JP7801646B2 - Image processing device, image processing method and program - Google Patents

Description

This disclosure relates to an image processing device, an image processing method, and a program.

The ground on which objects such as utility poles are installed usually has multiple land classifications, such as public land and private land. It is important for the administrator who manages the object to understand the land classification where the object is installed.

One method for determining the land classification of the installation location of an object is to determine the land classification of the installation location of an object from a point cloud obtained by a laser scanner (see non-patent document 1).

Minemura et al.: Extraction and classification of objects using point clouds and images obtained by moving measurements (2nd report), Proceedings of the 2021 Autumn Meeting of the Japan Society for Precision Engineering, 2021

The above-mentioned method has the problem that it requires a lot of time and resources, and is labor-intensive, due to labeling each point that makes up the point cloud and generating a three-dimensional model based on the point cloud.

The purpose of this disclosure, made in consideration of the above-mentioned problems, is to provide an image processing device, image processing method, and program that can more easily determine the land classification of the installation location of an object.

To solve the above problem, the image processing device disclosed herein is an image processing device that determines the land classification of the installation location of an object installed on the ground consisting of multiple land classifications from an image captured of the object installed on the ground, and includes: a pixel area detection unit that generates a pixel area detection image from the captured image that shows the pixel area of the object and the pixel area of each of the multiple land classifications; an object area extraction unit that generates an object area extraction image that extracts an image of the area near the bottom end of the object that abuts the ground from the pixel area detection image; and a classification determination unit that determines the land classification of the installation location of the object based on the number of pixels in the pixel area of each of the multiple land classifications in the object area extraction image.

In addition, to solve the above-mentioned problem, the image processing method disclosed herein is an image processing method using an image processing device that determines the land classification of the installation location of an object installed on the ground consisting of multiple land classifications from an image captured of the object installed on the ground, and includes the steps of: generating a pixel area detection image from the captured image that shows the pixel area of the object and the pixel area of the land of each of the multiple land classifications; generating an object area extraction image by extracting an image of the area near the bottom end of the object that abuts the ground in the pixel area detection image; and determining the land classification of the installation location of the object based on the number of pixels in the pixel area of the land of each of the multiple land classifications in the object area extraction image.

In addition, to solve the above problem, the program disclosed herein causes a computer to operate as the above-mentioned image processing device.

The image processing device, image processing method, and program disclosed herein make it easier to determine the land classification of the installation location of an object.

1 is a diagram illustrating an example of the configuration of an image processing device according to a first embodiment of the present disclosure. FIG. 1 is a diagram illustrating an example of a captured image including an object. 2 is a diagram showing an example of a pixel region detection image generated by a pixel region detection unit shown in FIG. 1; FIG. 2 is a diagram for explaining the operation of a judgment target selection unit shown in FIG. 1; FIG. 2 is a diagram for explaining the operation of a judgment target selection unit shown in FIG. 1; FIG. 2 is a diagram for explaining the operation of a judgment target selection unit shown in FIG. 1; FIG. 2 is a diagram for explaining the operation of a judgment target selection unit shown in FIG. 1; FIG. 2 is a diagram for explaining the operation of the object region extraction unit shown in FIG. 1 . FIG. 1. FIG. 4 is a diagram for explaining the operation of the lower end region expansion unit shown in FIG. 2 is a flowchart showing an example of the operation of the image processing device shown in FIG. 1 . FIG. 10 is a diagram illustrating an example of a configuration of an image processing device according to a second embodiment of the present disclosure. 9A and 9B are diagrams for explaining the operation of the lower end area detection unit shown in FIG. 8 . FIG. 10 is a diagram illustrating an example of the configuration of an image processing device according to a third embodiment of the present disclosure. FIG. 1 is a diagram illustrating an example of a hardware configuration of an image processing device according to the present disclosure.

Embodiments of the present disclosure are described below with reference to the drawings.

