Object recognition device
The purpose of the present invention is to obtain reliable recognition performance even in a lens which is capable of long-distance and close proximity photography. This object recognition device 1 comprises an image acquisition unit which acquires an image 1000 which includes a plurality of image regions 1001, 1002, 1003 which have differing resolutions, and a recognition processing unit which carries out a recognition process upon an object within the screen 1000 . One image region 1003 , among the plurality of image regions 1001, 1002, 1003 of the screen 1000 , has a lower resolution than the other image regions 1001, 1002 . The recognition processing unit carries out the recognition process of the object on the basis of assessment references which differ among the one image region 1003 and the other image regions 1001, 1002.
1. An object recognition device comprising:
an image acquirer, including a lens, configured to acquire an image including a plurality of image regions having different resolutions; and
at least one processor configured to perform a recognition process on an object in the image,
wherein
one image region of the image among the plurality of image regions has a lower resolution than another image region of the image, and
the at least one processor is configured to perform the recognition process on the object on the basis of different assessment references for each of the one image region and the another image region,
the another image region corresponds to an inner image region located on an inner side of the image and an outer image region located on an outer side of the image,
the one image region corresponds to an intermediate image region located between the inner image region and the outer image region,
the intermediate image region has a lower resolution than the inner image region and the outer image region,
a threshold of an assessment reference in the intermediate image region is smaller than a threshold of an assessment reference in the inner image region and the outer image region, and
the intermediate image region has a toroidal shape.
2. The object recognition device according to claim 1 , wherein the at least one processor is configured to perform the recognition process on the object on the basis of different thresholds related to a luminance value.
3. The object recognition device according to claim 1 , wherein the at least one processor is configured to perform the recognition process on the object by pattern matching on the basis of patterns different for each of the one image region and the another image region.
4. The object recognition device according to claim 1 , wherein the at least one processor is configured to extract a contour of the object in the image, and
wherein in a case where a coordinate position of the object in the image is estimated to be included in the intermediate image region, the at least one processor uses a contour extraction threshold smaller than a contour extraction threshold to be used in the inner image region and the outer image region.
5. The object recognition device according to claim 1 ,
wherein the lens is arranged to have a magnification of a first incident angle site, having an incident angle smaller than an inflection point incident angle, and greater than a magnification of a second incident angle site, having an incident angle larger than the inflection point incident angle.
6. An object recognition device comprising:
an image acquirer, including a lens, configured to acquire an image including a plurality of image regions having different resolutions; and
at least one processor configured to perform a recognition process on an object in the image,
wherein one image region of the image among the plurality of image regions has a lower resolution than another image region,
an image to be obtained is an image photographed at a photographing timing at which the object relatively moving with respect to an own vehicle is included in the another image region,
the another image region corresponds to an inner image region located on an inner side of the image and an outer image region located on an outer side of the image,
the one image region corresponds to an intermediate image region located between the inner image region and the outer image region,
the intermediate image region has a lower resolution than the inner image region and the outer image region, and
the intermediate image region has a toroidal shape.
7. The object recognition device according to claim 6 ,
wherein in a case where the coordinate position of the object in the image is estimated to be included in the one image region by photographing with a predetermined fixed cycle, the photographing timing is changed so as to allow the coordinate position of the object in the image to be included in the another image region.
8. The object recognition device according to claim 6 ,
wherein the lens is arranged to have a magnification of a first incident angle site, having an incident angle smaller than an inflection point incident angle, and greater than a magnification of a second incident angle site, having an incident angle larger than the inflection point incident angle.