Abstract：Rock classification and boundary determination of rock slopes are very important for the analysis of slope stability. At present, the artificial methods are inefficient and affected by subjective factors. So a convolution neural network model for the image set analysis of a rock slope is established based on Tensorflow. Through convolution operation and pooling operation, the feature information of 8000 original rock slope images is extracted and compressed respectively. Then the network model is trained to realize the automatic recognition and classification of the rock slope. The model is tested and analyzed by using the images of rock slopes in training set and testing set. The accuracy rate of the training set and the testing set is 98% and 90%, respectively. It is shown that the network model after training has good robustness and achieves ideal training effect. Next, the color of different rocks on the slope is taken as the main basis. The boundary of different types of rock on the rock slope is calibrated by the deep learning boundary extraction technology. To verify the effectiveness of the algorithm, the standard color image of the rock slope is selected for simulation experiment, and the results of boundary detection are accurate. The network model established by deep learning realizes the requirements of rapid and automatic rock identification and boundary range division of rock slopes, and introduces the rock slope information acquired by image recognition into the GeoSMA-3D software independently developed by the team,as an important parameter for determining the grade of rock slopes.
王鹏宇, 王述红. 四类常见边坡岩石类别识别和边界范围确定的方法[J]. 岩土工程学报, 2019, 41(8): 1505-1512.
WANG Peng-yu, WANG Shu-hong. Method for identifying four common rock types of slopes and determining boundary range. Chinese J. Geot. Eng., 2019, 41(8): 1505-1512.
[1] SINGH C, KAUR K P.A fast and efficient image retrieval system based on color and texture features[J]. Journal of Visual Communication & Image Representation, 2016, 41(10): 41-52. [2] SMITH S M, BRADY J M.SUSAN: a new approach to low level image processing[J]. International Journal of Computer Vision, 2015, 23(1): 45-78. [3] 王洋, 王述红, 郭牡丹, 等. 隧道掌子面节理信息快速数字识别及稳定性分析[J]. 岩土工程学报, 2011, 33(11): 1734-1739. (WANG Yang, WANG Shu-hong, GUO Mu-dan, et al.Fast digital identification of joint information of tunnel work face and its stability analysis[J]. Chinese Journal of Geotechnical Engineering, 2011, 33(11): 1734-1739. (in Chinese)) [4] BOONSIVANON K, MEESOMBOON A.IKDSIFT: an improved keypoint detection algorithm based-on sift approach for non-uniform illumination[J]. Procedia Computer Science, 2016, 86(2): 269-272. [5] ZISSERMAN A, VEDALDI A.Sparse kernel approximations for efficient classification and detection[C]// IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, Hawaii, 2012: 2320-2327. [6] KOBAYASHI T.BFO Meets HOG: feature extraction based on histograms of oriented pdf gradients for image classifi- cation[C]// IEEE Conference on Computer Vision and Pattern Recognition. Portland OR, 2013, 91(9): 747-754. [7] ZAINI N, van der MEER F, van der WERFF H. Deter- mination of carbonate rock chemistry using laboratory-based hyperspectral imagery[J]. Remote Sensing, 2014, 6(5): 4149-4172. [8] 郭清宏, 周永章, 曹姝旻. 广绿玉玉石的矿物学研究[J]. 中山大学学报(自然科学版), 2010, 49(3): 146-151. (GUO Qing-hong, ZHOU Yong-zhang, CAO Shu-min.Study on mineralogy of Guangning jade[J]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2010, 49(3): 146-151. (in Chinese)) [9] 肖凡, 陈建国, 侯卫生. 2017. 钦—杭结合带南段庞西垌地区Ag-Au致矿地球化学异常信息识别与提取[J]. 岩石学报, 2017, 33(3): 779-790. (XIAO Fan, CHEN Jian-guo, HOU Wei-sheng.Identification and extraction of Ag-Au mineralization associated geochemical anomaly in Pangxitong district, southern part of the Qinzhou-Hangzhou Metallogenic Belt, China[J]. Acta Petrologica Sinica, 2017, 33(3): 779-790. (in Chinese)) [10] 张旗, 金惟俊, 李承东. 