摘要 |
A feature point calculating section binarizes the image data to obtain a centroid of a consecutive component in which pixels are connected as a feature point, reverses the image data, obtains a centroid as a feature point from the reversed image data similarly, and adds them as a feature point of the image data. A features calculating section calculates a predetermined invariant based on the feature point containing the feature point obtained from the reversed image data, and calculates a hash value based on the predetermined invariant. A vote process section retrieves a hash table based on the calculated hash value, votes for a document of an index stored in association with the hash value, and accumulatively adds the vote. A similarity determination process section compares the number of votes calculated by the vote process section with a predetermined threshold value to determine a similarity.
|