An improved similarity measure algorithm based on point feature histogram

Currently, in 3D point cloud data field, different methods based on multi-value characteristics, which are utilized to measure the similarity between different point cloud data, are developed. These features are more dependent on low dimensional normal vector and curvature. In this paper, feature histogram of each point of the point cloud is calculated in high dimensional space. Through global feature information clustering, the mathematical distribution of feature is established in order to obtain the unique feature expression of point cloud data. According to the feature of point cloud data, this paper puts forward a new algorithm suitable for the similarity measure of point cloud data. The experiment result shows the improved algorithm works well for some occasions.