A Method of Solving the Specular Parameters of Cone-mirror Catadioptric Camera Using a Straight Line

Abstract

The invention relates to a method for solving a mirror surface parameter of a conical mirror surface refraction and reflection camera by utilizing a straight line. The method comprises the steps of firstly, extracting pixel coordinates of image points from an image by utilizing a function in Matlab; secondly, calculating a coefficient and a cone apex angle 2 theta of a line image according to the pixel coordinates of the image points, wherein theta is an included angle between any generatrix of a conical mirror surface and a rotational symmetric axis of the conical mirror surface; and finally, calculating Plucker coordinates of the spatial straight line and a distance from a vertex of the conical mirror surface to an optical center of the camera, namely, the mirror surface parameter xi. The method does not have a requirement on the physical size of the spatial straight line, so that coordinates of the spatial straight line in a world coordinate system do not need to be known.

Description

technical field [0001] The invention belongs to the field of computer vision, and relates to a method for calculating mirror parameters of a conical mirror catadioptric camera by using coefficients of straight lines and line images. Background technique [0002] Computer vision is the use of computer software and hardware technology to realize and simulate the functions of human vision to perceive, recognize and understand the objective world scene. Its central task is to understand the shape, position, and motion of the image, and use the information contained in the two-dimensional image of the three-dimensional object to obtain geometric information such as the spatial position and shape of the three-dimensional object to identify the three-dimensional object. In the field of computer vision, ordinary cameras have a small range, limited imaging angles, less information and dead angles, so they cannot imitate the human eye well. With the continuous progress of science and...

AI Technical Summary

Problems solved by technology

The literature "Real-time omnidirectional image sensor for vision-guided navigation", (YagiY, Kawato S, Tsuji S., IEEE Transactions on Robotics and Automation, 1994, 10(1): 11-22) analyzes and discusses the axicon mirror surface The projection relationship under the reflective system, but did not analyze and discuss the trajectory of the cone-mirror viewpoint combined with the symmetry properties of the cone-mirror surface, the literature "Line-images in cone mirrorcatadioptric systems", (Bermudez-Cameo J, Lopez-Nicolas G, Guerrero J. J., Proceedings of the 22 nd International Conference on Pattern Recognition, 2014, 1:2083-2088) obtained the line image equation, and used the line image equation to obtain the mirror angle, but did not give a method for calculating the distance from the mirror vertex to the optical center of the camera. The literature "Unitary torus model for conical mirror based catadioptric system”, (Lopez-Nicolas G, Sagues C, Computer Vision and Image Understanding, 2014, 126:67-79) proposed a unit ring model based on the unit ball model

Images