论文部分内容阅读
针对具有视角差异的无人机低空航拍遥感图像配准存在的误匹配、低效率等问题,提出了一种基于粒子群优化和互信息融合的配准优化算法.该算法通过对图像的变换参数的仿射采样来模拟建筑物在多视角遥感图像中的变形.在此基础上,将图像配准问题转换为仿射变换的优化问题,以粒子群为工具,研究了配准参数的搜索空间和适应度函数的合理设定.对4对无人机低空航拍遥感图像对进行了实验.结果表明:提出的算法能够实现具有较大视角变化的同时相遥感图像的配准,且比穷举互信息搜索法速度更快,比单纯型法的精度更高,有效地提高了算法对视角变化的鲁棒性.
Aiming at the mismatch and inefficiency of low altitude aerial image registration of unmanned aerial vehicles with different viewing angles, a registration optimization algorithm based on particle swarm optimization and mutual information fusion is proposed. The algorithm, , The author simulates the deformation of buildings in multi-view remote sensing images.On the basis of this, the problem of image registration is transformed into the optimization problem of affine transformation. The particle swarm optimization is used to study the search space of registration parameters And fitness function, the experiments were carried out on four pairs of UAV low-altitude aerial remote sensing images.The results show that the proposed algorithm can realize the simultaneous remote sensing image registration with larger viewing angle, The mutual information search method is faster and more accurate than the simplex method, which effectively improves the robustness of the algorithm to the change of perspective.