Target pose measurement method based on calibration object

Abstract

The invention discloses a target pose measurement method based on a calibration object. The method comprises the steps: setting the calibration object on the surface of a measured object, collecting an image of the measured object through a camera, recognizing the calibration object through an image processing technology, eliminating the wrong recognition condition according to the features of thedesigned calibration object, and collecting the feature points on the calibration object; finally, obtaining the pose information of the calibration object through the corresponding relationship of the 2D3D points, and replacing the pose information of the target with the pose information of the calibration object. The target pose measuring method based on the calibration object is adopted, the pose information of the target can be accurately measured, and great convenience can be provided for the mechanical arm to grab the target object.

Description

technical field [0001] The invention belongs to the field of computer vision, in particular to a target pose measurement method based on calibration objects. Background technique [0002] Computer vision refers to using a camera instead of the human eye to observe the target, combined with image processing technology, to extract the required key information from the image captured by the camera. Today, computer vision is often used to control the motion of robotic arms. If the robotic arm wants to grab an object, it needs to know the relevant information of the target object, such as the distance of the target object from the claw at the end of the robotic arm and the spatial position and orientation information of the target object. With this information, the robotic arm can plan the movement path and grasping angle, and complete the target object and grasping. [0003] In the actual application process, it is often difficult to only recognize the target object because th...

AI Technical Summary

Problems solved by technology

[0003] In the actual application process, it is often difficult to only recognize the target object because the target object is in a complex background. Due to environmental factors such as light, the camera may not be able to recognize the characteristic information of the target object or misidentify it.

This causes considerable difficulty in planning the motion path of the robot arm and selecting the grasping angle.

In addition, different target objects may have differences in shape. At this time, some parameter settings will be different in the process of target feature point recognition. It is extremely inconvenient to manually adjust the parameters for each target object.

Images