BKR-SIFT: A High-Precise Matching Algorithm

详细信息    查看全文

• 关键词：SIFT descriptor ; BBF ; KL divergence similarity score ; RANSAC
• 刊名：Lecture Notes in Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：9864
• 期：1
• 页码：433-445
• 全文大小：3,068 KB
• 参考文献：1.Abdel-Hakim, A.E., Farag, A.A.: CSIFT: a SIFT descriptor with color invariant characteristics. In: 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 1978–1983 (2006)
  2.Kumar, N.A.M., Sathidevi, P.S.: Image match using Wavelet-colour SIFT features. In: 7th IEEE International Conference on Industrial and Information System, pp. 1–6 (2012)
  3.Yang, Z., Kurita, T.: Improvements to the descriptor of SIFT by BOF approaches. In: 2013 2nd IAPR Asian Conference on Pattern Recognition, pp. 95–99 (2013)
  4.Mikolajczyk, K., Schmid, C.: A performance evaluation of local descriptors. IEEE Trans. Pattern Anal. Mach. Intell. 27, 1615–1630 (2005)CrossRef
  5.Li, J., Wang, G.: An improved SIFT matching algorithm based on geometric similarity. In: 2015 5th International Conference on Electronics Information and Emergency Communication (ICEIEC), pp. 16–19 (2015)
  6.Zhu, W., Jiang, Y., Wang, M., Lai, C.H.: Weighted least squares based improved SIFT matching algorithm. In: 2012 5th International Congress on Image and Signal Processing (CISP), pp. 500–504 (2012)
  7.Fotouhi, M., Kasaei, S., Mirsadeghi, S.E., Faez, K.: BSIFT: boosting SIFT using principal component analysis. In: 2014 22nd Iranian Conference on Electrical Engineering (ICEE), pp. 1130–1135 (2014)
  8.Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. 60, 91–110 (2004)CrossRef
  9.Johnson, D.H., Sinanovic, S.: Symmetrizing the Kullback-Leibler distance. IEEE Trans. Inf. Theor. 18, 96–99 (2003)
  10.Xiao, P., Cai, N., Tang, B., Weng, S.W., Wang, H.: Efficient SIFT descriptor via color quantization. In: 2014 IEEE International Conference on Consumer Electronics, pp. 1–3 (2014)
  11.Mikolajczyk dataset for images information. http://​www.​robots.​ox.​ac.​uk/​~vgg/​research/​affine/​index.​html
  
• 作者单位：Jiancai Wu (22) (23)
  Shunyan Wang (22) (23)
  Wenchi Sun (22) (23)
  
  22. State Key Laboratory of Software Engineering, Wuhan University, Wuhan, China
  23. School of Computer Science and Technology, Wuhan University of Technology, Wuhan, China
  
• 丛书名：Internet and Distributed Computing Systems
• ISBN：978-3-319-45940-0
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349
• 卷排序：9864

文摘

Scaled Invariant Feature Transform (SIFT) is the state-of-the-art local image descriptor for its invariance to image translation, rotation, scaling, and change in illumination. However, its matching precision is not satisfactory in many situations. In this paper, we proposed a more precise matching algorithm—BKR-SIFT. We apply the Best-Bin-First (BBF) algorithm to achieve rough matching firstly. Then, the Kullback-Leibler (KL) divergence similarity score is used as the coarse pruning algorithm. Finally, we apply the Random Sample Consensus (RANSAC) algorithm to refine the matched features furtherly. Experimental results show that our proposed algorithm can reach a higher matching precision with approximately the same time compared to the SIFT.