Feature point identification method aiming at pose measuring of cooperation marker

Abstract

The invention discloses a feature point identification method aiming at pose measuring of a cooperation marker. The method comprises the steps of: (1) threshold value division and judgment of connected regions; (2) calculating shape factors of the connected regions and solving a round connected region; (3) solving a square identification box according to the diameter of the round connected region, dividing four quadrants in the identification box, and calculating quantity of the connected regions, the shape factors, diameters and distance to the center of the identification box in each quadrant. The cooperation marker and respective feature points in an image are identified. The method solves the problem of identification and positioning of image feature points under the condition that the feature points are partially shielded and interfered in the visual pose measuring process of rendezvous and docking of a spacecraft, and an identification result is accurate and reliable.

Description

technical field [0001] The invention relates to a feature point recognition method, in particular to a feature point recognition method for pose measurement of cooperative markers. Background technique [0002] Vision-based target pose (position, attitude) measurement is the frontier direction of key research in the field of photoelectric precision measurement technology, and plays a pivotal role in space detection, industrial manufacturing, robotics and other fields. Especially in the field of space, accurate measurement of the pose of space objects is an important prerequisite for the success of space tasks (rendezvous and docking, target capture, on-orbit assembly and maintenance, etc.). [0003] Vision-based pose measurement is divided into cooperative target measurement and non-cooperative target measurement. Compared with non-cooperative target measurement, cooperative target measurement has the advantages of high measurement accuracy, strong reliability, strong anti-i...

AI Technical Summary

Problems solved by technology

This method establishes the recognition frame according to the proportional relationship between the size of the feature point and the size of the target, which suppresses the interference of the complex background outside the image recognition frame on the recognition; divides into 4 quadrants, counts the total number of recognized targets in the 4 quadrants, and solves the problem that one feature point of the target is blocked. How to stably recognize the target under occlusion; judge the size, shape, and distance to the center of the target in the four quadrants respectively, and solve the problem of accurate recognition of feature points of the target under displacement, rotation, and scale changes

Images