Three-Dimensional Surface Defect Detection Of Parts Based On Line Laser Scanning

Part surface defect detection is one of the important means to guarantee the quality of parts,and with the increasing number of global users of products,the requirements for product quality are also getting higher and higher.The quality of the product is not only reflected in its use function,but also has a great relationship with its appearance surface.However,most of the current surface defect detection methods focus on flat surface defect detection,and most defects have certain three-dimensional information.Obviously,such detection methods cannot meet the needs of production.In order to better express the three-dimensional surface information of the object and perform defect detection based on its three-dimensional information,this thesis proposes a three-dimensional surface defect detection system for parts based on line laser scanning.Firstly,the camera is calibrated,which can be used to rectify the image collected by the camera.And at the same time,the camera calibration result,which will be applied for coordinate conversion in subsequent image processing,can be obtained.Secondly,we can use the image acquisition system to acquire images and then perform Gaussian filtering,image difference,Gaussian smoothing,stripe centerline extraction,and coordinate conversion on the image in order to obtain the three-dimensional point cloud data of the surface of the part.Finally,the difference between the obtained data and the standard data is compared,and the difference is compared with the set threshold to determine whether there are defects on the surface of the part,thereby reducing the loss caused by not detecting the defects in time.The main research contents of this thesis are the following aspects.(1)The relevant parameter requirements of the image acquisition system is determined by analyzing the detection conditions of the part three-dimensional surface detection system based on line laser scanning.And through the characteristics of the line laser projected by the line laser and the selection method of the CMOS camera,the hardware composition of the image acquisition system and the acquisition scheme of laser fringe images can be determined.According to the characteristics of the collected images with laser fringe patterns,the preprocessing process of laser fringe patterns is drawn up.And the image coordinates can be linked with the world coordinates through the principle of laser triangulation,which makes foundation for the subsequent three-dimensional reconstruction of objects and surface defect detection.(2)A method of extracting laser fringe patterns based on image difference is proposed.Firstly,the image of the part without laser irradiation can be obtained,and it will be used as the one to be differentiated.Then the line laser is allowed to scan the entire surface of the part at the suitable speed and the camera takes pictures.Then the difference between the image with laser irradiation and the image without laser irradiation can be calculated.Finally,the laser stripe pattern can be obtained.This method can extract the laser stripe pattern relatively quickly and avoid the interference of the background pattern.At the same time,the calculation speed is fast,which can meet the basic requirements of surface defect detection.(3)A method for extracting the centerline of stripes based on “Hough” line detection is proposed.Firstly,the computer will traverse each pixel of the image to find the brightest point of each column of pixels in the image and then perform “Hough” straight line detection to connect the brightest points detected into a straight line.This straight line is the center line of the laser stripe.(4)According to the image acquisition system and other hardware used in this thesis and the developed image preprocessing,laser stripe centerline extraction and coordinate conversion algorithm and other software,a three-dimensional surface defect detection method based on line laser scanning is proposed.At the same time,an experimental platform for the threedimensional surface defect detection system of parts based on line laser scanning corresponding to the detection method is built.The feasibility of the detection method is verified by analyzing the detection results of the detection system.