A method, system and device for identifying and marking scattered workpieces

Abstract

The present application relates to a method, system and device for identifying and marking scattered workpieces, wherein the method includes: collecting the actual point cloud image and the workpiece pose matrix of the current workpiece in the batch to be collected through the camera, and according to the internal parameter matrix and external parameter matrix of the camera, The matrix constructs a virtual camera model, calculates the template point cloud projection under the virtual camera model according to the workpiece pose matrix of the current workpiece and the template point cloud data of the template workpiece, calculates the template depth map according to the template point cloud projection, and compares the actual points Cloud map and template depth map to determine whether the current workpiece matches the template workpiece. If they match, mark the current workpiece and repeat the above steps to complete the labeling of all workpieces in the batch to be collected. Through this application, the identification and labeling of workpieces is solved. Due to the problems of low efficiency and low accuracy, the matching of depth map information based on the workpiece template is realized, the time required for workpiece identification and labeling is shortened, and the accuracy of identification and labeling is improved.

Description

technical field [0001] The present application relates to the field of computer vision, and in particular, to a method, system and device for identifying and labeling scattered workpieces. Background technique [0002] At present, in the process of robot automatic assembly, the identification and annotation of the workpiece and the determination of the three-dimensional point cloud pose of the workpiece are the prerequisites for the robot to perform related operations such as clamping. When the robot is equipped with a camera to capture a frame of workpiece image, it is usually necessary to manually mark all the workpieces captured on the image one by one, or use a pixel recognition algorithm to identify the complete workpiece contained in the image. [0003] The above labeling methods all have certain problems: the manual labeling method requires manual observation, and its labeling speed is slow. Although the pixel recognition method can quickly identify the complete work...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application provides a method, system and device for identifying and marking scattered workpieces, so as to at least solve the problems of low efficiency and low accuracy in the identification and marking of workpieces in the related art

Images