Block mass association method and device based on gradient histogram features

The invention discloses a block mass association method and device based on gradient histogram features, and the method comprises the steps: obtaining a contour point set corresponding to each block mass in a to-be-associated block mass group, so as to obtain a feature point position set; extracting corresponding feature point information according to the feature point position set; randomly extracting n pieces of feature point information from a feature point information set corresponding to the first associated block mass group, and performing distance judgment on the n pieces of feature point information and all feature point information corresponding to the second associated block mass group to obtain corresponding matching objects; calculating an affine matrix according to the feature point positions corresponding to the n pieces of feature point information and the feature point positions corresponding to the matching objects; adopting an affine matrix to perform affine transformation on the feature poi