基于半投影翘曲的无人机图像拼接方法

UAV image stitching method based on half-projection warping

为了解决无人机图像在拼接过程中出现的错位、重影等造成图像失真的问题,提出一种基于半投影翘曲的无人机图像拼接方法。首先估计图像的全局投影变换矩阵;然后根据待拼接图像的重叠情况,以及投影变换和相似性变换的连续性,求解出过渡变换矩阵和相似性变换矩阵,得到最终的半投影变换矩阵;最后构建图像重叠区域的差异矩阵,以此为基础获取重叠区域的差异性区域。使用分区融合策略,在差异性区域进行单采样,在其他区域进行距离加权融合。实验结果表明,该方法可以很好地拼接由于视角变化、地势起伏等造成视差的无人机图像,得到的拼接结果自然清晰,效果优于其他先进算法。该方法有效地解决了无人机图像拼接错位、重影的问题,在多项定量指标上表现良好。

To solve the problems of distortion and ghosting in the stitching of UAV images,this paper proposed a UAV image stitching method based on half-projection warping.Firstly,it estimated the global projection transformation matrix of the image.Then it solved the transition transformation matrix and the similarity transformation matrix according to the overlap of the images,the continuity of the projection transformation and the similarity transformation,and obtained the final semi-projection transformation matrix.Finally,it constructed the difference matrix of the overlapping region of the images and obtained the different regions of the overlapping region based on the difference matrix.Using the partition fusion strategy,it performed single sampling in the different regions and distance-weighted fusion in other regions.The experimental results show that this method can splice the images caused by parallax due to the change of perspective and relief,and the result of the splicing is natural and clear,and the effect is better than other advanced algorithms.The method effectively solves the problems of misalignment and double image stitching of UAV images and performs well in many quantitative indexes.