论文部分内容阅读
针对MATLAB标定工具箱的标定精度与所拍图像数量成正比的问题,即拍摄照片数量越多标定精度越高,提出了一种基于粒子群算法的摄像机内参数优化方法,从而达到拍摄少量图片也可以有较好精度的效果。首先摄像机从不同角度拍摄4张和20张标定板图片,利用MATLAB标定工具箱分别求取它们的内参数。然后根据标定点的实际坐标和反投影坐标建立目标函数,再由粒子群算法对标定箱求取的内参数进行优化。实验结果对比表明:与MATLAB标定工具箱相比,此方法能够在一定程度上提高少量标定板图片的标定精度。
Aiming at the problem that the calibration accuracy of MATLAB calibration kit is proportional to the number of images taken, that is, the more the number of photos taken, the higher the calibration accuracy, a method based on particle swarm optimization is proposed to optimize the camera parameters so as to achieve a small number of pictures You can have better precision results. First of all, the camera took 4 pictures and 20 pictures of the calibration plate from different angles and used MATLAB Calibration Toolbox to find their internal parameters respectively. Then the objective function is established according to the actual coordinate and the back-projection coordinate of the calibration point, and then the internal parameters obtained from the calibration box are optimized by the particle swarm optimization algorithm. The comparison of experimental results shows that compared with MATLAB calibration kit, this method can improve the calibration accuracy of a small amount of calibration plate pictures to a certain extent.