly detected when the detection target is detected in duplicate. An object is specified as a detection target object to be detected. As a result, duplicate detection of the detection target is removed, and the detection target is appropriately detected. The coordinates in the image of the detection target specified by the detection target specifying unit 108 are registered in, for example, the external memory 211 or the like.

  Next, an example of a hardware configuration of the object detection device 100 will be described with reference to FIG. FIG. 2 is a diagram illustrating a hardware configuration of the object detection device 100.

  2, the object detection device 100 includes a CPU 201, a RAM 202, a ROM 203, a system bus 204, an input controller 205, a video controller 206, a memory controller 207, a communication I / F controller 208, and a keyboard 209. , A CRT 210, and an external memory 211.

  The CPU 201 comprehensively controls operations in various devices, and comprehensively controls various devices and various controllers connected to the system bus 204 as necessary.

  The RAM 202 functions as a main memory or a work area of the CPU 201. The CPU 201 loads various programs and the like necessary for executing processing from the ROM 203 or the external memory 211 to the RAM 202, and implements various operations by executing the loaded programs.

  The ROM 203 stores various programs. The various programs include a basic input / output system (BIOS) that is a control program of the CPU 201, an operating system program (hereinafter, referred to as an OS), or various programs necessary for realizing a function executed by each server or each PC. And so on. Note that these programs may be stored in the external memory 211.

  The system bus 204 connects the CPU 201, the RAM 202, the ROM 203, the input controller 205, the video controller 206, the memory controller 207, and the communication I / F controller 208 so that they can communicate with each other.

  The input controller 205 controls input from a keyboard 209 or a pointing device such as a mouse.

  The video controller 206 controls display on a display unit such as a CRT display (CRT) 210. The display unit is not limited to a CRT display, but may be another display unit such as a liquid crystal display.

  The memory controller 207 controls access to the external memory 211. The external memory 211 is a hard disk (HD), a flexible disk (FD), a compact flash (registered trademark) memory, or the like. The external memory 211 stores a boot program, browser software, various applications, font data, user files, edit files, various data, and the like. The CompactFlash (registered trademark) memory is connected to, for example, a PCMCIA card slot via an adapter.

  The communication I / F controller 208 connects and communicates with an external device via a network, and executes communication control processing on the network. For example, Internet communication using TCP / IP is possible.

  Note that the CPU 201 enables the display on the CRT 210 by executing, for example, an outline font development (rasterization) process on a display information area in the RAM 202. Further, the CPU 201 enables a user instruction with a mouse cursor or the like on the CRT 210.

  Various programs for implementing the present invention are recorded in the external memory 211 and are executed by the CPU 201 by being loaded into the RAM 202 as needed. Further, a definition file and various information tables used when executing the program are also stored in the external memory 211.

(Image acquisition method)
Next, a detailed functional configuration of the image acquisition unit 103 will be described with reference to FIG.

  FIG. 3 is a block diagram illustrating a detailed functional configuration of the image acquisition unit 103. As shown in FIG. 3, the image obtaining unit 103 includes a panel specifying unit 301 (panel specifying unit) and a panel corresponding image obtaining unit 302 (panel corresponding image obtaining unit).

  The panel specifying unit 301 specifies a plurality of panels 700 corresponding to the detection range 800 based on the zoom level and the detection range information received by the input reception unit 101. As a result, the panel specifying unit 301 specifies a plurality of panels 700 used to acquire the aerial image 400 to be analyzed among the panels 700 managed by the panel management unit 102.

  For example, the panel specifying unit 301 calculates the panel coordinates PP of the plurality of panels 700 including the detection range 800 by a calculation method described in step S1501 of FIG. In the example shown in FIG. 8, the panel coordinates PP are (0, 0), (7, 0), (14, 0), (0, 7), (7, 7), (14, 7), (0, 14), (7, 14), and (14, 14) are calculated. Thus, a total of nine panels 700 are specified. The panel specifying unit 301 outputs the calculated panel coordinates PP to the panel corresponding image obtaining unit 302.

