A Six-Degree-of-Freedom Measurement Method Based on Swinging Multi-Camera Tracking

At present, for six-degrees-of-freedom measurement of industrial targets, the visual measurement technology has difficulty in taking account of the measurement efficiency, accuracy, and range. Therefore, this paper proposes a six-degree-of-freedom measurement method based on swinging multi-camera tracking to achieve high efficiency, high accuracy, and large range pose measurement by combining multiple images in different angles of same position to form redundant constraints for large-size coordinate calculation. First, a camera rotation-swing motion model is proposed to realize fast estimation of the camera pose, and use it as a priori information to perform image matching and space resection to obtain accurate camera pose, to perform space forward intersection. Then, a pose estimation method based on image point adjustment is proposed, which directly uses the image point information of the object's moving as the observation value to get its optimal pose estimation. Experimental results show that the measurement efficiency is increased by 4 times, the maximum error of single-point accuracy is no more than 0.2 mm, and the maximum error of pose accuracy is no more than 0.043 degrees, which proves that the proposed method can effectively measure the six-degree-of-freedom posture and balance the measurement efficiency, accuracy, and range.