Crankshaft positioning identification method and device, storage medium and equipment

Abstract

The invention provides a crankshaft positioning identification method and device, a storage medium and equipment, and the method comprises the steps: obtaining an initial minimum bounding box of a crankshaft based on the point cloud data of a to-be-identified crankshaft, and extracting a first length of the initial minimum bounding box, the first length being the axial length of the initial minimum bounding box; judging whether the first length is smaller than a preset reference length or not, wherein the reference length is the axial length of the crankshaft; and if the first length is smaller than a preset reference length, extending along the initial minimum bounding box to search for missing point clouds, and obtaining a final minimum bounding box of the crankshaft. According to the crankshaft sorting method and device, the problem that crankshaft recognition is inaccurate due to the fact that the crankshaft eccentric position point cloud is broken in the sorting process is effectively solved, and the crankshaft sorting reliability is improved.

Description

technical field [0001] The present invention relates to the technical field of robots, in particular to a crankshaft positioning identification method, device, storage medium and equipment. Background technique [0002] In the manufacturing industry, material sorting is a very critical link. Material sorting is mainly to place various types of production materials in specified positions according to certain classification principles. In this work, the 3D visual positioning recognition method is used to guide the robot for material sorting, which helps to reduce the use of labor costs. For different materials, designing a suitable recognition algorithm is a key factor to improve the overall detection rate. [0003] figure 1 It is a physical picture of a crankshaft. Due to the particularity of the crankshaft structure, the area with a cylindrical block at the head is not on the same axis as other areas. Here is the eccentric position of the crankshaft. In the case of stacke...

AI Technical Summary

Problems solved by technology

[0003] figure 1 It is a physical picture of a crankshaft. Due to the particularity of the crankshaft structure, the area with a cylindrical block at the head is not on the same axis as other areas. Here is the eccentric position of the crankshaft. In the case of stacked crankshafts, in When using a 3D camera to scan the crankshaft to obtain point cloud data, the following will be obtained: figure 2 The point cloud image of the crankshaft shown in (a), composed of figure 2 (a) It can be seen that the point cloud data is disconnected at the eccentric position of the crankshaft. In this case, the initial minimum bounding box of the crankshaft is obtained as figure 2 As shown in (b), the minimum bounding box cannot completely surround the entire crankshaft, which brings certain errors to the complete identification of the crankshaft, which in turn causes the subsequent inability to identify the head position of the crankshaft, which brings great inconvenience to the sorting work of the crankshaft. Difficulty

Images