再论花岗岩按照Sr-Yb的分类:标志[J]. 岩石学报, 2010, 26(4): 985-1015. (ZHANG Qi, JIN Wei-jun, LI Cheng-dong.Revisiting the new classification of granitic rocks based on whole-rock Sr and Yb contents: index[J]. Acta Petrologica Sinica, 2010, 26(4): 985-1015. (in Chinese)) [11] 张旗, 金惟俊, 李承东. 三论花岗岩按照Sr-Yb的分类: 应用[J]. 岩石学报, 2010, 26(12): 3431-3455. (ZHANG Qi, JIN Wei-jun, LI Cheng-dong.On the classification of graniti crocks based on whole-rock Sr and Yb concentrations III: practice[J]. Acta Petrologica Sinica, 2010, 26(12): 3431-3455. (in Chinese)) [12] 张旭, 蒋建国, 洪日昌. 基于朴素贝叶斯K近邻的快速图像分类算法[J]. 北京航空航天大学学报, 2015, 41(2): 302-310. (ZHANG Xu, JIANG Jian-guo, HONG Ri-chang.Accelerated image classification algorithm based on naive Bayes K-nearest neighbor[J]. Journal of Beijing University of Aeronautics and Astronautics, 2015, 41(2): 302-310. (in Chinese)) [13] 康丽萍, 孙显, 许光銮.加权KNN的图文数据融合分类[J]. 中国图象图形学报, 2016, 21(7): 854-864. (KANG Li-ping, SUN Xian, XU Guang-luan.Fusion method via KNN with weight adjustment for the classification of image-text co-occurrence data[J]. Journal of Image and Graphics, 2016, 21(7): 854-864. (in Chinese)) [14] 张嘉凡, 张雪娇, 杨更社. 基于聚类算法的岩石CT图像分割及量化方法[J]. 西安科技大学学报, 2016, 36(2): 171-175. (ZHANG Jia-fan, ZHANG Xue-jiao, YANG Geng-she.A method of rock CT image segmentation and quantification based on clustering algorithm[J]. Journal of Xi'an University of Science and Technology, 2016, 36(2): 171-175. (in Chinese)) [15] LI N, HAO H Z, GU Q, et al.A transfer learning method for automatic identification of sandstone microscopic image. Computers ＆ Geosciences, 2017, 103(2): 111-121. [16] 谭永杰. 地质大数据体系建设的总体框架研究[J]. 中国地质调查, 2016, 3(3): 1-6. (TAN Yong-jie.Study on the overall framework of the construction of large geological data system[J]. Geological Survey of China, 2016, 3(3): 1-6. (in Chinese)) [17] 郑泽宇, 顾思宇. TENSORFLOW实战GOOGLE深度学习框架[M]. 北京: 电子工业出版社, 2017. (ZHENG Ze-yu, GU Si-yu.TENSORFLOW practical GOOGLE deep learning framework[M]. Beijing: Electronic Industry Press, 2017. (in Chinese)) [18] WU Y C, YIN F, LIU C L.Improving handwritten Chinese text recognition using neural network language models and convolutional neural network shape models[J]. Pattern Recognition, 2017, 65(5): 251-264. [19] BIANCO S, BUZZELLI M, MAZZINI D, et al.Deep learning for logo recognition[J]. Neurocomputing, 2017, 245(1): 23-30. [20] ESTEVA A, KUPREL B, NOVOA R A, et al.Dermatologist- level classification of skin cancer with deep neural networks[J]. Nature, 2017, 542(7639): 115-118. [21] GONG M G, YANG H L, ZHANG P Z.Feature learning and change feature classification based on deep learning for ternary change detection in SAR images[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 129(2): 212-225. [22] 柳厚祥, 李汪石, 查焕奕. 基于深度学习技术的公路隧道围岩分级方法[J]. 岩土工程学报, 2018, 40(10): 1809-1816. (LIU Hou-xiang, LI Wang-shi, ZHA Huan-yi.Method for surrounding rock mass classification of highway tunnels based on deep learning technology[J]. Chinese Journal of Geotechnical Engineering, 2018, 40(10): 1809-1816. (in Chinese)) [23] 杨善统, 姜清辉, 尹涛, 等. 边坡临界滑面搜索的改进粒子群优化算法[J]. 岩土工程学报, 2015, 37(8): 1411-1417. (YANG Shan-tong, JIANG Qing-hui, YIN Tao, et al.Search of critical slip surface of slopes using improved particle swarm optimization method[J]. Chinese Journal of Geotechnical Engineering, 2015, 37(8): 1411-1417. (in Chinese)) [24] 康石磊, 阳军生, 杨峰. 含裂隙类岩试样破坏行为的宏细观数值分析[J]. 长江科学院院报, 2016, 33(4): 71-77. (KANG Shi-lei, YANG Jun-sheng, YANG Feng.Macro- and meso-analysis of failure behavior of rock-like samples with flaws using a numerical method[J]. Journal of Yangtze River Scientific Research Institute, 2016, 33(4): 71-77. (in Chinese))