  The panel corresponding image obtaining unit 302 obtains the aerial image 400 in association with each panel 700 specified by the panel specifying unit 301. The panel-corresponding image acquisition unit 302 specifies, for example, a predetermined parameter (latitude and longitude of the center of the image, a zoom level, and an image size) for each panel 700 using a predetermined Web service capable of distributing map information. .

  Here, the latitude and longitude of the center of the image as the predetermined parameters are calculated based on the panel coordinates PP calculated by the panel specifying unit 301 by the calculation method of step S1502 in FIG. The panel corresponding image acquisition unit 302 acquires the aerial image 400 using the latitude and longitude of the image center calculated based on the panel coordinates PP as parameters. Thereby, the aerial image 400 is acquired in one-to-one correspondence with the panel 700, and the corresponding aerial image 400 is acquired for all the specified panels 700. As a result, a plurality of aerial images 400 having an overlapping portion 1001 having a width of one map tile 501 are obtained.

  Next, an example of an overall processing flow for detecting a detection target by the target detection device 100 will be described with reference to the flowchart of FIG. 14 and the like. FIG. 14 is a flowchart illustrating an example of the flow of a process performed by the target object detection device 100.

  As shown in FIG. 14, when the object detection processing starts, first, in step S1401, it is determined whether to continue the processing. Whether or not to continue the process is determined, for example, by whether or not a user has input a process end instruction. If it is determined that the process is not to be continued (step S1401; NO), the process ends. If it is determined that the processing is to be continued (step S1401; YES), the processing of steps S1402 to S1408 is performed.

  In step S1402, the input receiving unit 101 receives a zoom level and detection range information. The input receiving unit 101 receives a numerical value of, for example, “20” as a zoom level according to a user input. In addition, the input receiving unit 101 receives, as detection range information, the latitude and longitude of the upper left corner 800a of the detection range 800 and the latitude and longitude of the lower right corner 800b of the detection range 800 as input by the user.

  Next, the image acquisition unit 103 acquires the aerial image 400 based on the information received in step S1401 (step S1403). Although details will be described later with reference to the flowchart of FIG. 15, the image acquisition unit 103 acquires a plurality of aerial images 400 having an overlapping portion 1001 of one map tile 501.

  Subsequently, the image management unit 104 stores the plurality of aerial images 400 acquired in step S1403 on a hard disk or the like of the object detection device 100, and records a database of the aerial images 400 as illustrated in FIG. S1404). That is, the image management unit 104 records each aerial image 400 in association with the panel coordinates PP.

  Next, the target object detection unit 105 reads a plurality of aerial images 400 from the database recorded in step S1404, and detects a detection target included in each aerial image 400 (step S1405). Specifically, the target detection unit 105 outputs rectangular information (vectors V0 and V1) indicating the area of the detection target as a result of detecting the detection target.

  Next, the latitude / longitude conversion unit 106 converts the rectangular information (vectors V0, V1) detected in step S1405 into latitude and longitude (step S1406). Details of the flow of the latitude / longitude conversion process by the latitude / longitude conversion unit 106 will be described later with reference to the flowchart in FIG.

  Subsequently, the duplication determination unit 107 determines whether or not the detection target has been redundantly detected based on the latitude and longitude converted in step S1406 (step S1407).

  For example, when the degree of overlap of the rectangles indicated by latitude and longitude is less than a predetermined threshold, the overlap determination unit 107 determines that the same detection target has not been detected in duplicate (step S1407; NO). Then, the process returns to step S1401.

  When the degree of overlap of the rectangles represented by latitude and longitude is equal to or greater than a predetermined threshold, the overlap determination unit 107 determines that the same detection target has been repeatedly detected (step S1407; YES), and step S1408. Move to.

  In step S1408, the detection target specifying unit 108 specifies any one of the detection targets determined to have been redundantly detected in step S1407 as the detection target to be detected.

