| 摘要 |
PROBLEM TO BE SOLVED: To provide a technique capable of increasing detection accuracy in a detection object image.SOLUTION: A detection unit 3 sets a plurality of detection frames having different positions from each other, to a processing object image and performs detection processing of a detection object image to an image in the detection frames for each of the plurality of detection frames. A map generation unit 4 generates a map showing distribution in the processing object image of detection accuracy values showing accuracy as the detection object image on the basis of a detection result of the detection unit 3. A maximum point retrieval unit 5 retrieves a maximum point of the detection accuracy values in the map generated by the map generation unit 4. A detection object image determination unit 6 determines a predetermined area including a pixel having the same position as the maximum point retrieved by the maximum point retrieval unit 5, as a detection object image in the processing object image. |
