Abstract Extracting the image features with strong representation is critical to complete different vision tasks. The traditional features only describe one aspect of the image information, and therefore their representation capability is limited. In this paper, deep hierarchical feature(DHF) extraction algorithm based on the convolutional neural networks(CNN) is proposed. The essential information hidden inside the image is effectively mined by abstractly expressing the image in different layers. Firstly, the feature maps of the image are created based on CNN, and those in the convolutional layers are selected to construct the hierarchical structure of the image. Then, the best layer combination is determined according to the matching experiment. The feature maps in the low layers are described by the information entropy, and the ones in high layers are described by averaging the pixels in specified region. the DHF with strong representative capability is ultimately constructed. The experiment demonstrates that the proposed DHF has evident advantages compared with the existing features, and it can complete the matching task with high efficiency.
Fund:Supported by National Natural Science Foundation of China(No.41201390), Project of Science and Technology Innovation of Henan Province(No.142101510005)
About author:: (LI Qin, born in 1990, Ph.D. candidate. His research interests include computer vision and deep learning.)(YOU Xiong, born in 1962, Ph.D., professor. His research interests include virtual geographic environment(VGE) and geographic information system(GIS).)(LI Ke(Corresponding author), born in 1977, Ph,D., associate professor. His research interests include image processing and computer vision.) (TANG Fen, born in 1991, Ph.D. candidate. His research interests include map design.)
[1] ROY K, MUKHERJRR J. Image Similarity Measure Using Color Histogram, Color Coherence Vector, and Sobel Method. International Journal of Science and Research, 2013, 2(1): 538-543. [2] 熊芳敏,岑宇森,曾碧卿.熵权法融合局部匹配Gabor特征的鲁棒人脸识别.计算机工程与应用, 2016, 52(5): 134-140. (XIONG F M, CEN Y S, ZENG B Q. Robust Face Recognition Based on Fusion of Entropy Weight Method and Local Matching Gabor Features. Computer Engineering and Applications, 2016, 52(5): 134-140.) [3] 徐天阳,吴小俊.基于Gabor变换的快速跟踪算法.计算机工程与应用, 2016, 52(6): 134-138. (XU T Y, WU X J. Fast Tracking Algorithm Based on Gabor Transformation. Computer Engineering and Applications, 2016, 52(6): 134-138.) [4] 夏 瑜.基于结构的纹理特征及应用研究.博士学位论文.合肥:中国科学技术大学, 2014. (XIA Y. Research on the Structural Texture Feature Descriptor and its Application. Ph.D Dissertation. Hefei, China: University of Science and Technology of China, 2014.) [5] LOWE D G. Object Recognition from Local Scale-Invariant Features // Proc of the 7th IEEE International Conference on Computer Vision. Washington, USA: IEEE, 1999. DOI: 10.1109/ICCV.1999.790410. [6] LOWE D G. Distinctive Image Features from Scale-Invariant Keypoints. International Journal of Computer Vision, 2004, 60(2): 91-110. [7] 李 钦,游 雄,李 科,等.PCA-SIFT 特征匹配算法研究.测绘工程, 2016, 25(4): 19-24. (LI Q, YOU X, LI K, et al. Research on the Matching Algorithm of PCA-SIFT. Engineering of Surveying and Mapping, 2016, 25(4): 19-24) [8] 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. [9] BAY H, ESS A, TUYTELAARS T, et al. Speeded-Up Robust Features(SURF). Computer Vision and Image Understanding, 2008, 110(3): 346-359. [10] KE Y, SUKTHANKAR R. PCA-SIFT: A More Distinctive Representation for Local Image Descriptors // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2004: 506-513. [11] LI Q, LI K, YOU X G, et al. Place Recognition Based on Deep Feature and Adaptive Weighting of Similarity Matrix. Neurocomputing, 2016, 199: 114-127. [12] ABDEL-HAMID O, LI D, YU D. Exploring Convolutional Neural Network Structures and Optimization Techniques for Speech Recognition[C/OL]. [2016-08-24]. https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/CNN-Interspeech2013_ pub.pdf. [13] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet Classification with Deep Convolutional Neural Networks[C/OL]. [2016-08-24]. http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf. [14] CIRESAN D C, MEIER U, GAMBARDELLA L M, et al. Convolutional Neural Network Committees for Handwritten Character Classification // Proc of the International Conference on Document Analysis and Recognition. Washington, USA: IEEE, 2011: 1135-1139.
[15] 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. [16] HAN J W, CHEN C Y, SHAO L, et al. Learning Computational Models of Video Memorability from fMRI Brain Imaging. IEEE Transactions on Cybernetics, 2015, 45(8): 1692-1703.
2017, Vol. 30 
