Research On 3D Quality Inspection Method For Complex Additive Manufacturing Parts Based On Structured Light Sensor

Structured light measurement technology is an important technology in the field of 3D quality inspection.It has the advantages of non-contact,high speed,high precision and superior real-time measurement performance.It can calculate the high-precision 3D point cloud model within the measurement range.Widely used in face recognition,workpiece quality detection,3D shape measurement,motion tracking,pattern recognition,reverse engineering and other fields.However,when performing three-dimensional quality inspection on complex additive manufacturing parts,the following problems occur: the projection range of the projector is rectangular,and the additive manufacturing parts are complex and irregular,resulting in a point cloud generated by the measurement containing a large amount of redundant information;The optical measurement system has a field of view limitation,and the generated incomplete point cloud needs to be registered,and the registration residual is still an obstacle to generating an accurate,large-area three-dimensional point cloud model.The main work of this paper is as follows:1)Additive manufacturing parts detection meansDevelopment of a measurement system for additive manufacturing based on structured light.The structural light three-dimensional measurement system used in the whole text was designed and constructed,the mathematical model of the structured light measurement system was established,and the structured light system was calibrated.The structured light measurement system is a necessary tool for obtaining subsequent image information and point cloud information,and has specific requirements for system accuracy.After the plane measurement experiment,the accuracy of the structured light measurement system is verified to meet the required accuracy requirements.The work of this chapter lays the hardware foundation for the subsequent method research.2)High-efficiency additive manufacturing point cloud generation methodAiming at the problem of redundant information generated by structured light measurement,a point cloud extraction method based on two-dimensional image saliency is proposed.The difference between the color values of the additive manufacturing parts and their surrounding environment on the two-dimensional image is used to extract the additive manufacturing parts.The pixel coordinates of the salient region are used to generate a point cloud for this part of the pixel,which reduces the amount of calculation and data generated during the generation of the point cloud,and improves the pertinence,purpose,and measurement efficiency.3)Research on point cloud model of high precision and complex additive manufacturing partsAiming at the problem that the generated point cloud is incomplete and the point cloud registration residual is large,the point cloud registration optimization method based on plane loop is designed.This method utilizes the planar feature pair of the substrate in additive manufacturing.After registration,the point cloud is used for plane detection to obtain plane feature data,and the point cloud is closed-loop optimized according to the data,which reduces the registration residual and improves the accuracy of the generated point cloud model.In the experimental part,the quality and effect of the two methods and the combined application of the two methods were verified by testing the physical objects.