Research on digital detection method based on edge point extraction
Jiahao Chen, Yunxia Chen, Honggang Yang · 2022
With the rapid progress of industrial level, the inspection of product manufacturing deviation by scanning point cloud has become one of the key steps of quality inspection automation. In order to improve the accuracy and efficiency of machining quality detection, this paper proposes a grid division method based on normal line. After point cloud model segmentation, noise reduction and downsampled , the secondary optimization of threshold parameters, combined with the radius outlier elimination method, a more complete and accurate edge point extraction is realized. Finally, taking the diesel engine runner shell as the study object, the error rate of the machining size of the key parts is calculated by using the edge point parameters. The results show that this method can avoid the edge point extraction error caused by improper threshold, effectively reducing the error rate of key parts detection, improve the detection accuracy and speed of product manufacturing deviation, and provide reference for the digital detection of product processing accuracy.