  Specifically, for example, in the overlapping portion 1001 of the aerial images 400 arranged side by side, if the same detection target is detected in duplicate, the detection target included in the right aerial image 400 is invalidated, and the left The detection target object included in the aerial image 400 is valid. In addition, in the overlapping portion 1001 of the vertically arranged aerial images 400, when the same detection target is detected in duplicate, the detection target included in the lower aerial image 400 is invalidated, and the upper aerial image 400 is invalidated. The object to be detected included in is valid.

  Upon completion of the detection of the detection target by the detection target specifying unit 108, the process returns to step S1401. Until it is determined in step S1401 that the processing is not to be continued, the object detection processing in steps S1402 to 1408 is repeated.

  Next, with reference to FIG. 15, the details of the processing flow of the image acquisition method by the image acquisition unit 103 will be described. FIG. 15 is a flowchart illustrating details of the process in step S1403 in FIG.

  As shown in FIG. 15, when the image acquisition process by the image acquisition unit 103 starts, first, the panel identification unit 301 of the image acquisition unit 103 identifies a plurality of panels 700 including the detection range 800 (step S1501). .

  Specifically, the panel specifying unit 301 converts the detection range information (latitude and longitude of the upper left corner 800a of the detection range 800 and latitude and longitude of the lower right corner 800b of the detection range 800) received by the input reception unit 101 into: By converting to the panel coordinates PP, the panel 700 specified by the panel coordinates PP is specified.

  The conversion from the latitude and longitude to the panel coordinates PP is performed as follows using the conversion formulas of FIGS. 17 to 19 and the conversion formula of FIG. These conversion equations are stored in, for example, the external memory 211 of the object detection device 100 or the like.

  First, the panel specifying unit 301 converts the latitude and longitude of each of the upper left corner 800a and the lower right corner 800b into world coordinates using the conversion formula in FIG.

  Subsequently, the panel specifying unit 301 converts the converted world coordinates into pixel coordinates using the conversion formula in FIG. Here, the zoom level z in the conversion formula in FIG. 18 is the zoom level received in step S1402.

  Subsequently, the panel specifying unit 301 converts the converted pixel coordinates into the tile coordinates TP using the conversion formula in FIG.

  Subsequently, the panel specifying unit 301 converts the tile coordinates TP into panel coordinates PP using the conversion formula in FIG.

  By the above conversion, for example, in the example of FIG. 8, the latitude and longitude of the upper left corner 800a of the detection range 800 are converted into the panel coordinates PP (0,0). The panel coordinates PP (0,0) are the minimum values of the panel coordinates PP that specify the panel 700 including the detection range 800. In addition, the latitude and longitude of the lower right corner 800b of the detection range 800 are converted to panel coordinates PP (14, 14). The panel coordinates PP (14, 14) are the maximum values of the panel coordinates PP that specify the panel 700 including the detection range 800.

  The panel specifying unit 301 calculates all panel coordinates PP in a range equal to or larger than the minimum value panel coordinate PP and equal to or less than the maximum value panel coordinate PP. In the example of FIG. 8, the panel coordinates PP are (0, 0), (7, 0), (14, 0), (0, 7), (7, 7), (14, 7), (0 , 14), (7, 14), and (14, 14). Thus, a total of nine panels 700 are specified.

  When the panel coordinates PP are calculated in step S1501, the process proceeds to step S1502. In step S1502, the panel corresponding image acquisition unit 302 of the image acquisition unit 103 calculates the latitude and longitude of the image center (parameters for acquiring the aerial image 400) based on the panel coordinates PP calculated in step S1501 ( Step S1502).

  For example, as shown in FIG. 9, the panel-corresponding image acquisition unit 302 determines the tile coordinates TP (11) located at the lower right of four map tiles 501 from the panel coordinates PP (7,7) calculated in step S1501. , 11) is an image center position 901 indicating the center of the image. That is, a position indicated by coordinates obtained by adding the number of map tiles 501 constituting the aerial image 400 to １ ／ (here, 4) to the panel coordinates PP is defined as an image center position 901.

