[1]吴含前,李程超,谢珏.一种改进的A-KAZE算法在图像配准中的应用[J].东南大学学报(自然科学版),2017,47(4):667-672.[doi:10.3969/j.issn.1001-0505.2017.04.007]
 Wu Hanqian,Li Chengchao,Xie Jue.Application of improved A-KAZE algorithm in image registration[J].Journal of Southeast University (Natural Science Edition),2017,47(4):667-672.[doi:10.3969/j.issn.1001-0505.2017.04.007]
点击复制

一种改进的A-KAZE算法在图像配准中的应用()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
47
期数:
2017年第4期
页码:
667-672
栏目:
计算机科学与工程
出版日期:
2017-07-20

文章信息/Info

Title:
Application of improved A-KAZE algorithm in image registration
作者:
吴含前12李程超12谢珏3
1东南大学计算机科学与工程学院, 南京 210018; 2东南大学计算机网络和信息集成教育部重点实验室, 南京 210018; 3东南大学蒙纳士大学苏州联合研究生院, 苏州 215123
Author(s):
Wu Hanqian12 Li Chengchao12 Xie Jue3
1School of Computer Science and Engineering, Southeast University, Nanjing 210018, China
2Key Laboratory of Computer Network and Information Integration of Ministry of Education, Southeast University, Nanjing 210018, China
3Southeast University-Monash University Joint Graduate School, Suzhou 215123, China
关键词:
A-KAZE 非线性扩散滤波 FED KNN匹配 仿射变换
Keywords:
A-KAZE nonlinear diffusion filter fast explicit diffusion(FED) K-nearest neighbor matching affine transformation
分类号:
TP315.69
DOI:
10.3969/j.issn.1001-0505.2017.04.007
摘要:
针对现有图像配准过程中难以保持图像的局部精度和边缘细节的问题,在A-KAZE算法的基础上提出了一种改进的图像特征提取算法AKAZE-ILDB.该算法首先利用非线性扩散滤波方程构造图像金字塔,采用快速显示扩散(FED)求得数值解,得到具有亚像素精度的图像特征点坐标;然后利用改进的LDB(ILDB)描述子构造具有尺度和旋转不变性的图像特征向量,对特征向量采用汉明距离进行KNN匹配;最后基于仿射变换模型计算空间映射参数矩阵来实现图像配准.实验结果表明:在保持相同图像特征匹配正确率的情况下, AKAZE-ILDB算法比A-KAZE算法平均配准时间缩短了300 ms; 在配准精度方面,比A-KAZE算法提高了3.7%,比传统特征提取算法SURF匹配正确率提高了29%.
Abstract:
Aiming at the problem that local precision and edge details are difficult to preserve in the existing process of image registration, an improved image feature extraction algorithm AKAZE-ILDB(accelerated KAZE-improved local difference binary)is proposed based on the A-KAZE algorithm. First, this algorithm uses nonlinear diffusion filtering equation to construct the image pyramid. The numerical solution is obtained by the fast explicit diffusion(FED)method. The coordinates of the image feature points with subpixel precision are obtained. Then, the invariant image feature vectors are constructed by the improved LDB descriptor. The eigenvectors are matched by KNN(K-nearest neighbor)with Hamming distance. Finally, the spatial mapping parameter matrix is computed based on the affine transformation model to realize image registration. The experimental results show that in terms of registration efficiency, the AKAZE-ILDB algorithm reduces average registration time by 300 ms compared with the original A-KAZE algorithm in the condition of maintaining the same matching accuracy. Meanwhile, the matching accuracy of the same image feature is also improved by 3.7% higher than the A-KAZE algorithm and 29% higher than the traditional feature extraction algorithm SURF(speed up robust feature).

参考文献/References:

[1] Rohilla J, Bhatnagar M A. Image registration: A review[J]. International Journal for Innovative Research in Science & Technology, 2016,3(2):90-93.
[2] Nasrollahi K, Moeslund T B. Super-resolution: A comprehensive survey[J]. Machine Vision and Applications, 2014, 25(6):1423-1468. DOI:10.1007/s00138-014-0623-4.
[3] Tian J, Ma K K. A survey on super-resolution imaging[J]. Signal, Image and Video Processing, 2011, 5(3):329-342. DOI:10.1007/s11760-010-0204-6.
[4] Li Y, Wang S, Tian Q, et al. A survey of recent advances in visual feature detection[J]. Neurocomputing, 2015, 149:736-751. DOI:10.1016/j.neucom.2014.08.003.
[5] Sun Y B, Zhao L, Huang S D, et al. L2-SIFT: SIFT feature extraction and matching for large images in large-scale aerial photogrammetry[J]. ISPRS Journal of Photogrammetry & Remote Sensing, 2014, 91:1-16.
[6] Gui Y, Su A, Du J. Point-pattern matching method using SURF and shape context[J]. Optik—International Journal for Light and Electron Optics, 2013, 124(14): 1869-1873. DOI:10.1016/j.ijleo.2012.05.037.
[7] Alcantarilla P F, Bartoli A, Davison A J. KAZE features[C]//Computer Vision—ECCV 2012. Berlin: Springer, 2012:214-227. DOI:10.1007/978-3-642-33783-3_16.
[8] Weickert J, Grewenig S, Schroers C, et al. Cyclic schemes for PDE-based image analysis[J]. International Journal of Computer Vision, 2016, 118(3): 275-299.DOI:10.1007/s11263-015-0874-1.
[9] Alcantarilla P F, Jesú N, Bartoli A. Fast explicit diffusion for accelerated features in nonlinear scale spaces[C]//Electronic Proceedings of the British Machine Vision Conference. Bristol, UK, 2013:1-11.
[10] Grewenig S, Weickert J, Bruhn A. From box filtering to fast explicit diffusion[J]. Lecture Notes in Computer Science, 2010, 6376: 533-542. DOI:10.1007/978-3-642-15986-2_54.
[11] Yang X, Cheng K T. LDB: An ultra-fast feature for scalable augmented reality on mobile devices[C]//IEEE International Symposium on Mixed and Augmented Reality. Atlanta, USA, 2012:49-57.
[12] Tuytelaars T, Mikoljczyk K. Affine covariant features[EB/OL].(2008)[2016-03-03]. http://www.robots.ox.ac.uk/~vgg/research/affine.
[13] Mikolajczyk K, Schmid C. A performance evaluation of local descriptors[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(10): 1615-1630. DOI:10.1109/TPAMI.2005.188.

备注/Memo

备注/Memo:
收稿日期: 2016-11-13.
作者简介: 吴含前(1972—),男,博士,副教授, hanqian@seu.edu.cn.
基金项目: 国家高技术研究发展计划(863计划)资助项目(2015AA015904).
引用本文: 吴含前,李程超,谢珏.一种改进的A-KAZE算法在图像配准中的应用[J].东南大学学报(自然科学版),2017,47(4):667-672. DOI:10.3969/j.issn.1001-0505.2017.04.007.
更新日期/Last Update: 2017-07-20