A High Precision Location Method for Point Cloud Registration and Improved Least Squares Method
In order to solve the problem that the accuracy of chip contour detection is easily disturbed by nozzle imaging,an algorithm based on point cloud registration and improved least square method is pro-posed.The algorithm describes the chip contour through joint optimization of nozzle positioning and rectan-gular positioning.In the nozzle positioning stage,the point cloud registration algorithm of fusion KDTree is used to accurately locate the nozzle position and remove the outline of the nozzle part.In the stage of rec-tangular positioning,a linear fitting method is created to obtain the coordinate point set of four sides of the rectangle,and the normal distribution and radian information are introduced to improve the least square method to fit the rectangular edge to obtain the precise coordinate information of the chip.The proposed al-gorithm solves the problem of inaccurate positioning when the rectangular chip is disturbed by the nozzle imaging.The experimental results show that the algorithm can achieve more than 98％ accuracy in the rec-tangular chip image with interference,the accuracy is controlled within 0.1 mm and the execution time of the algorithm is about 50 ms.In summary,the proposed algorithm has good robustness and accuracy,and pro-vides a reliable solution for solving the rectangular chip positioning problem in complex industrial scenes.
point cloud registrationrectangle detectionstraight line bondingleast square
万方数据
维普
