一种基于颜色域的低空无人机图像拼接算法

Low-altitude UAV Image Stitching Approach Based on the Color Domain

拼接图像重合处的痕迹是衡量图像拼接质量好坏的标准之一。尽管低空无人机图像拼接技术已经研究多年,但自动拼接图像中的错位、重影、锯齿仍然存在。通过研究低空无人机遥感平台图像拼接存在的问题,提出一种新的无人机图像无缝拼接方法,并通过实验进行了拼接结果好坏分析与评价。具体而言,新方法首先通过SURF匹配方法使待拼接无人机图像进行配准,将具有重叠区域的两幅图像配准并转换到同一坐标系中;接着基于Boykov提出的图割能量优化理论,研究如何利用待拼接图像构建图结构,并引入图像的全局光流场来定义图像像素颜色域的能量代价函数;最后通过搜索最优拼接线实现无人机图像无缝拼接。

The traces at the overlapping area of stitching image are one of the criteria for measuring the quality of image stitching.Although UAV image stitching technology has been studied for many years,there are still problems in stitched image like misalignment,ghosting and aliasing.Therefore,we proposed a new UAV image stitching seamless fusion approach by studying the existing problems in the image stitching of low-altitude UAV remote sensing platform,analyzed and evaluated the experimental results of new approach.Specifically,we performed image registration on UAV image pairs by SURF matching method,so as to align and transform the two images with overlapping areas into the same coordinate system at first.Then,we studied how to construct a graph structure using images to be stitched based on the graph cuts energy optimization theory proposed by Boykov,and introduced the global optical flow field of image to define the energy cost function in the color domain of image pixels.Finally,we realized UAV image seamless stitching by searching for the optimal seam-line.