(First embodiment)
Fig. 1 is a diagram illustrating an example configuration of an image processing device 10 according to a first embodiment of the present disclosure. The image processing device 10 according to the present disclosure determines the land classification of the installation location of an object 2 from a captured image of the object 2 installed on ground 1 consisting of multiple land classifications, as shown in Fig. 2. In the following, an example will be described in which the ground 1 consists of two land classifications, namely, public land 1A and private land 1B, but the present disclosure is not limited to this. The object 2 is installed on the ground 1, such as a utility pole, a guide sign, a traffic sign, a street light, or a traffic light.

As shown in Figure 1, the image processing device 10 of this embodiment includes an input unit 11, a pixel area detection unit 12, a judgment target selection unit 13, an object area extraction unit 14, a lower end area extension unit 15, a classification judgment unit 16, and an output unit 17.

The input unit 11 receives an image captured of a specified range including the object 2, as shown in Figure 2.

From the captured image input to the input unit 11, the pixel area detection unit 12 generates a pixel area detection image showing the pixel area of the object 2, the pixel area of each of the multiple land divisions (in this embodiment, public land 1A and private land 1B), and the pixel area of the background other than the object 2 and the ground 1, as shown in Figure 3. In the example shown in Figure 2, the pixel area detection unit 12 generates a pixel area detection image showing the pixel area of the object 2, the pixel area of the road (public land 1A), the pixel area of the gutter (public land 1A), the pixel area of private land 1B, and the pixel area of the background.

The pixel area detection unit 12 detects the pixel area of the object 2, for example, by image recognition of the captured image. Furthermore, for example, information indicating whether each pixel corresponding to the ground 1 is included in the pixel area of public land 1A or the pixel area of private land 1B is input to the pixel area detection unit 12, and the pixel area of public land 1A and the pixel area of private land 1B are detected based on this information. Furthermore, for example, information indicating the coordinates of the pixel area of public land 1A and the pixel area of private land 1B in the captured image is input to the pixel area detection unit 12, and the pixel area of public land 1A and the pixel area of private land 1B are detected based on this information. Furthermore, for example, the pixel area detection unit 12 detects pixel areas other than the pixel area of the object 2, the pixel area of public land 1A, and the pixel area of private land 1B as background pixel areas.

Referring again to Figure 1, the determination target selection unit 13 selects an object 2 to be used to determine the land classification of the installation location from the pixel area detection image generated by the pixel area detection unit 12. Specifically, the determination target selection unit 13 excludes from the determination targets those objects 2 included in the pixel area detection image that are difficult to determine, and selects the remaining objects 2 as determination targets.

Figure 4A is a diagram showing an example of a pixel region detection image. As shown in Figure 4A, the determination target selection unit 13 sets an X coordinate along one direction of the pixel region detection image (the horizontal direction in Figure 4A) and sets a Y axis in a direction perpendicular to the X coordinate (the vertical direction in Figure 4A). The determination target selection unit 13 determines whether the coordinate of the edge of the pixel region of the object 2 on the ground 1 side is the maximum or minimum coordinate in the X-axis or Y-axis direction of the pixel region detection image. As shown in Figure 4A, if the coordinate of the edge of the pixel region of the object 2 on the ground 1 side is the maximum or minimum coordinate in the X-axis or Y-axis direction of the pixel region detection image, the bottom edge of the object 2 that abuts the ground 1 may not be included in the pixel region detection image. If the bottom edge of the object 2 that abuts the ground 1 is not included in the pixel region detection image, it is difficult to determine the land classification of the installation location of the object 2. Therefore, the determination target selection unit 13 excludes from the determination target any object 2 whose bottom edge that abuts the ground 1 is not included in the pixel region detection image.

Figure 4B is a diagram showing another example of a pixel area detection image. As shown in Figure 4B, the determination target selection unit 13 calculates the size (number of pixels) of the object 2 in a predetermined direction (the direction (X-axis direction) perpendicular to the extension direction (Y-axis direction) of the object 2). An object 2 whose size in the predetermined direction is smaller than the threshold value is considered to be located far away from the shooting location and to appear small. It is difficult to determine the land classification of the installation location of such an object 2. Therefore, the determination target selection unit 13 excludes from the determination target any object 2 whose size in the predetermined direction in the pixel area detection image is equal to or smaller than the threshold value.

