STEREO IMAGE FEATURE MATCHING SYSTEM AND METHOD BASED ON HARRIS CORNER VECTOR CLUSTERING ALGORITHM

摘要

The present invention relates to a stereo image feature point matching system and method using a Harris corner vector clustering algorithm comprising: a feature point extraction unit for extracting feature points constituting each object by changing a contrast value of two images each obtained by a stereo camera; a clustering unit for clustering feature points within a predetermined range from each feature point of the plurality of adjacent feature points extracted by the feature point extraction unit into one cluster; and a matching unit for comparing the respective clusters clustered by the clustering unit in the two binocular images of the stereo camera to find a best match and detect the same object sharing each other.