基于流行学习的路面破损图像多特征融合与可视化

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (578 KB) HTML ()
输出: BibTeX | EndNote (RIS)

摘要 For multi-feature fusion in automatic recognition of pavement distress images, we proposed a multi-feature fusion method based on manifold learning. In this method, the intrinsic features of pavement distress images are extracted through mapping the high dimensional data combing projection, mixture density factor and second order moment invariant into the low dimensional space. The multiple features are fused and the visualization of pavement distress images is implemented. In the experiments, we applied the multi-feature fusion method in the detection of pavement distress images. Two-dimensional features are firstly extracted from the 8 combining features, then the recognition effects on the 2D features of 4 methods including ELM, KNN, SVM and BP network are compared. The experiment result shows that the proposed method effectively improved the detection accuracy of pavement distress images. Simultaneously, the physical meaning of the 2D features is obtained through visualizing, one feature preliminary denotes the complexity and damaged extent of the cracks in images, the other describes the direction of the cracks.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	石陆魁
	周浩
	刘文浩

关键词 ： road engineering, multiple feature fusion, manifold learning, pavement distress image, visualization

Abstract：For multi-feature fusion in automatic recognition of pavement distress images, we proposed a multi-feature fusion method based on manifold learning. In this method, the intrinsic features of pavement distress images are extracted through mapping the high dimensional data combing projection, mixture density factor and second order moment invariant into the low dimensional space. The multiple features are fused and the visualization of pavement distress images is implemented. In the experiments, we applied the multi-feature fusion method in the detection of pavement distress images. Two-dimensional features are firstly extracted from the 8 combining features, then the recognition effects on the 2D features of 4 methods including ELM, KNN, SVM and BP network are compared. The experiment result shows that the proposed method effectively improved the detection accuracy of pavement distress images. Simultaneously, the physical meaning of the 2D features is obtained through visualizing, one feature preliminary denotes the complexity and damaged extent of the cracks in images, the other describes the direction of the cracks.

Key words： road engineering multiple feature fusion manifold learning pavement distress image visualization

收稿日期: 2016-05-15

基金资助:Supported by the Tianjin Research Project of Application Foundation and Advanced Technology of China (No. 14JCZDJC31600);the Natural Science Foundation of Hebei Province of China (No. F2016202144).

通讯作者: SHI Lu-kui,E-mail address:shilukui@scse.hebut.edu.cn E-mail: shilukui@scse.hebut.edu.cn

引用本文:

石陆魁, 周浩, 刘文浩. 基于流行学习的路面破损图像多特征融合与可视化[J]. Journal of Highway and Transportation Research and Development, 2017, 11(1): 14-22.
SHI Lu-kui, ZHOU Hao, LIU Wen-hao. Multiple Feature Fusion and Visualization of Pavement Distress Images Based on Manifold Learning. Journal of Highway and Transportation Research and Development, 2017, 11(1): 14-22.

链接本文:

http://manu27.magtech.com.cn/Jwk_gljtkj_en/CN/ 或 http://manu27.magtech.com.cn/Jwk_gljtkj_en/CN/Y2017/V11/I1/14

