摘要 |
<P>PROBLEM TO BE SOLVED: To further increase resistance against image loss, noise, and the like in techniques related to image recognition. <P>SOLUTION: A predetermined feature point obtained from an input image is extracted. A locus image that indicates a locus specifying a predetermined graphic included in the input image and corresponds to a feature point is acquired using a Hough transform. A recognition target object is detected from an input image, based on feature quantities, using an identifier generated by statistical learning using the plurality of feature quantities obtained from a locus image obtained based on a learning image including a recognition target object and a locus image obtained based on a learning image including no recognition target object. <P>COPYRIGHT: (C)2013,JPO&INPIT |