A precision pose measurement technique based on multi-cooperative logo

Abstract

Abstract The use of image recognition technology to complete the precise pose measurement of objects has always been a difficulty in the field of computer vision. This paper proposes a precision pose measurement technology based on multi-cooperation logo, which uses OpenCV and Aruco to complete the generation of cooperative logo. Through the uPnP algorithm to complete the relative pose detection of the camera and the cooperation logo, and then the precise pose measurement of the multi-cooperation sign is completed by Gaussian filtering and iteration of the nearest neighbor. Compared with the traditional positioning method, it has higher accuracy and robustness. This technology has important reference value in robot positioning.