  The panel-corresponding image acquisition unit 302 converts the tile coordinates TP (11, 11) indicating the image center position 901 into latitude and longitude using the conversion formulas shown in FIGS. The result of this conversion is defined as the latitude and longitude of the center of the image.

  Subsequently, the panel corresponding image acquisition unit 302 acquires a plurality of aerial images 400 corresponding to each of the plurality of panels 700 specified in step S1501 (step S1503).

  Specifically, by specifying a predetermined parameter in a predetermined Web service capable of distributing map information, the aerial image 400 is acquired one-to-one with the panel 700. Here, the specified parameters are the latitude and longitude of the image center calculated in step S1502, the zoom level received in step S1402, and a preset image size (2048 pixels × 2048 pixels).

  Through the above processing, a plurality of aerial images 400 are acquired in association with the panel 700, and the image acquisition processing by the image acquisition unit 103 ends.

  Next, with reference to FIG. 16, details of the processing flow of the latitude / longitude conversion method by the latitude / longitude conversion unit 106 will be described. FIG. 16 is a flowchart showing details of the processing in step S1406 in FIG.

  As shown in FIG. 16, when the processing by the latitude / longitude conversion unit 106 starts, first, the latitude / longitude conversion unit 106 calculates the pixel coordinates of the reference position P2 (see FIG. 12) of the aerial image 400 (step S1601). .

  Specifically, by converting the panel coordinates PP (that is, the tile coordinates TP divisible by 7) recorded as a database as shown in FIG. 11 in step S1404 using the conversion formula in FIG. 20, the pixel at the reference position P2 is converted. Calculate the coordinates. That is, based on the panel coordinates PP of the panel 700 (the tile coordinates TP for specifying the map tile 501 located at the upper left corner among the map tiles 501 included in the aerial image 400) corresponding to the time of acquiring the aerial image 400, Performs conversion processing.

  Subsequently, the latitude / longitude conversion unit 106 calculates the pixel coordinates of the positions P0 and P1 (see FIG. 12) indicating the object area 1201 (step S1602).

  Specifically, based on the pixel coordinates of the reference position P2 calculated in step S1601 and the rectangular information (vectors V0 and V1) calculated in step S1405, the pixel coordinates of the positions P0 and P1 indicating the object region 1201 are calculated. calculate.

  The pixel coordinates of the upper left position P0 are calculated by adding the vector V0 from the reference position P2 to the position P0 to the pixel coordinates of the reference position P2. The pixel coordinates of the lower right position P1 are calculated by adding the vector V1 from the reference position P2 to the position P1 to the pixel coordinates of the reference position P2.

  Subsequently, the latitude / longitude conversion unit 106 converts the pixel coordinates calculated in step S1602 into world coordinates using the conversion formula in FIG. 21 (step S1603). Subsequently, the latitude / longitude conversion unit 106 converts the world coordinates calculated in step S1603 into latitude and longitude using the conversion formula in FIG. 22 (step S1604).

  With the above processing, the processing of conversion into latitude and longitude by the latitude / longitude conversion unit 106 ends.

  The various processes described above may be realized as a program for causing the computer of the object detection device 100 to function as each of the above-described functional units.

  Next, the operation and effect of the object detection device 100 according to the present embodiment will be described in comparison with the related art.

In a conventional method for acquiring an aerial image, for example, when a predetermined Web service capable of distributing map information is used, one parameter is specified by specifying predetermined parameters (latitude and longitude of the center of the image, zoom level, and image size). Images had been obtained. In this case, it has been difficult to obtain a plurality of aerial images suitable for detecting the detection target.

  For example, when the user appropriately specifies the latitude and longitude of the center of the image, there is a possibility that a plurality of aerial images are acquired with a gap without being adjacent to each other. In other words, there has been a problem of omission of aerial image acquisition, in which image information has not been acquired for the gap. As a result, the detection target existing in the gap could not be properly detected. In order to acquire a plurality of aerial images that are adjacent without a gap, it is necessary for the user to adjust and set parameters (latitude and longitude of the center of the image) specified for each of the aerial images. Did not.

