河道山地灾害的卷积神经网络快速识别方法
|
摘要
河道山地灾害如泥石流、滑坡、山洪、水土流失等严重危害着河道周边公路、铁路、桥梁、大型水利工程等重要基础设施的安全。快速识别已发生的河道山地灾害意义重大,而传统巡检方式具有极高的危险性和滞后性,迫切需要新方法来替代。以深度卷积神经网络为代表的深度学习技术具有局部感知、参数共享、池化等多个特性,相比传统机器学习方法具有更强大的特征学习和特征表达能力。在深度学习开源框架下,利用大量河道山地灾害图片数据完成了Caffenet等多个深度模型的训练,并结合迁移学习方法,使河道山地灾害识别准确率最终达到90%以上,为河道山地灾害快速识别、群防群测体系的完善提供了新思路。
The mountain hazards in the river courses such as the debris flow, landslide, mountain flood and soil erosion seriously threaten the safety of the important infrastructures such as roads, railways, bridges and large-scale hydroprojects around the river. More than 90% of the traffic interruption in the Sichuan-Tibet highway are caused by the mountain hazards, which restrict the development of economy and society in Tibet. It is of great significance to quickly identify the mountain hazards in the river courses for timely adoption of appropriate emergency disaster plans and release evacuation information. The observation of the mountain hazards in the river courses by professional inspectors in a traditional way is of great danger and obvious lag. Therefore, it is urgent to study a new method as an alternative. With the arrival of the big data era, a deep learning technology represented by the convolution neural network has many characteristics such as local connectivity, parameter sharing and pooling, and has more powerful features and learning skills compared with the traditional machine learning method. This study has completed the training of Caffenet and other depth models by using a large number of image data of the mountain hazards in the river courses with the deep learning open source framework. By use of the transfer learning, the recognition accuracy eventually reached more than 90 percent. The analysis results show that the recognition method provides a new idea for the fast recognition of the typical mountain hazards in the river courses and the improvement of the group measuring system.
The mountain hazards in the river courses such as the debris flow, landslide, mountain flood and soil erosion seriously threaten the safety of the important infrastructures such as roads, railways, bridges and large-scale hydroprojects around the river. More than 90% of the traffic interruption in the Sichuan-Tibet highway are caused by the mountain hazards, which restrict the development of economy and society in Tibet. It is of great significance to quickly identify the mountain hazards in the river courses for timely adoption of appropriate emergency disaster plans and release evacuation information. The observation of the mountain hazards in the river courses by professional inspectors in a traditional way is of great danger and obvious lag. Therefore, it is urgent to study a new method as an alternative. With the arrival of the big data era, a deep learning technology represented by the convolution neural network has many characteristics such as local connectivity, parameter sharing and pooling, and has more powerful features and learning skills compared with the traditional machine learning method. This study has completed the training of Caffenet and other depth models by using a large number of image data of the mountain hazards in the river courses with the deep learning open source framework. By use of the transfer learning, the recognition accuracy eventually reached more than 90 percent. The analysis results show that the recognition method provides a new idea for the fast recognition of the typical mountain hazards in the river courses and the improvement of the group measuring system.
引文
[1] 崔鹏.中国山地灾害研究进展与未来应关注的科学问题[J].地理科学进展,2014,33(2):145- 152.(CUI Peng.Progress and prospects in research on mountain hazards in China [J].Progress in Geography,2014,33(2):145- 152.(in Chinese))
[2] 宋章,张广泽,蒋良文,等.川藏铁路主要地质灾害特征及地质选线探析[J].铁道标准设计,2016,60(1):14- 19.(SONG Zhang,ZHANG Guangze,JIANG Liangwen,et al.Analysis of the characteristics of major geological disasters and geological alignment of Sichuan-Tibet Railway[J].Railway Standard Design,2016,60(1):14- 19.(in Chinese))
[3] 李宏恩,何勇军.水库与山洪灾害防治协同预警模式[J].水利水运工程学报,2017(1):37- 42.(LI Hongen,HE Yongjun.Early collaborative warning mode for non-engineering measures of reservoirs and flash flood disaster prevention [J].Hydro-Science and Engineering,2017(1):37- 42.(in Chinese))
[4] 郭志学,彭清娥,汤雷,等.堆积体作用下陡坡河道流速沿程及横向分布规律[J].水利水运工程学报,2011(4):27-31.(GUO Zhixue,PENG Qing’e,TANG Lei,et al.Streamwise and lateral velocity distribution of steep river under the action of deposit body[J].Hydro-Science and Engineering,2011(4):27-31.(in Chinese))
[5] SZEGEDY C,LIU N W,JIA N Y,et al.Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE Computer Society,2015.
[6] 崔永锋,刘伟.远程采集图像特征的优化识别过程仿真[J].控制工程,2016,23(7):1053- 1056.(CUI Yongfeng,LIU Wei.Simulation of optimization and recognition of image features in remote collection[J].Control Engineering of China,2016,23(7):1053- 1056.(in Chinese))
[7] LECUM Y L,BOTTOU L,BENGIO Y,et al.Gradient-based learning applied to document recognition [J].Proceedings of the IEEE,1998,86(11):2278- 2324.
[8] KRIZHEVSKY A,SUTSKEVER I,HINTON G.ImageNet classification with deep convolutional neural networks[J].Advances in Neural Information Processing Systems,2012,25(2):1097-1105.
[9] ZHENG Q,YANG M,YANG J,et al.Improvement of generalization ability of deep CNN via implicit regularization in two-stage training process[J].IEEE Access,2018,6:15844- 15869.
