基于RANSAC和互信息的图像配准算法及其硬件实现

An Image Registration Algorithm Based on RANSAC and Mutual Information and Its Hardware Implementation

图像融合时,两路异源视频输入往往会因为光学系统和空间位置的差异而造成两路图像存在视差,进而造成图像融合时出现虚边、伪影问题,因此在图像融合前需要进行图像配准工作。针对当前异源图像配准存在的实时性差导致无法在实际工程中应用的问题,提出了一种基于RANSAC和互信息的图像配准算法,通过将不同维度的配准运算划分到初始配准和实时配准中,同时引入Powell优化算法加速搜索,在不降低配准精度的同时保持了较高的实时性。最终用所提算法进行了基于Zynq的嵌入式硬件移植和算法加速,实现了高精度、高实时性的嵌入式图像配准系统。

In the process of image fusion,two different-source video inputs often cause parallax between the two images due to the difference in optical system and spatial position,which in turn causes the problem of blurred edges and artifacts during image fusion.Therefore,image registration is required before image fusion.Regarding the problem that the current heterogeneous image registration has poor real-time performance, which makes it impossible to be applied in practical engineering,an image registration algorithm based on RANSAC and mutual information is proposed,which divides the registration operations of different dimensions into the initial registration and real-time registration,in the meantime the POWELL optimization algorithm is introduced to speed up the search,which maintains a high real-time performance without reducing the registration accuracy.Finally,the proposed algorithm is transplanted to embedded hardware and accelerated based on Zynq,and an embedded image registration system with high accuracy and real-time performance is realized.