  Further, even if a plurality of aerial images can be obtained adjacently without a gap, for example, when the detection target is located across two adjacent aerial images, one aerial image is obtained. When the image is viewed as an image, the object to be detected is cut off. As a result, the detection target could not be properly detected.

  In order to appropriately detect the detection target, it is only necessary to acquire one large aerial image in which the detection target is included in one aerial image. There is an upper limit to the image size that can be analyzed by. For this reason, it is required to appropriately acquire a plurality of aerial images having an image size equal to or smaller than the upper limit such that omission of detection of an object to be detected can be suppressed.

  On the other hand, according to the target object detection device, the control method, and the program according to the present embodiment, when the image size of the aerial image 400 is a size including the predetermined number of map tiles 501, the predetermined number is set. Sections that collectively divide the map tiles 501 for each corresponding number are managed as a panel 700. Then, each aerial image 400 is obtained in association with the panel 700. Therefore, for example, the aerial images 400 adjacent to each other so as to have a plurality of aerial images 400 so as to be arranged without a gap and the overlapping portion 1001 can be easily acquired by associating them with the panel 700. As a result, it is possible to suppress detection omission of the detection target object included in the aerial image. As described above, it is possible to acquire a plurality of aerial images 400 that are appropriate for suppressing omission of detection of a detection target.

  Here, FIG. 23 is a diagram showing an aerial image 400 acquired without overlapping. As shown in FIG. 23, when the aerial image 400 is acquired without overlapping, for example, the detection target M (a diamond-shaped object) is located just over two adjacent aerial images 400. However, the detection target object M can be missed from each aerial image 400. For this reason, there was a possibility that the detection target M could not be properly detected.

  On the other hand, in the present embodiment, when the aerial image 400 has a size including 8 × 8 map tiles 501, the aerial image 400 corresponds to 7 × 7 images smaller than 8 × 8 images. Is obtained. As a result, a plurality of aerial images 400 having the overlapping portion 1001 can be easily acquired.

  FIG. 24 is a diagram showing a plurality of aerial images 400 having an overlapping portion 1001. As shown in FIG. 24, by acquiring the aerial image 400 so as to have the overlapping portion 1001, the detection target M positioned between two adjacent aerial images 400 in FIG. Are located in the overlapping portion 1001. Therefore, the detection target M is included in each aerial image 400, and the detection omission of the detection target M can be suppressed.

  According to the present embodiment, when the user inputs the detection range information and the zoom level to the input reception unit 101, a plurality of panels 700 corresponding to each aerial image 400 are specified based on the detection range information. . Then, a plurality of aerial images 400 are acquired corresponding to the panel 700. Therefore, unlike the related art, there is no need to input predetermined parameters such as the latitude and longitude of the center of the image each time the image is acquired, and it is not necessary for the user to adjust the parameters for appropriate image acquisition. According to the present embodiment, a plurality of panels 700 necessary for acquiring each aerial image 400 are specified only by inputting the detection range information and the zoom level, and the omission of detection of the detection target is detected in correspondence with the panels 700. A plurality of aerial images 400 suitable for suppression can be acquired at a time.

  According to the present embodiment, the identity of the detection target detected in the overlapping portion 1001 is determined. When it is determined that the same detection target is detected in duplicate, any one of the detected detection targets is identified as a detection target to be detected, and the detection target is detected. Can be eliminated. Therefore, both the suppression of detection omission and the suppression of duplicate detection can be achieved, and the detection target can be appropriately detected.

  Here, the rectangle information (vector V0, V1) indicating the area of the detection target that is output as a result of detection of the detection target is the coordinates in the image indicating the position in each aerial image 400, and therefore the rectangular information Even if the (vectors V0, V1) themselves are compared, it is difficult to compare the position of the detection target. On the other hand, according to the present embodiment, the rectangular information (vectors V0 and V1) indicating the area of each detection target is converted into latitude and longitude indicating the position on the earth of each detection target. The position of each detection object can be compared based on the latitude and longitude.

