Local identification of 3D point cloud based on structured light

Structured light scanning technology has the characteristics of high accuracy, low environmental impact, low power consumption, and fast acquisition speed, which is very suitable for real-time 3D point cloud extraction. Therefore, a local feature extraction method based on structured light is proposed in this paper. In this method, a structured light system based on digital fringe projection technology was used to obtain the actual measurement area according to the phase constraint condition and the flood filling algorithm, and the overdetermined equations were established to solve the point cloud data with the internal and external parameters obtained by the structured light system calibration. Then the ISS method is used to extract the key points in the point cloud. According to the key points and normal, PFH is used to establish local feature descriptors. Finally, the point correspondence between the two groups of point cloud data is obtained by the RANSAC algorithm to complete feature matching. Experiments show that the algorithm can complete the feature recognition work better and faster, and meet the requirements of industrial occasions to a certain extent.