An Inner Surface Image Acquisition and Fusion Method of Narrow and Long Atypical Cavity Based on Multi-camera Parallel Moving Image Acquisition

When the electromagnetic rail launcher accelerates the projectile, the inner bore of the track is in a harsh environment such as high temperature, high pressure and high current, which will cause irreversible and cumulative structural damage on the surface of the inner bore. It is important to study the damage image of the inner cavity of the electromagnetic rail transmitter for evaluating the durability and life decay of the tube. In this paper, a new measuring structure is designed according to the special cavity structure of electromagnetic orbital transmitter with large aspect ratio and special shape. It can be equipped with a micro-multi-camera array to obtain high definition image of the inner cavity. The image detection system in the cavity of the electromagnetic track transmitter is composed of ARM master, micro-multi-camera combination module, laser ranging module and motion mechanism, and upper computer which is composed of wireless control and image data fusion. According to the characteristics of ring stitching and parallel fusion of multi-camera images in this paper, an optimal stitching algorithm is proposed, which mainly includes SIFT feature extraction, BBF feature matching, RANSAC screening optimization, image transformation mapping, in-out edge fusion and image stitching. The experimental results show that this method can obtain high resolution inner surface image and achieve better image Mosaic effect.