论文部分内容阅读
针对无人机(UAV)自主起飞与着陆机器视觉导引系统中摄像机的标定问题,讨论了标定摄像机内参数的重要意义,选定了包含畸变因素的摄像机透视投影模型。采用改进两步法进行摄像机内参数标定,结果中主要参数相对误差均小于1%,位置误差小于0.2像素,证明了所选择模型和标定方法的合理性。分析并验证了影响标定精度的因素,包括模型选择、图像拍摄质量和标定靶精度。针对机器视觉导引的应用条件,基本解决了准确、快速标定摄像机内参数的实际问题。
In order to solve the problem of camera calibration in UAV autonomous takeoff and landing machine vision guidance system, the importance of calibrating camera parameters is discussed. The camera perspective projection model with distortion factors is selected. The improved two-step method was used to calibrate the camera parameters. The relative error of the main parameters was less than 1% and the position error was less than 0.2 pixels, which proves the rationality of the selected model and calibration method. The factors affecting calibration accuracy were analyzed and verified, including model selection, image quality and calibration target accuracy. The application of machine vision guidance, the basic solution to the accurate and rapid calibration of the camera parameters of the actual problem.