| In this era of rapid industrial development,three-dimensional high-precision vision systems have been widely used to measure object morphology.3D morphology measurement is one of the main research topics in visual measurement,and 3D stitching technology based on binocular stereo vision is an effective method for measuring the 3D morphology of large or high-precision objects.The key step in 3D morphology measurement is to accurately calculate the 3D morphology of various parts of irregular parts and the splicing technology of 3D point clouds.In the process of restoring the three-dimensional morphology of objects,the stitching effect is an important indicator to measure the overall accuracy.Therefore,the focus of this article is on computational accuracy and efficiency.This article first adopts the Three-Dimensional Digital Image Correlation(3D-DIC)non-contact three-dimensional binocular measurement system,and uses the established binocular vision three-dimensional morphology measurement platform to measure the surface morphology of components in blocks and calculate the three-dimensional morphology of each part of the component.Secondly,the Iterative Closest Point(ICP)algorithm was used to concatenate multi region data and obtain a complete three-dimensional morphology map of the component.In the stage of partial 3D morphology measurement,first perform exhaustive search to obtain accurate initial values;Then,the forward Newton’s method is used to classify pixels,and a first-order nine point interpolation is designed to quickly obtain Jacobian matrix and Hessian matrix elements,so as to achieve accurate sub-pixel positioning.The experimental results show that the improved method has high accuracy,and its stability of mean error and standard deviation is better than other algorithms;Compared with the traditional forward Newton’s method method,the total iteration time of the improved forward Newton method in the sub-pixel iteration phase is reduced,and the calculation efficiency is 3.8 times of the traditional NR algorithm.The entire process of the algorithm proposed in this article is concise and efficient,and has application value in precision situations with high precision requirements.To obtain the overall morphology data of the tested area,the next step is to use the ICP algorithm to concatenate the 3D-DIC data calculated from each part,convert the 3D morphology information into point cloud data,and visualize it.This article provides a detailed analysis of the theoretical methods and process derivation of point cloud stitching,including the definition of rigid body transformation and the parameter solution of rigid body transformation;A 3D point cloud data mosaic method based on ICP algorithm is proposed,which is combined with k-d tree.This method can effectively reduce the complexity of operations,significantly improve measurement accuracy,and reduce research costs.Compared with traditional splicing measurement methods,this method has significant advantages,ensuring measurement accuracy while improving the efficiency of reconstruction.It is a more effective and fast method for splicing the three-dimensional morphology of irregular parts. |
