An improved APAP image matching algorithm

doi:10.13474/j.cnki.11-2246.2020.0092

Abstract

Abstract: Aiming at the problems of image ghosting, perspective distortion and long time consumption when UAV image matching, this paper proposes an improved APAP algorithm. First, this algorithm uses SIFT algorithm to select feature points and removes the mismatch point by improving RANSAC algorithm. Then, according to the APAP algorithm, the image is meshed, the homography matrix of each mesh is obtained, and the homography matrix is linearized. Finally, the image matching is performed according to the linearized homography matrix, and the homography matrix is Linearization not only weakens the ghosting phenomenon caused by image matching, but also reduces the perspective distortion of non-overlapping regions. The experimental results show that the proposed method is effective in matching efficiency and matching effect.

Key words: UAV image, APAP algorithm, RANSAC algorithm, homography, linear

CLC Number:

P237

ZHANG Dongmei, LU Xiaoping, MIAO Peiji, ZHOU Yushi, MA Liangting. An improved APAP image matching algorithm[J]. Bulletin of Surveying and Mapping, 2020, 0(3): 123-128.

References

[1] 魏嘉磊.基于SIFT算法的无人机高分影像几何配准研究[J].北京测绘,2019,33(1):49-53.
[2] 李德仁,李明.无人机遥感系统的研究进展与应用前景[J].武汉大学学报(信息科学版),2014,39(5):505-513.
[3] 陈晓龙,陈显龙,彭瑶瑶.无人机影像处理关键技术研究[J].北京测绘,2016,30(3):24-27.
[4] LOWE D G. Distinctive image features from scale-invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[5] BAY H, ESS A,TUYTELAARS T, et al. Speed-up robust features[J].Computer Vision and Image Understanding,2008,110(3):404-417.
[6] CALONDER M, LEPETIT V,STRECHA C,et al.BRIEF:binary robust independent elementary features[C]//European Conference on Computer Vision. [S.l.]: Springer-Verlag, 2010:778-792.
[7] CALONDER M, LEPETIT V,OZUYSAL M,et al.BRIEF:computing a local binary descriptor very fast [J].IEEE Transactions on Pattern Analysis and Machine Interlligence,2012,34(7):1281-1298.
[8] ZARAGOZA J,CHIN T,TRAN Q,et al.As-projective-as-possible image stitching with moving DLT[J].IEEE Transactions on Pattern Analysis & Machine Intelligence, 2014, 36(7):1285-1298.
[9] LI Nan,XU Yifang,WANG Chao. Quasi-homography warps in image stitching[J]. IEEE Transactions on Multimedia,2018,20(6):1365-1375.
[10] 何川,周军.具有直线结构保护的网格化图像拼接[J].中国图象图形学报,2018,23(7):973-983.
[11] 薛佳乐,赵萌,张哲,等.针对大视差图像拼接的显性子平面配准[J].中国图象图形学报,2018,23(3):323-332.
[12] XIANG T Z,XIA G S, BAI X, et al.Image stitching by line-guided local warping with global similarity constraint[J].Pattern Recognition,2018,83:481-497.
[13] FISCHLER M A, BOLLES R C. Random sample consensus: a paradign for model fitting with applications to image analysis and automated cartography[J]. Comm. of the ACM, 1981, 24(6):381-395.
[14] ZARAGOZA J, CHIN T J, BROWN M S, et al. Asprojective-as-possible image stitching with moving DLT[C]//2013 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), [S.l.]: IEEE, 2013.
[15] 王可,贾松敏,李秀智.基于全概率更新的改进RANSAC算法[J].控制与决策, 2017,32(3):427-434.