Figures 4C and 4D are diagrams showing yet another example of a pixel region detection image. The determination target selection unit 13 determines whether the pixel at the end of the pixel region of the object 2 on the ground 1 side in the pixel region detection image is adjacent to a pixel of the background. As shown in Figure 4C, if the pixel at the end of the pixel region of the object 2 on the ground 1 side in the pixel region detection image is adjacent to a pixel other than the background (government land 1A or private land 1B), the object 2 is considered to be in contact with the ground 1. In this case, the determination target selection unit 13 selects the object 2 as the target for determining the installation position.

On the other hand, as shown in Figure 4D, if the pixel at the ground 1 side edge of the pixel area of object 2 in the pixel area detection image is adjacent to a pixel in the background, it is considered that object 2 is not in contact with ground 1. It is difficult to determine the land classification of the installation location of such object 2. Therefore, the judgment target selection unit 13 excludes from the judgment target any object 2 whose ground 1 side edge pixel in the pixel area detection image is not adjacent to a pixel of at least one land among multiple land classifications.

The determination target selection unit 13 excludes from the determination of the land classification of the installation location at least one of the following objects 2, as described with reference to Figures 4A to 4D: objects 2 whose pixel region detection image does not include a lower end that abuts the ground 1; objects 2 whose size in a predetermined direction in the pixel region detection image is equal to or smaller than a threshold; and objects 2 whose pixel at the edge on the ground 1 side in the pixel region detection image is not adjacent to pixels of at least one of the land classifications. The object region extraction unit 14, lower end region expansion unit 15, and classification determination unit 16 process the objects 2 selected as the determination target by the determination target selection unit 13.

Referring again to Figure 1, the object region extraction unit 14 generates an object region extraction image by extracting an image of the region near the bottom edge of the object 2 that abuts the ground 1 from the pixel region detection image. Specifically, as shown in Figure 5, the object region extraction unit 14 identifies a circumscribing rectangle of the object 2 in the pixel region detection image. The object region extraction unit 14 identifies a circumscribing rectangle of the object 2 based on the maximum/minimum values of the coordinates in the X-axis and Y-axis directions of the pixel region of the object 2. The object region extraction unit 14 extracts an image of the region of the identified circumscribing rectangle from the pixel region detection image, as shown in Figure 5.

Next, the object region extraction unit 14 divides the extracted image into n parts (n is any positive integer). Specifically, as shown in FIG. 5, if the size of the extracted image in the X-axis direction is B [pixels] and the size of the extracted image in the Y-axis direction is H [pixels], the object region extraction unit 14 divides the extracted image into n parts in the Y-axis direction. For example, the object region extraction unit 14 may divide the extracted image in the Y-axis direction by a predetermined number of divisions n. Alternatively, the object region extraction unit 14 may divide the extracted image by varying the number of divisions n so that the size of the image after division is fixed (B × (n/H) is constant).

The object region extraction unit 14 outputs the image furthest from the bottom of the object 2 as the object region extraction image, out of the n images obtained by dividing the extracted image. The object region extraction unit 14 may also output an image with a fixed size of B [pixels] x H [pixels] from the bottom of the circumscribing rectangle as the object region extraction image. As described above, the object 2 selected as the object for determination by the determination target selection unit 13 is in contact with the ground 1 in the pixel region detection image. Therefore, of the n images obtained by dividing the image of the circumscribing rectangle of the object 2, the image furthest from the bottom of the object 2 includes the bottom edge of the object 2 that is in contact with the ground 1. Therefore, the object region extraction unit 14 can generate an object region extraction image that includes the vicinity of the bottom edge of the object 2 that is in contact with the ground 1. By generating an object region detection image that includes the vicinity of the bottom edge of the object 2 that is in contact with the ground 1 in this way, it is possible to extract information necessary for determining the land classification of the installation location of the object 2.

