Monocular vision pose estimation method based on auxiliary point geometric constraint

The invention discloses a monocular vision pose estimation method based on auxiliary point geometric constraint, which comprises the following steps: S1, target point extraction: shooting a target object containing a plurality of target points by using a single lens reflex camera to obtain a color picture, preprocessing the image to obtain a binary image, performing eight-field boundary tracking on the binarized image, calculating the area s and perimeter c of the contour, setting a constraint criterion according to the size and characteristics of the target point so as to obtain the contour of the target point, and carrying out center positioning on the target point; S2, pose calculation: enabling the part to adopt a pose estimation algorithm based on auxiliary point geometric constraintto finally obtain a target point coordinate Pic under a camera coordinate system; and S3, iterative optimization: performing iterative optimization on the Pic to obtain an optimized target point pose.According to the method,