Computer-readable recording medium storing position identification program, position identification method, and information processing apparatus

1. A non-transitory computer-readable recording medium storing a position identification program for causing a computer to execute processing comprising:

acquiring a captured image and a depth image that represents a distance from an imaging position and corresponds to the captured image;

identifying a road region and another region in contact with the road region from the captured image;

calculating a change in a depth of a first region that corresponds to the road region included in the depth image and a change in a depth of a second region that corresponds to the another region included in the depth image;

determining whether or not the another region is a detection target on the basis of the change in the depth of the first region and the change in the depth of the second region; and

identifying a position of a subject included in the another region on the basis of the depth of the second region and the imaging position in a case where it is determined that the another region is the detection target,

wherein the processing of determining includes

processing of determining that the another region is the detection target in a case where the change in the depth of the first region before and after the second region matches the second region that corresponds to the another region and the change in the depth of the second region that corresponds to the second region is different from the change in the depth of the first region.

2. The non-transitory computer-readable recording medium storing the position identification program according to claim 1 , wherein

the processing of acquiring includes

processing of acquiring the depth image that corresponds to the captured image by inputting the captured image into a machine learning model that outputs a depth image according to an input of a Red Green Blue (RGB) image.

3. The non-transitory computer-readable recording medium storing the position identification program according to claim 1 , wherein

the processing of calculating includes

processing of calculating a change in a depth of the first region that corresponds to the road region and a change in a depth of the second region that corresponds to the another region from a lower portion toward an upper portion of the depth image in a depth direction, and

the processing of determining includes

processing of determining that the another region is the detection target in a case where tendency of the change in the depth of the second region is different from tendency of the change in the depth of the first region by a value equal to or more than a threshold.

4. The non-transitory computer-readable recording medium storing the position identification program according to claim 1 , wherein

the processing of calculating includes

processing of calculating a change in a depth of the first region that corresponds to the road region and a change in a depth of the second region that corresponds to the another region from a lower portion toward an upper portion of the depth image in a depth direction and

converting each of the change in the depth of the first region and the change in the depth of the second region into each piece of linear regression equation information that indicates a relationship between the number of pixels in the depth direction and a depth of the depth image, and

the processing of determining includes

processing of determining that the second region is not a detection target in a case where there is a statistical significant difference between the pieces of linear regression equation information and determining that the second region is the detection target in a case where there is no statistical significant difference.

5. The non-transitory computer-readable recording medium storing the position identification program according to claim 4 , wherein

the processing of identifying a position of a subject includes

processing of identifying the position of the subject included in the another region on the basis of the depth of the second region and the imaging position in a case where it is determined that the second region is the detection target on the basis of each piece of the linear regression equation information and

processing of associating image data of the subject with a position of the subject on map data.

6. The non-transitory computer-readable recording medium storing the position identification program according to claim 4 , wherein

the processing of identifying a position of a subject includes

processing of identifying the position of the subject on the basis of the depth of the top of the first region and the imaging position in a case where it is determined that the second region is not the detection target on the basis of each piece of the linear regression equation information and

processing of associating image data of the subject with a position of the subject on map data.

7. An information processing apparatus comprising:

a memory; and a

processor coupled to the memory and configured to:

acquire a captured image and a depth image that represents a distance from an imaging position and corresponds to the captured image;

identify a road region and another region in contact with the road region from the captured image;

calculate a change in a depth of a first region that corresponds to the road region included in the depth image and a change in a depth of a second region that corresponds to the another region included in the depth image;

determine whether or not the another region is a detection target on the basis of the change in the depth of the first region and the change in the depth of the second region; and

identify a position of a subject included in the another region on the basis of the depth of the second region and the imaging position in a case where it is determined that the another region is the detection target,

wherein a processing to determine includes

processing of determining that the another region is the detection target in a case where the change in the depth of the first region before and after the second region matches the second region that corresponds to the another region and the change in the depth of the second region that corresponds to the second region is different from the change in the depth of the first region.

8. A position identification method comprising:

acquiring, by a computer, a captured image and a depth image that represents a distance from an imaging position and corresponds to the captured image;

identifying a road region and another region in contact with the road region from the captured image;

calculating a change in a depth of a first region that corresponds to the road region included in the depth image and a change in a depth of a second region that corresponds to the another region included in the depth image;

determining whether or not the another region is a detection target on the basis of the change in the depth of the first region and the change in the depth of the second region; and

identifying a position of a subject included in the another region on the basis of the depth of the second region and the imaging position in a case where it is determined that the another region is the detection target,

wherein the determining includes

processing of determining that the another region is the detection target in a case where the change in the depth of the first region before and after the second region matches the second region that corresponds to the another region and the change in the depth of the second region that corresponds to the second region is different from the change in the depth of the first region.