Referring again to FIG. 1, the bottom end region expansion unit 15 expands the region of the image extracted as the object region extraction image in the pixel region detection image in a direction other than the direction in which the object 2 extends from the ground 1, as shown in FIG. 6. In the example shown in FIG. 6, the object 2 extends from the ground 1 upward on the paper. In this case, the bottom end region expansion unit 15 expands the region of the object region extraction image in the downward, leftward, and rightward directions on the paper, as shown in FIG. 6, to expand the object region extraction image. The bottom end region expansion unit 15 may, for example, expand the region of the object region extraction image in a direction other than the direction in which the object 2 extends by a predetermined number of pixels. The bottom end region expansion unit 15 may also, for example, expand the region of the object region extraction image in a direction other than the direction in which the object 2 extends by a predetermined multiple.

As described above, the object area extraction image generated by the object area extraction unit 14 is generated based on the circumscribing rectangle of the object 2 in the pixel area detection image. In this case, the object area extraction image may not include pixels of the government land 1A and pixels of the private land 1B. By expanding the object area extraction image using the bottom edge area expansion unit 15, it is possible to include in the object area extraction image the pixels of the government land 1A and pixels of the private land 1B necessary for the classification determination unit 16 (described later) to determine the land classification of the installation location of the object 2.

Referring again to FIG. 1, the classification determination unit 16 determines the land classification of the installation location of the object 2 based on the number of pixels in the pixel area of each of the multiple land classifications in the object area extraction image. Specifically, the classification determination unit 16 determines the land classification of the installation location of the object 2 based on the number of pixels in the pixel area of each of the multiple land classifications in the object area extraction image after expansion by the bottom area expansion unit 15.

The classification determination unit 16 counts the number of pixels C1 in the pixel area of government land 1A and the number of pixels C2 in the pixel area of private land 1B in the object area extraction image. Then, for example, if the number of pixels C1 is greater than or equal to the number of pixels C2, the classification determination unit 16 determines that the land classification of the installation location of object 2 is government land 1A, and if the number of pixels C1 is less than the number of pixels C2, the classification determination unit 16 determines that the land classification of the installation location of object 2 is private land 1B. The classification determination unit 16 may also determine the land classification of the installation location of object 2 based on the ratio between the number of pixels C1 and the number of pixels C2.

The output unit 17 outputs the judgment result made by the classification judgment unit 16. The output unit 17 outputs the judgment result, for example, by displaying it on a display unit provided in the image processing device 10 or another device. The output unit 17 also outputs the judgment result, for example, by audio output from an audio output unit provided in the image processing device 10 or another device.

In the image processing device 10 according to this embodiment, of the components for performing the image processing according to the present disclosure (pixel region detection unit 12, judgment target selection unit 13, object region extraction unit 14, bottom end region extension unit 15, and division determination unit 16), the judgment target selection unit 13 and bottom end region extension unit 15 are not essential components. Therefore, the image processing device 10 according to this embodiment is only required to include the pixel region detection unit 12, object region extraction unit 14, and division determination unit 16.

Next, the operation of the image processing device 10 according to this embodiment will be described. Figure 7 is a flowchart showing an example of the operation of the image processing device 10 according to this embodiment, and is a diagram for explaining the image processing method by the image processing device 10.

The input unit 11 receives an image captured within a predetermined range including the object 2 and acquires the input image (step S11).

The pixel area detection unit 12 generates a pixel area detection image from the captured image, showing the pixel area of the object 2, the pixel area of each of the multiple land divisions, and the pixel area of the background (step S12).

The determination target selection unit 13 selects from the pixel region detection image an object 2 to be used to determine the land classification of the installation location (step S13). Specifically, the determination target selection unit 13 excludes from the determination targets those objects 2 included in the pixel region detection image that are difficult to determine, and selects the remaining objects 2 as determination targets. The determination target selection unit 13 excludes from the determination targets the following objects 2: objects 2 whose lower ends that contact the ground 1 are not included in the pixel region detection image; objects 2 whose size in a specified direction in the pixel region detection image is equal to or less than a threshold; and objects 2 whose pixel at the edge on the ground 1 side in the pixel region detection image is adjacent to a pixel of at least one of the land segments of multiple land segments. The following processing is performed on the object 2 selected as a determination target by the determination target selection unit 13.

