基于改进SIFT的无人机航拍图像快速匹配
|
摘要
针对无人机航拍图像快速匹配问题,传统的SIFT算法复杂度高,处理时间长,为了满足实时性的要求,提出一种改进的SIFT算法.首先将特征点的矩形区域改为圆形区域,对描述子进行降维,然后借助绝对距离和余弦相似度进行双重匹配,最后再通过RANSAC算法剔除误匹配点.实验证明,改进的SIFT算法在尺度缩放、旋转、光照等情况下均有良好的匹配效果,与原算法相比,在保证匹配精度的同时很好的提高了匹配速率,验证了算法的实时性、有效性.
To realize fast matching of UAV aerial images, the traditional SIFT algorithm has high complexity and long processing time, which cannot meet the requirements of real-time. A improved SIFT algorithm is proposed. The improved SIFT algorithm changes the rectangular area of the feature point into circular area and reduces the Dimensions of descriptor at first, and then do the match by means of the absolute distance and cosine similarity. Finally, the RANSAC algorithm is used to eliminate the false matching points. The experimental results show that the improved SIFT algorithm has a good matching result on the conditions of image zooming, rotating, lighting. Compared with the original algorithm, the matching speed is improved while the matching accuracy is ensured, which verifies the real-time and effectiveness of the improved algorithm.
To realize fast matching of UAV aerial images, the traditional SIFT algorithm has high complexity and long processing time, which cannot meet the requirements of real-time. A improved SIFT algorithm is proposed. The improved SIFT algorithm changes the rectangular area of the feature point into circular area and reduces the Dimensions of descriptor at first, and then do the match by means of the absolute distance and cosine similarity. Finally, the RANSAC algorithm is used to eliminate the false matching points. The experimental results show that the improved SIFT algorithm has a good matching result on the conditions of image zooming, rotating, lighting. Compared with the original algorithm, the matching speed is improved while the matching accuracy is ensured, which verifies the real-time and effectiveness of the improved algorithm.
引文
1 Arican Z,Frossard P.Scale-invariant features and polar descriptors in omnidirectional imaging[J].IEEE Transactions on Image Processing,2012,21(5):2 412-2 423.
2 Lowe D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
3 Soyel H,Demirel H.Improved SIFT matching for pose robust facial expression recognition[C/OL]//IEEE:Automatic Face and Gesture Recognition and Workshops,Santa Barbara,CA,USA,March 21-25,2011.[2017-08-27].https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5771463.
4朱进,丁亚洲,肖雄武,等.基于SIFT改进算法的大幅面无人机影像特征匹配方法[J].计算机应用研究,2015,32(10):3 156-3 159.
5 Ke Y,Sukthankar R.PCA-SIFT:A more distinctive representation for local image descriptors[C/OL]//Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition,Washington,DC,USA,June 27-July 2,2004.[2017-08-27].https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1315206.
6 Mikolajczyk K,Schmid C.A performance evaluation of local descriptors[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1 615-1 630.
7 Bay H,Ess A,Tuytelaars T,et al.Speeded-up robust features(SURF)[J].Computer Vision and Image Understanding,2008,110(3):346-359.
8张春美,龚志辉,孙雷.改进SIFT特征在图像匹配中的应用[J].计算机工程与应用,2008,44(2):95-97.
9陈抒2),李勃,董蓉,等.Contourlet-SIFT特征匹配算法[J].电子与信息学报,2013,35(5):1 215-1 221.
10曾峦,顾大龙.一种基于扇形区域分割的SIFT特征描述符[J].自动化学报,2012,38(9):1 513-1 519.
2 Lowe D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
3 Soyel H,Demirel H.Improved SIFT matching for pose robust facial expression recognition[C/OL]//IEEE:Automatic Face and Gesture Recognition and Workshops,Santa Barbara,CA,USA,March 21-25,2011.[2017-08-27].https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5771463.
4朱进,丁亚洲,肖雄武,等.基于SIFT改进算法的大幅面无人机影像特征匹配方法[J].计算机应用研究,2015,32(10):3 156-3 159.
5 Ke Y,Sukthankar R.PCA-SIFT:A more distinctive representation for local image descriptors[C/OL]//Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition,Washington,DC,USA,June 27-July 2,2004.[2017-08-27].https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1315206.
6 Mikolajczyk K,Schmid C.A performance evaluation of local descriptors[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1 615-1 630.
7 Bay H,Ess A,Tuytelaars T,et al.Speeded-up robust features(SURF)[J].Computer Vision and Image Understanding,2008,110(3):346-359.
8张春美,龚志辉,孙雷.改进SIFT特征在图像匹配中的应用[J].计算机工程与应用,2008,44(2):95-97.
9陈抒2),李勃,董蓉,等.Contourlet-SIFT特征匹配算法[J].电子与信息学报,2013,35(5):1 215-1 221.
10曾峦,顾大龙.一种基于扇形区域分割的SIFT特征描述符[J].自动化学报,2012,38(9):1 513-1 519.