[1] CHU Xiu-min, WANG Rong-ben, CHU Jiang-wei, et al. Asphalt Pavement Surface Distress Image Recognition Based on Moment Invariant Feature[J]. Journal of Jilin University:Engineering and Technology Edition, 2003,33(1):1-7. (in Chinese)
[2] XIAO Wang-xin, YAN Xin-ping, ZHANG Xue. Research on the Automatic Pavement Distress Recognition Based on Synthetical Distress Density Factor[J]. Journal of Transportation Engineering and Information, 2005, 3(2):19-26. (in Chinese)
[3] YING L, SALARI E. Beamlet Transform-Based Technique for Pavement Crack Detection and Classification[J]. Computer-Aided Civil and Infrastructure Engineering, 2010, 25(8):572-580.
[4] FUJITA Y, HAMAMOTO Y. A Robust Automatic Crack Detection Method from Noisy Concrete Surfaces[J]. Machine Vision and Applications, 2011, 22(2):245-254.
[5] NEJAD F M, ZAKERI H. An Expert System Based on Wavelet Transform and Radon Neural Network for Pavement Distress Classification[J]. Expert Systems with Applications, 2011, 38(6):7088-7101.
[6] YU Yong-bo, LI Wan-heng, ZHANG Jin-quan, et al. Bridge Cracks Extraction Method Based on Image Connection Domain[J]. Journal of Highway and Transportation Research and Development, 2011, 28(7):90-93. (in Chinese)
[7] MA Rong-gui, XU Kun, LIU Fei-fei. Highway Surface Crack Images Identifying Algorithm[J]. Journal of Transport Information and Safety, 2014, 32(2):90-94. (in Chinese)
[8] RONG Jing, PAN Yu-li. Digital Image Based Crack Detection of Grooved Cement Concrete Pavement[J]. Journal of Highway and Transportation Research and Development, 2012,29(3):45-50. (in Chinese)
[9] OLIVEIRA H, CORREIA P L. Automatic Road Crack Detection and Characterization[J]. IEEE Transactions on Intelligent Transportation Systems, 2013, 14(1):155-168.
[10] SHEN Zhao-qing, PENG Yu-hua, SHU Ning. A Road Damage Identification Method Based on Scale-span Image and SVM[J]. Geomatics and Information Science of Wuhan University, 2013, 38(8):993-997. (in Chinese)
[11] SUN Chao-yun, ZHAO Hai-wei, LI Wei, et al. 3D Pavement Crack Identification Method Based on Dual-phase Scanning Detection[J]. China Journal of Highway and Transport, 2015, 28(2):26-32. (in Chinese)
[12] HUANG G B, ZHOU H, DING X, et al. Extreme Learning Machine for Regression and Multiclass Classification[J]. IEEE Transactions on Systems, Man & Cybernetics Part B:Cybernetics, 2012, 42(2):513-529.
[13] ZUO Y X, WANG G Q, ZUO C C. The Segmentation Algorithm for Pavement Cracking Images Based on the Improved Fuzzy Clustering[J]. Applied Mechanics and Materials, 2013, 319:362-366.
[14] RADUCANU B, DORNAIKA F. A Supervised Non-linear Dimensionality Reduction Approach for Manifold Learning[J]. Pattern Recognition, 2012, 45(6):2432-2444.
[15] SUN B Y, ZHANG X M, LI J, et al. Feature Fusion Using Locally Linear Embedding for Classification[J]. IEEE Transactions on Neural Networks, 2010, 21(1):163-168.
[16] HU Jian-zhong, WU Yao, XIE Xiao-xin. A Method for Fault Recognition Based on LLE Feature Fusion[J]. China Mechanical Engineering, 2013, 24(24):3345-3348. (in Chinese)
[17] CRIMINISI A, SHOTTON J, KONUKOGLU E. Decision Forests:A Unified Framework for Classification, Regression, Density Estimation, Manifold Learning and Semi-Supervised Learning[J]. Foundations and Trends in Computer Graphics and Vision, 2012, 7(2/3):81-227.
[18] TENENBAUM J B, DE SILVA V, LANGFORD J C. A Global Geometric Framework for Nonlinear Dimensionality Reduction[J]. Science, 2000, 290(5500):2319-2323.
[19] ROWEIS S T, SAUL L K. Nonlinear Dimensionality Reduction by Locally Linear Embedding[J]. Science, 2000, 290(5500):2323-2326.
[20] BELKIN M, NIYOGI P. Laplacian Eigenmaps for Dimensionality Reduction and Data Representation[J]. Neural Computation, 2003, 15(6):1373-1396.
[21] ZHANG Z Y, ZHA H Y. Principal Manifolds and Nonlinear Dimensionality Reduction via Local Tangent Space Alignment[J]. Journal of Shanghai University, 2004, 26(4):406-424.
[22] XIAO Wang-xin, ZHANG Xue, HUANG Wei, et al. One New Algorithm of Automatic Classification for Pavement Distress[J]. Journal of Highway and Transportation Research and Development, 2005, 22(11):75-78. (in Chinese)
[23] LIU Guo-hai, JIANG Zhi-jia. Recognition of Porcelain Bottle Crack Based on Modified ART-2 Network and Invariant Moment[J]. Chinese Journal of Scientific Instrument, 2009, 30(7):1420-1425.(in Chinese)