| 摘要 |
<P>PROBLEM TO BE SOLVED: To accurately calculate the probability that an object image belongs to given classification information even when the probability distribution of image features learned for classification is divided into plural. <P>SOLUTION: An image processor 10 acquires the probability distribution of image feature information for each sub classification for which the image feature information belonging to respective classifications is classified into plural, calculates an area likelihood corresponding to the probability that an image region belongs to a given classification, on the basis of the image feature information obtained from the image region set for the object image, the probability distribution of the image feature information to each sub classification for the given classification and weight determined for each sub classification, calculates a posterior probability indicating the probability that the object image belongs to each sub classification on the basis of the area likelihood calculated for each sub classification, updates the weight determined for each sub classification on the basis of the posterior probability calculated for each sub classification, and calculates an evaluation value corresponding to the probability that the object image belongs to the given classification, on the basis of the area likelihood calculated after the weight determined for each sub classification is updated. <P>COPYRIGHT: (C)2012,JPO&INPIT |