The object region extraction unit 14 generates an object region extraction image by extracting an image of the region near the bottom end of the object 2 that abuts the ground surface 1 from the pixel region detection image (step S14). Specifically, the object region extraction unit 14 identifies a circumscribing rectangle of the object 2 in the pixel region detection image, extracts an image of the region near the bottom end of the object 2 from the identified circumscribing rectangle, and generates the object region extraction image.

The lower end area expansion unit 15 expands the area of the image extracted as the object area extraction image in the pixel area detection image in a direction other than the extension direction of the object 2 from the ground 1, thereby expanding the object area extraction image (step S15).

The classification determination unit 16 determines the land classification of the installation location of the object 2 based on the number of pixels in the pixel area of each of the multiple land classifications in the object area extraction image (step S16). Specifically, the classification determination unit 16 determines the land classification of the installation location of the object 2 based on the number of pixels in the pixel area of each of the multiple land classifications in the object area extraction image after expansion by the bottom area expansion unit 15.

The output unit 17 outputs the result of the classification determination unit 16's determination of the land classification of the installation location of the object 2 (step S17).

As described above, in this embodiment, the judgment target selection unit 13 and the bottom end area extension unit 15 are not required components. Therefore, in the image processing method by the image processing device 10 according to this embodiment, the processing of step S13 performed by the judgment target selection unit 13 and the processing of step S15 performed by the bottom end area extension unit 15 are not required.

As such, the image processing device 10 according to this embodiment comprises a pixel area detection unit 12, an object area extraction unit 14, and a classification determination unit 16. The pixel area detection unit 12 generates a pixel area detection image showing the pixel area of the object 2 and the pixel area of each of the multiple land areas from a captured image of the object 2 placed on ground 1 consisting of multiple land areas. The object area extraction unit 14 generates an object area extraction image by extracting an image of the area near the bottom end of the object 2 that abuts on the ground 1 from the pixel area detection image. The classification determination unit 16 determines the land classification of the installation location of the object 2 based on the number of pixels in the pixel area of each of the multiple land areas in the object area extraction image.

According to the image processing device 10 of this embodiment, the land classification of the installation location of the object 2 can be determined more easily using only the captured image of the object 2.

Second Embodiment
Fig. 8 is a diagram showing an example of the configuration of an image processing device 10A according to the second embodiment of the present disclosure. In Fig. 8, the same components as those in Fig. 1 are denoted by the same reference numerals, and description thereof will be omitted.

As shown in Figure 8, the image processing device 10A according to this embodiment includes an input unit 11, a pixel region detection unit 12, an object region extraction unit 14A, a bottom region extension unit 15A, a division determination unit 16, an output unit 17, and a bottom region detection unit 18. Compared to the image processing device 10 according to the first embodiment, the image processing device 10A according to this embodiment differs in that the determination target selection unit 13 has been removed, the object region extraction unit 14 has been changed to an object region extraction unit 14A, the bottom region extension unit 15 has been changed to a bottom region extension unit 15A, and a bottom region detection unit 18 has been added.

The bottom edge area detection unit 18 receives the captured image from the input unit 11. As shown in Figure 9, the bottom edge area detection unit 18 detects the bottom edge area in the captured image, which is the area near the bottom edge of the object 2 that is in contact with the ground 1. The bottom edge area detection unit 18 detects the bottom edge area using, for example, image recognition technology that detects a rectangular area such as a bounding box.

Referring again to Figure 8, the lower end area extension unit 15A extends the lower end area in the captured image in a direction other than the extension direction of the object 2 from the ground 1.