[10] 李明威.图像分类中的卷积神经网络方法研究[D].南京:南京邮电大学,2016.(LI Mingwei.Research of convolutional neural network in image classification[D].Nanjing:Nanjing University of Posts and Telecommunications,2016.(in Chinese))
[11] YOSINSKI J,CLUNE J,BENGIO Y,et al.How transferable are features in deep neural networks[J].Eprint Arxiv,2014,27:3320-3328.
[12] 中国科学院水利部成都山地灾害与环境研究所,西藏自治区交通科学研究所.川藏公路典型山地灾害研究[M].成都:成都科技大学出版社,1999.(Institute of Mountain Hazards and Environment,Chinese Academy of Sciences and Water Conservancy Ministry,Institute of the Traffic Science,Tibet Autonomous Region.A study of typical mountain hazards along Sichuan-Tibet Highway [M].Chengdu:Chengdu University of Science and Technology Press,1999.(in Chinese))
[13] 罗德富,毛济周,朱平一.川藏公路南线 (西藏境内)山地灾害及防治对策[M].北京:科学出版社,1995.(LUO Defu,MAO Jizhou,ZHU Pingyi.Mountain disasters and prevention countermeasures in Sichuan-Tibet Highway South (Tibet) [M].Beijing:Science Press,1995.(in Chinese))
[14] 王培清,徐国涛,何强.西藏藏东南地区典型地质灾害成因及防治技术浅析[J].西藏大学学报(自然科学版),2013,28(1):16- 20.(WANG Peiqing,XU Guotao,HE Qiang.Analysis on the causes of typical geological disasters in the southeastern Tibet and its prevention technology [J].Journal of Tibet University,2013,28(1):16- 20.(in Chinese))
[2] 宋章,张广泽,蒋良文,等.川藏铁路主要地质灾害特征及地质选线探析[J].铁道标准设计,2016,60(1):14- 19.(SONG Zhang,ZHANG Guangze,JIANG Liangwen,et al.Analysis of the characteristics of major geological disasters and geological alignment of Sichuan-Tibet Railway[J].Railway Standard Design,2016,60(1):14- 19.(in Chinese))
[3] 李宏恩,何勇军.水库与山洪灾害防治协同预警模式[J].水利水运工程学报,2017(1):37- 42.(LI Hongen,HE Yongjun.Early collaborative warning mode for non-engineering measures of reservoirs and flash flood disaster prevention [J].Hydro-Science and Engineering,2017(1):37- 42.(in Chinese))
[4] 郭志学,彭清娥,汤雷,等.堆积体作用下陡坡河道流速沿程及横向分布规律[J].水利水运工程学报,2011(4):27-31.(GUO Zhixue,PENG Qing’e,TANG Lei,et al.Streamwise and lateral velocity distribution of steep river under the action of deposit body[J].Hydro-Science and Engineering,2011(4):27-31.(in Chinese))
[5] SZEGEDY C,LIU N W,JIA N Y,et al.Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE Computer Society,2015.
[6] 崔永锋,刘伟.远程采集图像特征的优化识别过程仿真[J].控制工程,2016,23(7):1053- 1056.(CUI Yongfeng,LIU Wei.Simulation of optimization and recognition of image features in remote collection[J].Control Engineering of China,2016,23(7):1053- 1056.(in Chinese))
[7] LECUM Y L,BOTTOU L,BENGIO Y,et al.Gradient-based learning applied to document recognition [J].Proceedings of the IEEE,1998,86(11):2278- 2324.
[8] KRIZHEVSKY A,SUTSKEVER I,HINTON G.ImageNet classification with deep convolutional neural networks[J].Advances in Neural Information Processing Systems,2012,25(2):1097-1105.
[9] ZHENG Q,YANG M,YANG J,et al.Improvement of generalization ability of deep CNN via implicit regularization in two-stage training process[J].IEEE Access,2018,6:15844- 15869.
[10] 李明威.图像分类中的卷积神经网络方法研究[D].南京:南京邮电大学,2016.(LI Mingwei.Research of convolutional neural network in image classification[D].Nanjing:Nanjing University of Posts and Telecommunications,2016.(in Chinese))
[11] YOSINSKI J,CLUNE J,BENGIO Y,et al.How transferable are features in deep neural networks[J].Eprint Arxiv,2014,27:3320-3328.
[12] 中国科学院水利部成都山地灾害与环境研究所,西藏自治区交通科学研究所.川藏公路典型山地灾害研究[M].成都:成都科技大学出版社,1999.(Institute of Mountain Hazards and Environment,Chinese Academy of Sciences and Water Conservancy Ministry,Institute of the Traffic Science,Tibet Autonomous Region.A study of typical mountain hazards along Sichuan-Tibet Highway [M].Chengdu:Chengdu University of Science and Technology Press,1999.(in Chinese))
[13] 罗德富,毛济周,朱平一.川藏公路南线 (西藏境内)山地灾害及防治对策[M].北京:科学出版社,1995.(LUO Defu,MAO Jizhou,ZHU Pingyi.Mountain disasters and prevention countermeasures in Sichuan-Tibet Highway South (Tibet) [M].Beijing:Science Press,1995.(in Chinese))
[14] 王培清,徐国涛,何强.西藏藏东南地区典型地质灾害成因及防治技术浅析[J].西藏大学学报(自然科学版),2013,28(1):16- 20.(WANG Peiqing,XU Guotao,HE Qiang.Analysis on the causes of typical geological disasters in the southeastern Tibet and its prevention technology [J].Journal of Tibet University,2013,28(1):16- 20.(in Chinese))