摄像机标定方法的建模与仿真研究

摘要

The accuracy of remote sensing image fusion problem. Image registration techniques has been widely used in remote sensing images, and other fields, for the traditional image registration algorithm and low efficiency and lack of precision in order to improve the accuracy of remote sensing image fusion, a remote sensing image based on minimum spanning tree registration algorithm, the minimum spanning tree algorithm and the optimization is applied to image fusion process, the algorithm first extracts uniform set of sub-sampling points, and on this basis construct minimum spanning tree, and then use the minimum spanning tree to estimate the entropy, the final image between the edge of the gradient information into the integration framework. Algorithm effectively overcomes the shortcomings of the traditional image fusion algorithms, simulation results show that remote sensing image fusion with traditional algorithm , this algorithm effectively improves the accuracy of image registration to verify the feasibility of the method is an effective The image registration algorithm.%研究摄像机定位优化控制问题,摄像机镜头存在多种非线性畸变,对标定路径和图像质量产生一定的影响,难以采用精确数学模型来描述,针对传统方法的摄像机标定准确率低.为了提高摄像机的标定准确率,利用LSSVM较好的处理非线性预测能力,建立一种PSO-LSSVM的摄像机标定模型.将摄像头采集到的图像坐标作为输入,将世界坐标作为输出,通过采用LSSVM精确逼近输入与输出的复杂非线性关系,采用PSO寻找LSSVM最优参数,提高标定准确率.通过标定模型进行对比实验,实验结果表明,PSO - LSSVM不仅加快标定速度,且提高了摄像机标定的准确率.