Research On 3D Fast Measurement And Classification System Based On Multi Line Laser

Three dimensional measurement technology is widely used in defects detection,3D printing and other fields.The field of industrial defect detection,mostly by measuring the object three-dimensional contour information,find out the defect position of objects with standard size matching method,because the testing environment is complex and changeable,often require three-dimensional detection method and apparatus of good stability,high precision,simple structure and can be adjusted in real time according to the test environment and the shape of the workpiece.The 3D measurement techniques are commonly used in stereo matching,structure light difference method,laser scanning method,the laser scanning line method because of its simple structure,high precision is widely used,but this method needs to control the motor to scan the entire surface of the object,seriously affect the measurement time,and the camera frame and the motor speed can be effective matching.In order to solve the shortcomings of line scanning,it is becoming more and more important to quickly measure the 3D information technology.Rapid 3D measurement technology has many existing products,such as the Microsoft Corp Kinect depth camera,portable 3D scanner,but the precision of equipment(>3mm)is low,and the high cost of hardware,software development is not easy,and can not meet the requirements of industrial inspection.Therefore,the focus of this paper is to achieve a three-dimensional rapid measurement and classification scheme to meet the requirements of industrial precision.The starting point of this paper is three dimensional rapid measurement and classification.Mainly from theoretical deduction,data modeling,matlab simulation analysis,environment construction,experimental measurement,matching correction,discrete fitting,three-dimensional classification and other ideas.The camera calibration by first,and then derive the linear structured light 3D measurement formula,mathematical modeling and simulation analysis of MATLAB factors affecting the resolution of the system;then the system hardware and software design and installation and debugging;image acquisition,optimization method,the fringe centerline extraction and eliminate the error matching,to obtain 3D results after the detection sensitivity of the system is analyzed;finally the 3D surface fitting results,and quantitative analysis of fitting results,and the KNN algorithm to classify the data based on 3D,and validate the classification accuracy.This paper finally set up the integrated software and hardware system of 3D measurement and classification,design and installation of the linear structure light device,write C++ language under the VS platform MFC programming interface based on optimized measurement algorithm,improve the measurement efficiency,design and experimental verification of 3D measurement sensitivity,3D fitting,3D the classification accuracy,the final completion of the experiment the measurement distance 600mm system sensitivity of 0.125mm/m(0.469mm/pixel),50mm can be resolved when the thickness of 0.10mm,surface fitting>0.82,fast measurement and classification system of three-dimensional classification accuracy of 97%.