  Further, in the present embodiment, the identity of the detection target detected in the overlapping portion 1001 is determined based on the result of converting the rectangular information (vectors V0 and V1) indicating the area of each detection target into latitude and longitude. I do. Therefore, it is possible to more accurately determine the identity of the detection target.

  Further, according to the present embodiment, it is possible to convert the coordinates in the image into the latitude and the longitude more easily than in the conventional method of converting the coordinates in the image into the latitude and the longitude.

  FIG. 25 is a conceptual diagram for explaining a conventional method for converting coordinates in an image into latitude and longitude. FIG. 26 is a flowchart showing the flow of a conventional process for converting coordinates in an image into latitude and longitude. As shown in FIGS. 25 and 26, conventionally, coordinates in an image are converted into latitude and longitude based on the latitude and longitude of the image center Pa.

  Specifically, first, the latitude and longitude of the image center Pa are converted into world coordinates by the conversion formula of FIG. 17 (step S2601). Subsequently, the world coordinates are converted into pixel coordinates by the conversion formula in FIG. 18 (step S2602).

  Subsequently, rectangular information (vectors V0 and V1) indicating the area of the detection target is calculated as vectors V2 and V3 from the image center Pa to the detection target (step S2603). Subsequently, based on the pixel coordinates of the image center Pa calculated in steps S2601 and S602 and the vectors V2 and V3 calculated in step S2603, the pixel coordinates of the positions P0 and P1 indicating the object area 1201 are calculated (step S2601). 2604).

  Subsequently, the pixel coordinates calculated in step S2604 are converted into world coordinates using the conversion formula in FIG. 21 (step S2605). Subsequently, the world coordinates calculated in step S2605 are converted into latitude and longitude using the conversion formula in FIG. 22 (step S2606).

  As described above, in the conventional method of converting latitude and longitude, since the conversion process is performed based on the latitude and longitude of the image center Pa, the latitude and longitude of the image center Pa are determined as shown in steps S2601 and S2602. Is first converted to world coordinates and then to pixel coordinates. Further, as shown in steps S2603 and S2604, when calculating the pixel coordinates of the positions P0 and P1 indicating the object region 1201, the vectors V0 and V1 from the reference position P2 to the positions P0 and P1 are calculated based on the image center. It takes time to convert from Pa to vectors V2 and V3 directed to the detection target.

  On the other hand, in the method of converting coordinates in an image into latitude and longitude according to the present embodiment, as shown in the flowchart of FIG. 16, based on panel coordinates PP (tile coordinates TP) of reference position P2 instead of image center Pa. Then, a conversion process is performed. Therefore, it is only necessary to convert the panel coordinates PP (tile coordinates TP) indicating the reference position P2 into pixel coordinates by the conversion formula 20. Further, when calculating the pixel coordinates of the positions P0 and P1 indicating the object area 1201, the vectors V0 and V1 from the reference position P2 to the positions P0 and P1 are used without being converted into the vectors V2 and V3. The coordinates can be calculated. As described above, the conversion process can be facilitated.

  As described above, one embodiment of the present embodiment has been described. However, the present invention is not limited to the above embodiment, and various changes can be made without departing from the gist of the present invention.

  In the above embodiment, the example in which the aerial image 400 acquired by the image acquisition unit 103 is recorded by the image management unit 104 has been described, but the present invention is not limited to this. The image acquisition unit 103 may directly output the acquired aerial image 400 to the object detection unit 105. The target detection unit 105 may detect a detection target in the aerial image 400 output without being recorded by the image management unit 104.

  The image size of the aerial image 400 acquired by the image acquisition unit 103 is not limited to 2048 pixels × 2048 pixels and may be any size including an integer number of map tiles 501. Further, the image size may not be set in advance, and may be received by the input receiving unit 101 by, for example, a user input.

  Further, the number of map tiles 501 that are collectively classified by panel 700 may be set and managed by panel management unit 102 according to the image size and zoom level received by input reception unit 101. In this case, in response to the image size and zoom level being received by the input receiving unit 101, the number of map tiles 501 that one panel 700 classifies can be automatically determined by the panel management unit 102. .

