Plane Loop Closure Based Point Cloud Registration Using Structured Light Sensor

In the field of manufacturing such as additive manufacturing and rapid prototyping, the establishment of a 3D model of workpieces has great significance for the quality inspection, modeling, subsequent research, and application of the workpiece. However, there are still some problems in incompleteness and poor reduction for the complete 3D model of the workpiece, which limits the efficiency and accuracy of workpiece 3D quality inspection. In addition, due to the limitation of the field of view in the 3D measurement system, the generated local point cloud cannot completely describe the 3D information of the workpiece. To solve these problems, this paper proposes a point cloud registration method based on the structured light sensor, which combines loop closure optimization through the planar features in the local point clouds. This method considers the planar features to constrain the point clouds, and the point clouds were re-registered to improve the accuracy of point cloud registration. The feasibility and effect of the method are verified in the simulation and experiment, and the method can be applied to the fields of quality inspection.