| The machine vision technology has been widely used in the field of measurement using un-contact method.The system designed in this paper is used to get the shape size,position relationship and the relationship between the various features.Some work as following was done:The system can obtain the 3D profile information of the work piece to be measured by the scanning method in the fixed direction,and the camera is triggered by the self-designed closed-loop feedback trigger circuit.The trigger system breaks the limit of the measured object and the camera must obtain the absolute distance in the fixed position.It can be moving in the scanning process until scanning is finished,the scanning speed is improved and the jitter is reduced.After camera calibration,the points coordinates in the linear structure light can be obtained from the camera coordinate system,the light plane equation is the key to complete the transformation.In the earlier work,the calibration is finished with the method of cross ratio invariance.In this paper,the calibration was completed using positional relationship between the light plane,the projection image plane and the light plane calibration,the method increases the amount of data involved in the plane fitting,the validation of the algorithm has higher accuracy and robustness.The coordinate system of the scanning process is integrated into the unified coordinate system,which is based on the translation vector of the local coordinate systems.After obtaining the 3D point cloud of the whole shape of object,feature point is obtained by the open source 3D point cloud library.The final measurement results show that the closed-loop feedback trigger system can complete the trigger at the expected position,the light plane calibration,the direction cosine calibration with high precision,PCL can complete the point cloud segmentation and feature matching,the final measurement results have a certain accuracy. |
PDF Full Text Request