  In the above embodiment, an example has been described in which the latitude and longitude of the image center calculated based on the panel coordinates PP of the panel 700 are designated as parameters to acquire the aerial image 400, but the present invention is not limited to this. Any method may be used as long as it acquires the aerial image 400 corresponding to the panel 700. For example, the aerial image 400 may be acquired by designating the panel coordinates PP itself as a parameter.

  In the above-described embodiment, an example has been described in which the image acquisition unit 103 acquires a plurality of aerial images 400 having the overlapping portion 1001 of one map tile 501, but the invention is not limited thereto. The width of the overlapping portion 1001 may be larger or smaller than the width of one map tile 501. The number of map tiles 501 that panel 700 collectively divides may be appropriately adjusted and set so as to have overlapping portion 1001 of a predetermined width.

  Further, the image acquisition unit 103 may not have the overlapping portion 1001 and may acquire a plurality of aerial images 400 arranged adjacent to each other.

  The correspondence shown by the conversion formulas in FIG. 13 and FIGS. 17 to 22 is not limited to the format of the conversion formula shown, and may be stored as another conversion formula format, a database, or the like.

  The method by which the image acquiring unit 103 acquires the aerial image 400 is not limited to the method using a predetermined Web service capable of distributing map information as described in the above embodiment. The aerial image 400 may be acquired from an internal or external storage device in which the aerial image 400 is stored, or may be acquired by directly capturing the aerial image 400 in correspondence with the panel 700.

  The determination of the presence or absence of duplicate detection by the duplicate determination unit 107 is not limited to the method described in the above embodiment. For example, the shape of the detection target may be detected by the target detection unit 105, and presence or absence of duplicate detection may be determined based on the detected shape. In this case, the object detection device 100 may not necessarily include the latitude / longitude conversion unit 106.

  In the above-described embodiment, an example is described in which a process of converting coordinates in an image into latitude and longitude is performed based on the tile coordinates TP specifying the map tile 501 located at the upper left corner among the map tiles 501 included in the aerial image 400. Although described, it is not limited to this. For example, the conversion processing may be performed based on the tile coordinates TP specifying the map tile 501 located at any of the four corners among the map tiles 501 included in the aerial image 400. Further, the method of converting the coordinates in the image into the latitude and longitude is not limited to the processing shown in the flowchart of FIG. 16, and may be, for example, the processing shown in the flowchart of FIG.

  For example, the present invention can be embodied as a system, an apparatus, a method, a program, a storage medium, or the like. Specifically, the present invention may be applied to a system including a plurality of devices. You may apply to the apparatus which consists of three apparatuses.

  Note that the present invention includes a program that directly or remotely supplies a software program that realizes the functions of the above-described embodiments to a system or an apparatus. The present invention includes a case where the present invention is also achieved by reading and executing a supplied program code by a computer of the system or the apparatus.

  Therefore, the program code itself installed in the computer in order to realize (executable) the functional processing of the present invention by the computer also realizes the present invention. That is, the present invention includes the computer program itself for realizing the functional processing of the present invention.

  In that case, as long as it has the function of the program, it may be in the form of object code, a program executed by the interpreter, or script data supplied to the OS.

  Examples of the recording medium for supplying the program include a flexible disk, a hard disk, an optical disk, a magneto-optical disk, an MO, a CD-ROM, a CD-R, and a CD-RW. Further, there are a magnetic tape, a nonvolatile memory card, a ROM, and a DVD (DVD-ROM, DVD-R).

  As another program supply method, a client computer is connected to a homepage on the Internet using a browser. The computer program of the present invention or the compressed file including the automatic installation function can be supplied from the home page by downloading the file to a recording medium such as a hard disk.

  Also, the present invention can be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. In other words, the present invention also includes a WWW server that allows a plurality of users to download a program file for implementing the functional processing of the present invention on a computer.

