一种针对非凸颗粒的双平面镜摄像机内参标定改进算法

amera calibration is one of the foundations of photogrammetry. The internal parameters of the camera is determined by the intrinsic properties of the camera, and has nothing to do with the camera pose. This paper introduces a camera internal calibration algorithm based on double-plane mirror system. And aim at the non-convex particle characteristics of experiment objects,we propose an improved algorithm for the calculation of the camera intrinsic parameters. Through the improvements of the epipole location algorithm, this method can avoid the deviation appeared in the process of calculating the epipoles, in addition to the deviation for the calculation of internal. Simulation and real experiment results show that the method can quickly and efficiently calculate the camera internal parameters,and the measurement experiment of ballast shape parameters,also lay the foundation for the subsequent three-dimensional reconstruction of ballast and the discrete element method in analyzing.