The object region extraction unit 14A extracts, as the object region extraction image, an image of the region in the pixel region detection image that corresponds to the bottom edge region. Specifically, the object region extraction unit 14A extracts, as the object region extraction image, an image of the region in the pixel region detection image that corresponds to the bottom edge region after expansion by the bottom edge region expansion unit 15A.

The image processing device 10A of this embodiment can automatically detect the area near the bottom end of the object 2 that is in contact with the ground 1 from the captured image, and generate an object area extraction image, which is an image of the area near the bottom end of the object 2 in the pixel area detection image.

(Third embodiment)
10 is a diagram showing an example of the configuration of an image processing device 10B according to the third embodiment of the present disclosure. In Fig. 10, the same components as those in Fig. 1 are denoted by the same reference numerals, and description thereof will be omitted.

As shown in Figure 10, the image processing device 10B according to this embodiment includes an input unit 11, a pixel region detection unit 12B, an object region extraction unit 14, a bottom region extension unit 15, a division determination unit 16, an output unit 17, and a trial count unit 19. Compared to the image processing device 10 according to the first embodiment, the image processing device 10B according to this embodiment differs in that the determination target selection unit 13 has been removed, the pixel region detection unit 12 has been replaced with a pixel region detection unit 12B, and a trial count unit 19 has been added.

The pixel area detection unit 12B can detect the pixel area of the object 2 in the captured image using multiple different methods. If the pixel area detection unit 12B cannot detect the pixel area of the object 2 using one method, it attempts to detect the pixel area of the object 2 using another method.

The trial counting unit 19 counts the number of attempts made by the pixel area detection unit 12 to detect the pixel area of the object 2.

If the pixel area of the object 2 is not detected, the pixel area detection unit 12B repeats detecting the pixel area of the object 2 using different methods until the number counted by the trial counting unit 19 reaches a predetermined value.

Next, we will explain the hardware configuration of the image processing devices 10, 10A, and 10B related to the present disclosure.

Figure 11 is a diagram showing an example of the hardware configuration of an image processing device 10, 10A, or 10B according to the present disclosure. Figure 11 shows an example of the hardware configuration of an image processing device 10, 10A, or 10B when the image processing device 10, 10A, or 10B is configured by a computer capable of executing program instructions. Here, the computer may be a general-purpose computer, a dedicated computer, a workstation, a PC (Personal Computer), an electronic notepad, or the like. The program instructions may be program code, code segments, or the like for performing necessary tasks.

As shown in FIG. 11, the image processing device 10, 10A, 10B has a processor 21, a ROM (Read Only Memory) 22, a RAM (Random Access Memory) 23, a storage 24, an input unit 25, a display unit 26, and a communication interface (I/F) 27. Each component is connected to each other via a bus 29 so that they can communicate with each other. The processor 21 is specifically a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), SoC (System on a Chip), etc., and may be composed of multiple processors of the same or different types.

The processor 21 is a control unit that controls each component and performs various arithmetic operations. That is, the processor 21 reads a program from the ROM 22 or the storage 24 and executes the program using the RAM 23 as a work area. The processor 21 controls each component and performs various arithmetic operations in accordance with the program stored in the ROM 22 or the storage 24. In this embodiment, the ROM 22 or the storage 24 stores a program for operating a computer as the image processing device 10, 10A, or 10B according to the present disclosure. The program is read and executed by the processor 21 to realize each component of the image processing device 10, 10A, or 10B, namely, the pixel area detection unit 12, 12B, the judgment target selection unit 13, the object area extraction unit 14, 14A, the bottom area extension unit 15, 15A, the division determination unit 16, the bottom area detection unit 18, and the trial count unit 19.

The program may be provided in a form stored on a non-transitory storage medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. The program may also be provided in a form that can be downloaded from an external device via a network.

ROM 22 stores various programs and data. RAM 23 temporarily stores programs or data as a working area. Storage 24 is composed of an HDD (Hard Disk Drive) or SSD (Solid State Drive) and stores various programs and data, including the operating system.