  Further, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and downloaded to a user who satisfies predetermined conditions from a homepage via the Internet to download key information for decryption. Let it. It is also possible to execute the encrypted program by using the downloaded key information and install the program on a computer to realize the program.

  The functions of the above-described embodiments are implemented when the computer executes the read program. In addition, the OS or the like running on the computer performs part or all of the actual processing based on the instructions of the program, and the functions of the above-described embodiments can also be realized by the processing.

  Further, the program read from the recording medium is written to a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer. Thereafter, based on instructions of the program, a CPU or the like provided in the function expansion board or the function expansion unit performs part or all of actual processing, and the functions of the above-described embodiments are also realized by the processing.

  It should be noted that the above-described embodiment is merely an example of the embodiment for carrying out the present invention, and the technical scope of the present invention should not be interpreted in a limited manner. That is, the present invention can be implemented in various forms without departing from the technical idea or the main features.

100 Object detection device (information processing device)
101 input receiving unit (input receiving means)
102 Panel management unit (panel management means)
103 Image acquisition unit (image acquisition means)
105 Object detection unit (object detection means)
106 Latitude / Longitude Conversion Unit (Latitude / Longitude Conversion Means)
107 Duplication judgment unit (duplication judgment means)
108 Detection target specifying unit (detection target specifying unit)
301 Panel identification part (panel identification means)
302 Panel-compatible image acquisition unit (panel-compatible image acquisition means)
501 Map tile 700 Panel 1001 Overlapping part

Claims (7)

A target for detecting a detection target included in an aerial image of an image size including a predetermined number of map tiles that are obtained by dividing a map image and outputting coordinates in the image indicating a position of the detection target in the aerial image Object detection means;
Latitude-longitude conversion means for converting the coordinates in the image output by the object detection means into latitude and longitude indicating the position of the detection object on the earth based on the tile coordinates specifying the map tile; An information processing device. The coordinates in the image are coordinates indicating the position of the detection target with one of the four corners of the aerial image as the origin,
The information processing apparatus according to claim 1, wherein the latitude / longitude conversion unit converts the coordinates into the latitude and the longitude based on the tile coordinates corresponding to the origin.   2. The apparatus according to claim 1, further comprising: an overlap determination unit configured to determine whether the same detection target is redundantly detected by the target detection unit based on the latitude and the longitude converted by the latitude and longitude conversion unit. 3. The information processing device according to 2.   When it is determined by the duplication determination means that the detection target has been detected in duplicate, any one of the detection targets detected in duplicate is the detection target to be detected. The information processing apparatus according to claim 3, further comprising: a detection target specifying unit configured to specify as a detection target. Image acquisition means for acquiring a plurality of the aerial images having an image size including a predetermined number of the map tiles in accordance with the map tiles,
Panel management means for managing a section that collectively divides the map tiles into a number corresponding to the predetermined number as a panel,
The image acquisition unit acquires each of the aerial images in association with the panel managed by the panel management unit,
2. The latitude-longitude conversion unit converts the coordinates in the image into the latitude and the longitude based on panel coordinates specifying the panel corresponding to the acquisition of the aerial image by the image acquisition unit. 3. The information processing device according to any one of claims 4 to 4. A target for detecting a detection target included in an aerial image of an image size including a predetermined number of map tiles that are obtained by dividing a map image and outputting coordinates in the image indicating a position of the detection target in the aerial image An object detection step;
A latitude-longitude conversion step of converting the coordinates in the image output in the object detection step into latitude and longitude indicating a position on the earth of the detection object based on tile coordinates specifying the map tile; And a control method of the information processing apparatus. Information processing device,
Detects a detection target included in an aerial image of an image size including a predetermined number of map tiles that divide the image of the map of the entire earth and displays coordinates in the image indicating the position of the detection target in the aerial image. Object detection means for outputting;
The coordinates in the image output by the object detection means are made to function as latitude / longitude conversion means for converting the coordinates of the detection object into the latitude and longitude indicating the position on the earth based on the tile coordinates specifying the map tile. Program for.

Images