Object pose recognition method and device, visual processing equipment and readable storage medium

Abstract

The invention provides an object pose recognition method and device, visual processing equipment and a readable storage medium, and relates to the technical field of robot control. According to the method, after the color image and the depth image which are matched with the pixel content and collected by the grabbing robot for the target object are obtained, color normalization processing is carried out on the color image according to the color calibration parameter to obtain the first target image, and depth normalization compensation processing is carried out on the depth image to obtain the second target image; performing image channel splicing processing on the first target image and the second target image to obtain a corresponding target object image, and calling a pre-stored pose estimation model to directly recognize object pose information of a target object from the target object image; therefore, the object pose information of the target object is directly recognized after the color image and the depth image of the target object are fused, so that the pose recognition real-time performance is enhanced, and the scheme application range is effectively expanded while the pose recognition efficiency is ensured.

Description

technical field [0001] The present application relates to the technical field of robot control, in particular, to a method and device for recognizing object poses, a visual processing device, and a readable storage medium. Background technique [0002] With the continuous development of science and technology, robot technology has been widely valued by all walks of life because of its great research value and application value. In the actual application of robot technology, the robot's grasping action is one of the routine actions that intelligent robots can perform. In order to realize the robot's grasping action, it is usually necessary to first obtain the pose of the target object, and then estimate the effect on the The location of the grasping point on the target object, and then plan the trajectory of the robot's grasping action for execution. Therefore, to realize the grasping action of the robot, it is a necessary operation to recognize the specific pose of the targ...

AI Technical Summary

Problems solved by technology

Therefore, the object pose recognition of this general scheme takes a long time in specific implementation, and the overall recognition is not real-time

At the same time, the existing object positioning algorithm used to realize the function of the object positioning module often cannot achieve the balance between the scope of application of the scheme and the recognition efficiency, and the existing object pose estimation algorithm used to realize the function of the object pose estimation module is often unable to do so. To the balance between the scope of application of the scheme and the identification efficiency

Images