The input unit 25 includes a pointing device such as a mouse and a keyboard, and is used to make various inputs.

The display unit 26 is, for example, a liquid crystal display, and displays various information. The display unit 26 may also function as the input unit 25 by adopting a touch panel system. The output unit 17 outputs the results of the judgment by the classification judgment unit 16, for example, by displaying them on the display unit 26.

The communication interface 27 is an interface for communicating with other devices (e.g., a camera that has captured an image of an object), and is, for example, an interface for a LAN. The input unit 11 acquires captured images from other devices, for example, by communication via the communication interface 27.

A computer can be suitably used to function as each part of the image processing device 10 described above. Such a computer can be realized by storing a program describing the processing content that realizes the functions of each part of the image processing device 10 in the computer's memory, and having the computer's processor read and execute this program. In other words, the program can cause the computer to function as the image processing device 10 described above. The program can also be stored on a non-transitory storage medium. The program can also be provided via a network.

The following notes are further disclosed regarding the above embodiments.

[Additional note 1]
1. An image processing device that determines a land classification of an installation position of an object from a captured image of the object installed on a ground consisting of a plurality of land classifications,
Memory and
a control unit connected to the memory;
Equipped with
The control unit
generating a pixel area detection image indicating a pixel area of the object and a pixel area of each of the plurality of land divisions from the captured image;
generating an object region extracted image by extracting an image of a region near the bottom end of the object that is in contact with the ground surface from the pixel region detected image;
an image processing device that determines the land classification of the installation position of the target object based on the number of pixels in the pixel area of each of the plurality of land classifications in the object area extraction image.

[Additional note 2]
In the image processing device according to claim 1,
The control unit
excluding at least one of an object whose lower end that abuts on the ground is not included in the pixel region detection image, an object whose size in a predetermined direction in the pixel region detection image is equal to or smaller than a threshold, and an object whose pixel at the end on the ground side in the pixel region detection image is not adjacent to a pixel of at least one of the land of the plurality of land divisions from the object of determining the land division of the installation position;
an image processing device that generates the object region extraction image for the object selected as the object to be determined;

[Additional note 3]
In the image processing device according to claim 1 or 2,
The control unit
a circumscribing rectangle of the object in the pixel region detection image, and an image of the vicinity of the lower end of the object in the specified circumscribing rectangle is extracted to generate the object region extraction image;
expanding the object region extraction image by expanding the region of the image extracted as the object region extraction image in a direction other than the extending direction of the object from the ground;
an image processing device that determines the land classification of the installation position of the target object based on the number of pixels in the pixel area of each of the plurality of land classifications in the expanded object area extraction image.

[Additional note 4]
In the image processing device according to claim 1,
The control unit
a lower end region in the captured image that is a region near a lower end of the object that is in contact with the ground;
an image processing device that extracts an image of a region corresponding to the lower end region in the pixel region detection image as the object region extraction image;

[Additional note 5]
In the image processing device according to supplementary item 4,
The control unit
a lower end region in the captured image is expanded in a direction other than an extension direction of the object from the ground;
an image processing device that extracts, as the object region extraction image, an image of a region in the pixel region detection image that corresponds to the lower end region after the expansion;

[Additional note 6]
In the image processing device according to any one of claims 1 to 5,
The control unit detects the pixel area of the object in the captured image using a plurality of different methods.

[Additional note 7]
1. An image processing method for an image processing device that determines a land classification of an installation position of an object from a captured image of the object installed on a ground consisting of a plurality of land classifications, the method comprising:
generating a pixel area detection image indicating a pixel area of the object and a pixel area of each of the plurality of land divisions from the captured image;
generating an object region extracted image by extracting an image of a region near the bottom end of the object that is in contact with the ground surface from the pixel region detected image;
An image processing method for determining a land category in which the object is installed based on the number of pixels in a pixel area of each of the plurality of land categories in the object area extraction image.

[Additional Note 8]
A non-transitory storage medium storing a program executable by a computer, the non-transitory storage medium storing the program causing the computer to operate as the image processing device described in any one of appendix items 1 to 6.

