摘要 |
Information on a plurality of different templates (12a to 12c) that extract a predetermined number of pixels from pixels surrounding a pixel to be detected of a face part image is stored, A pixel is extracted using the stored different templates (12a to 12c), a face part image included in an image is detected using a feature amount of the extracted pixel, and a face image is detected from the image based on the detected face part image, so that accuracy of detecting a face image can be improved and the process time for detecting the face image can be shortened. |