Abstract Image similarity measurement is crucial to many vision applications. A similarity measurement method based on adaptive weighting of similarity matrix is proposed in this paper. The image is firstly divided into the same-sized patches, and the convolutional neural networks are adopted to construct the descriptor of each patch. The patch similarities are calculated to constitute the similarity matrix. The probability of image pair coming from the same place is evaluated by analyzing the data distribution in similarity matrix. And the similarity weight of each unit is calculated based on the data difference. Ultimately, the overall image similarity is determined. The experimental results show that the proposed method is more robust than the existing ones in image retrieval. Moreover, it effectively solves the loop closure detection in simultaneous localization and mapping.
Fund:Supported by National Natural Science Foundation of China(No.41201390), Project of Science and Technology Innovation of Henan Province(No.142101510005)
About author:: 李 钦,男,1990年生,博士研究生,主要研究方向为计算机视觉、深度学习.E-mail:leequer120419@163.com. 游 雄,男,1962年生,博士,教授,主要研究方向为虚拟地理环境、地理信息系统.E-mail:youarexiong@163.com. 李 科(通讯作者),男,1977年生,博士,副教授,主要研究方向为图像处理、计算机视觉.E-mail:Like19771223@163.com. 汤 奋,男,1991年生,博士研究生,主要研究方向为地图设计.E-mail:Tangfen1204@163.com.
[1] 朱道广,郭志刚,赵永威.基于空间上下文加权词汇树的图像检索方法.模式识别与人工智能, 2013, 26(11): 1050-1056. (ZHU D G, GUO Z G, ZHAO Y W. Image Retrieval with Spatial Context Weighting Based Vocabulary Tree. Pattern Recognition and Artificial Intelligence, 2013, 26(11): 1050-1056.) [2] ROY K, MUKHERJEE J. Image Similarity Measure Using Color Histogram, Color Coherence Vector, and Sobel Method. Internatio-nal Journal of Science and Research, 2013, 2(1): 538-543. [3] 张琳波,王春恒,肖柏华,等.基于Bag-of-Phrases的图像表示方法.自动化学报, 2012, 38(1): 46-54. (ZHANG L B, WANG C H, XIAO B H, et al. Image Representation Using Bag-of-Phrases. Acta Automatica Sinica, 2012, 38(1): 46-54.) [4] YOU F C, ZHANG C. The Technique of Color and Shape-Based Multi-feature Combination of Trademark Image Retrieval // Proc of the International Conference on Multimedia Technology. Washington, USA: IEEE, 2010. DOI: 10.1109/ICMULT.2010.5631275. [5] SIKIRIC/ I, BRKIC/ K, S‘EGVIC/ S. Classifying Traffic Scenes Using the Gist Image Descriptor[C/OL]. [2016-12-23]. https://arxiv.org/ftp/arxiv/papers/1310/1310.0316.pdf. [6] DOUZE M, JGOU H, SANDHAWALIA H, et al. Evaluation of GIST Descriptors for Web-Scale Image Search // Proc of the ACM International Conference on Image and Video Retrieval. New York, USA: ACM, 2009. DOI: 10.1145/1646396.1646421. [7] PASS G, ZABIH R. Histogram Refinement for Content-Based Image Retrieval // Proc of the 3rd IEEE Workshop on Applications of Computer Vision. Washington, USA: IEEE, 1996: 96-102. [8] LOWE D G. Object Recognition from Local Scale-Invariant Features // Proc of the 7th IEEE International Conference on Computer Vision. Washington, USA: IEEE, 1999: 1150-1157. [9] BAY H, TUYTELAARS T, VAN GOOL L. SURF: Speeded up Robust Features // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2006: 404-417. [10] BAY H, ESS A, TUYTELAARS T, et al. Speeded-Up Robust Features(SURF). Computer Vision & Image Understanding, 2008, 110(3): 346-359. [11] RUBLEE E, RABAUD V, KONOLIGE K, et al. ORB: An Efficient Alternative to SIFT or SURF // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2011: 2564-2571. [12] NOWAK E, JURIE F, TRIGGS B. Sampling Strategies for Bag-of-Features Image Classification // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2008: 490-503. [13] ROTH P M, WINTER M. Survey of Appearance-Based Methods for Object Recognition. Technical Report, ICG-TR-01/08. Graz, Austria: Graz University of Technology, 2008. [14] 梁志伟,陈燕燕,朱松豪,等.基于视觉词典的单目视觉闭环检测算法.模式识别与人工智能, 2013, 26(6): 561-570. (LIANG Z W, CHEN Y Y, ZHU S H, et al. Loop Closure Detection Algorithm Based on Monocular Vision Using Visual Dictionary. Pattern Recognition and Artificial Intelligence, 2013, 26(6): 561-570.) [15] CUMMINS M, NEWMAN P. FAB-MAP: Appearance-Based Place Recognition and Mapping Using a Learned Visual Vocabulary Model // Proc of the 27th International Conference on Machine Learning. Washington, USA: IEEE, 2010: 3-10. [16] VEDALDI A, LENC K. Matconvnet: Convolutional Neural Networks for Matlab // Proc of the 23rd ACM International Conference on Multimedia. New York, USA: ACM, 2015: 689-692. [17] 李 钦,游 雄,李 科,等.图像深度层次特征提取算法.模式识别与人工智能, 2017, 30(2): 127-136. (LI Q, YOU X, LI K, et al. Deep Hierarchical Feature Extraction Algorithm. Pattern Recognition and Artificial Intelligence, 2017, 30(2): 127-136.) [18] SMITH M, BALDWIN I, CHURCHILL W, et al. The New College Vision and Laser Data Set. The International Journal of Robotics Research, 2009, 28(5): 595-599. [19] CUMMINS M J, NEWMAN P M. FAB-MAP: Probabilistic Localization and Mapping in the Space Appearance. The International Journal of Robotics Research, 2008, 27(6): 647-665. [20] CUMMINS M, NEWMAN P. Highly Scalable Appearance-Only SLAM-FAB-MAP 2.0[C/OL]. [2016-12-23]. http://www.roboticsproceedings.org/rss05/p39.pdf. [21] LABBE M, MICHAUD F. Appearance-Based Loop Closure Detection for Online Large-Scale and Long-Term Operation. IEEE Transaction on Robotics, 2013, 29(3): 734-745. [22] 李永锋,张国良,王 蜂,等.一种基于历史模型集的改进闭环检测算法.机器人, 2015, 37(6): 663-673. (LI Y F, ZHANG G L, WANG F, et al. An Improved Loop Closure Detection Algorithm Based on Historical Model Set. Robot, 2015, 37(6): 663-673.)
2017, Vol. 30 