The above-described embodiments have been described as representative examples, but it will be apparent to those skilled in the art that numerous modifications and substitutions are possible within the spirit and scope of the present disclosure. Therefore, the present invention should not be construed as being limited to the above-described embodiments, and various modifications and alterations are possible without departing from the scope of the claims. For example, multiple component blocks shown in the configuration diagrams of the embodiments may be combined into one, or one component block may be divided.

REFERENCE SIGNS LIST 1 Ground 1A Public land 1B Private land 2 Object 10, 10A, 10B Image processing device 11 Input unit 12, 12B Pixel area detection unit 13 Determination target selection unit 14, 14A Object area detection unit 15, 15A Bottom end area extension unit 16 Classification determination unit 17 Output unit 18 Bottom end area detection unit 19 Trial number count unit 21 Processor 22 ROM
23 RAM
24 Storage 25 Input unit 26 Display unit 27 Communication I/F
29 Bus

Claims (8)

1. An image processing device that determines a land classification of an installation position of an object from a captured image of the object installed on a ground consisting of a plurality of land classifications,
a pixel area detection unit that generates, from the captured image, a pixel area detection image that indicates a pixel area of the object and a pixel area of each of the plurality of land divisions;
an object region extraction unit that generates an object region extraction image by extracting an image of a region near the bottom end of the object that is in contact with the ground surface from the pixel region detection image;
an area determination unit that determines the land area of the installation position of the object based on the number of pixels in the pixel area of each of the plurality of land areas in the object area extraction image. 2. The image processing device according to claim 1,
a determination object selection unit that excludes at least one of an object whose lower end that abuts on the ground is not included in the pixel region detection image, an object whose size in a predetermined direction in the pixel region detection image is equal to or smaller than a threshold, and an object whose pixel at the end on the ground side in the pixel region detection image is not adjacent to a pixel of at least one of the land of the plurality of land divisions, from targets for determining the land division of the installation position;
The object region extraction unit generates the object region extracted image for the object selected as the target for determination by the determination target selection unit. 2. The image processing device according to claim 1,
the object region extraction unit specifies a circumscribing rectangle of the object in the pixel region detection image, extracts an image of the vicinity of the lower end of the object in the specified circumscribing rectangle, and generates the object region extraction image;
a bottom end region extending unit that extends the region of the image extracted as the object region extracted image in the pixel region detection image in a direction other than the extending direction of the object from the ground surface, thereby extending the object region extracted image;
The division determination unit determines the land division of the installation position of the object based on the number of pixels in the pixel area of each of the plurality of land divisions in the object area extraction image after expansion by the lower end area expansion unit. 2. The image processing device according to claim 1,
a bottom edge region detection unit that detects a bottom edge region in the captured image, the bottom edge region being a region near the bottom edge of the object that is in contact with the ground;
The object region extraction unit extracts an image of a region corresponding to the lower end region in the pixel region detection image as the object region extraction image. 5. The image processing device according to claim 4,
a bottom region extending unit that extends a bottom region in the captured image in a direction other than an extension direction of the object from the ground;
The object region extraction unit extracts, as the object region extraction image, an image of a region in the pixel region detection image that corresponds to the bottom edge region after extension by the bottom edge region extension unit. 2. The image processing device according to claim 1,
The pixel region detection unit detects the pixel region of the object in the captured image using a plurality of different methods. 1. An image processing method for an image processing device that determines a land classification of an installation position of an object from a captured image of the object installed on a ground consisting of a plurality of land classifications, the method comprising:
generating a pixel area detection image indicating a pixel area of the object and a pixel area of each of the plurality of land divisions from the captured image;
generating an object region extracted image by extracting an image of a region near the bottom end of the object that is in contact with the ground from the pixel region detected image;
and determining the land classification of the installation position of the object based on the number of pixels in the pixel area of each of the plurality of land classifications in the object area extraction image. A program that causes a computer to operate as the image processing device described in claim 1.