Workpiece pose estimation method based on component model expression

Abstract

The invention discloses a workpiece pose estimation method based on component model expression, and the method comprises the following steps: carrying out the data enhancement of to-be-detected data,and obtaining a processed image; carrying out feature extraction on the image through a convolutional neural network to obtain feature maps of three scales; expressing a network structure of each feature map through a component model to obtain corresponding scores and model response maps of the whole and the components; performing parameter optimization on all the model response graphs to obtain acomprehensive response graph and key points; calculating n 3D points in the space through an EPnP algorithm to be matched with 2D points in the image, and then obtaining the corresponding pose of thecamera. According to the method, information marking only needs to be carried out on the whole workpiece, component information of the workpiece serves as recessive features, effective components areautomatically found through the neural network to carry out component marking, and the method has the advantages of being rapid and efficient in performance and capable of accurately calculating the6D pose of weak-texture objects such as the workpiece in real time.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a workpiece pose estimation method based on component model expression. Background technique [0002] In practical applications, it is very critical to obtain the 3D information of the target object. For example, in the task of manipulator grasping manipulation, recognizing the 6D pose (3D positioning and 3D orientation) of an object provides useful information for grasping and motion planning. The difficulty of estimating the pose of a workpiece is that the workpiece is a weakly textured object, which is easily affected by light and reflection, so that the texture reflected from the 2D image may not be the real texture of the 3D object surface. Moreover, when the resolution of the image changes, the calculated texture may have a large deviation, and the feature extraction algorithm is not easy to identify. In the actual process of pose estimation, there are still objective facto...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a workpiece pose estimation method based on component model expression, which solves the proble

Images