Research On Measuring Technique Of Body Tube Damage By Fusion Of 3D Point Cloud And Image

As the key component of the gun,the quality of the gun barrel has an important impact on the performance and safety of the gun.When the gun is launched,the high temperature and high pressure gas produced will cause ablation on the interior surface,and the projectile will also cause impact and wear on it,resulting in defects,affecting the normal launch of the projectile.Therefore,the quality inspection of the inner chamber has become one of the important items in the production and use of artillery.CCD,as a traditional method for detecting the inner bore of gun barrel,can obtain the panoramic image of the inner wall of gun barrel for defect detection,but it still has the disadvantages of easy misjudgment and unable to quantitatively analyze the defect degree.Therefore,how to obtain the effective information of the bore more accurately and comprehensively,give the overall situation and defect degree of the inner wall of the gun barrel quantitatively,and make the detection more automatic,has become the focus and focus of the current research on the bore detection of the gun barrel.In this paper,a method based on combining 2D and 3D information to identify the inner wall defects of gun barrel is proposed through the self-designed gun barrel bore detection system based on the combination of camera and laser displacement sensor.Firstly,the inner chamber detection system and its working principle are introduced,and the two-dimensional image of the body tube and the inner diameter depth data of the body tube are collected under the software drive,and the depth data is processed to obtain the point cloud data and format used for the threedimensional reconstruction.Secondly,the point cloud data is preprocessed: the holes in the point cloud are improved by the B-spline interpolation repair method,and the abnormal noise points are removed by the sparse and out-of-group point removal method according to the actual situation,and then the point cloud is reduced,so that the number of point clouds is simplified while retaining the point cloud characteristic information.According to the preprocessed point cloud data,the Poisson surface reconstruction algorithm is used to realize the surface reconstruction of the inner chamber of the gun barrel,and the three-dimensional contour image of the body tube is obtained.Finally,after the camera calibration,the coordinate conversion is completed,and the suspected defect area judged based on the two-dimensional image is mapped in the corresponding position in the three-dimensional image,and the defect detection is further carried out according to the depth information,and the degree of defect is calculated and located.In this paper,the characteristics of combining two-dimensional and three-dimensional characteristics for the defect detection method of the inner wall of the artillery barrel are proposed,and the defect detection of the inner wall of the barrel is completed,so that the inspectors can not only visually view the inner wall of the gun body,but also determine the specific location and damage degree of the defect through manual interaction,which enriches the technical means of artillery bore detection.