Abstract:In this study, a monocular vision measurement method based on AKAZE feature detection and the PnP algorithm is used to calculate the relative attitude of a camera and quickly and accurately determine the attitude relationship between two objects in the space. Specifically, the template image of the cooperative target is collected, and the pixel coordinates of the four feature points attached to the cooperative target are extracted. Then, the template image and the image to be measured are matched by key points of AKAZE, and the mapping matrix is calculated. After that, the pixel coordinates of the four feature points in the image to be measured are obtained through the mapping matrix. Finally, given the size information of the cooperative target, the PnP problem based on the four coplanar feature points is solved, and the relative positions of the camera and the cooperative target are calculated. The experimental analysis shows that the real-time image camera attitude calculated by this method is close to the real result, which verifies the effectiveness of this method.