Image recognition apparatus

1. An image recognition apparatus comprising:

circuitry configured to:

input an image of an object captured by a camera;

divide, based on a predetermined positioning point, an entirety of the image into a plurality of divided regions using rays;

set a process region that includes an entirety of one of the divided regions and at least a part of another one of the divided regions that is adjacent to the one of the divided regions, so as to fully include the object in the process region when the object is located on a boundary between the one of the divided regions and the another one of the divided regions;

set a rotation of the process region so that a positional relationship between up and down of the object in the process region matches up; and

perform the set rotation to a portion of the image corresponding to the process region and perform a recognition process to the portion of the image after the rotation to detect a presence of the object.

2. The apparatus according to claim 1 , wherein the circuitry is further configured to:

perform rotation of a result of the recognition process, the rotation of the result being a rotation opposite to the set rotation, to convert the result of the recognition process into a result of the image that has been input; and

output the result of the image that has been input.

3. The apparatus according to claim 1 , wherein the recognition process comprises a process for detecting a location of the object in the image.

4. The apparatus according to claim 1 , wherein the predetermined positioning point is a location in the image where two portions of the object are overlapping.

5. The apparatus according to claim 1 , wherein an angle of the set rotation is determined such that an angle of a straight line connecting two portions of the object in the process region falls within a predetermined range.

6. The apparatus according to claim 1 , wherein the process region is rectangular.

7. The apparatus according to claim 3 , wherein the object comprises a person.

8. The apparatus according to claim 4 , wherein the object comprises a person, and the two portions of the object comprise a head and a foot of the person.

9. The apparatus according to claim 1 , wherein, in the recognition process, an image is created by rotating the process region.

10. The apparatus according to claim 9 , wherein, in the recognition process, images are created by rotating the process region, and the images are connected to create a single image.

11. The apparatus according to claim 9 , wherein the circuitry is further configured to, in response to an input of a point in the image, generate images in which the point is set as the predetermined positioning point, and to display the generated images side by side.

12. The apparatus according to claim 1 , wherein the circuitry comprises:

an input circuit to which the image of the object captured by the camera is input;

a setting circuit configured to divide, based on the predetermined positioning point, the entirety of the image that has been input by the input circuit into the plurality of divided regions using rays, set the process region that comprises the respective one of the divided regions so as to fully include the object on a boundary of the respective one of the divided regions, and set the rotation of the process region so that the positional relationship between up and down of the object in the process region matches up; and

a recognition circuit configured to perform the set rotation set by the setting circuit to the portion of the image corresponding to the process region and perform the recognition process to the portion of the image after the rotation to detect the presence of the object.

13. The apparatus according to claim 12 , wherein the circuitry further comprises:

a conversion circuit configured to perform rotation of a result of the recognition processing performed by the recognition circuit, the rotation of the result being a rotation opposite to the set rotation set by the setting circuit, to convert the result of the recognition process into a result of the image that has been input; and

an output circuit configured to output the result converted by the conversion circuit.

14. The apparatus according to claim 1 , wherein the circuitry comprises at least one dedicated computer processor configured to:

input the image of the object captured by the camera;

divide, based on the predetermined positioning point, the entirety of the image that has been input into the plurality of divided regions using rays;

set the process region that comprises the respective one of the divided regions so as to fully include the object on a boundary of the respective one of the divided regions;

set the rotation of the process region so that the positional relationship between up and down of the object in the process region matches up;

perform the set rotation to the portion of the image corresponding to the process region; and

perform the recognition process to the portion of the image after the rotation to detect the presence of the object.

15. The apparatus according to claim 14 , wherein the at least one dedicated computer processor is further configured to:

perform rotation of a result of the recognition process, the rotation of the result being a rotation opposite to the set rotation, to convert the result of the recognition process into a result of the image that has been input; and

output the result of the image that has been input.

16. An image recognition method comprising:

inputting an image of an object captured by a camera;

dividing, based on a predetermined positioning point, an entirety of the image into a plurality of divided regions using rays;

configuring a process region that an entirety of one of the divided regions and at least a part of another one of the divided regions that is adjacent to the one of the divided regions, so as to fully include the object in the process region when the object is located on a boundary between the one of the divided regions and the another one of the divided regions;

configuring a rotation of the process region so that a positional relationship between up and down of the object in the process region matches up;

performing the rotation to a portion of the image corresponding to the process region; and

performing a recognition process to the portion of the image after performing the rotation to detect a presence of the object.

17. A non-transitory computer-readable storage medium having an image recognition program stored thereon, the image recognition program being executable to cause a computer to:

input an image of an object captured by a camera;

divide, based on a predetermined positioning point, an entirety of the image into a plurality of divided regions using rays;

set a process region that includes an entirety of one of the divided regions and at least a part of another one of the divided regions that is adjacent to the one of the divided regions, so as to fully include the object in the process region when the object is located on a boundary between the one of the divided regions and the another one of the divided regions;

set a rotation of the process region so that a positional relationship between up and down of the object in the process region matches up;

perform the rotation to a portion of the image corresponding to the process region; and

perform a recognition process to the portion of the image after the rotation to detect a presence of the object.

18. The image recognition apparatus according to claim 1 , wherein the process region further includes the predetermined positioning point.