Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (2): 293-303    DOI: 10.11873/j.issn.1004-0323.2021.2.0293
CNN 专栏     
基于SSD模型的京津冀地区尾矿库检测
李庆1,2(),陈俊杰1,李庆亭2,李柏鹏2,卢凯旋2,昝露洋2,陈正超2()
1.河南理工大学测绘与国土信息工程学院,河南 焦作 454000
2.中国科学院空天信息创新研究院,北京 100094
Detection of Tailings Pond in Beijing—Tianjin—Hebei Region based on SSD Model
Qing Li1,2(),Junjie Chen1,Qingting Li2,Baipeng Li2,Kaixuan Lu2,Luyang Zan2,Zhengchao Chen2()
1.School of Surveying and Land Information Engineering,Henan Polytechnic University,Jiaozuo 454000,China
2.Aerospace Information Research Institute, Chinese Academy of Sciences,Beijing 100094,China
 全文: PDF(11981 KB)   HTML
摘要:

我国尾矿库事故频发,所造成的危害极其严重。掌握尾矿库的数量及分布情况对预防尾矿库事故和开展尾矿库应急工作具有重大意义。传统的调查方法主要以地面调查为主,难以做到大范围高频次的监测。因此提出了一种基于深度学习的尾矿库目标检测方法,可以快速识别尾矿库的位置并掌握其地理分布。首先分析尾矿库在遥感图像上的特征,制作适合训练的样本,根据样本的情况优化调整训SSD (Single Shot Multibox Detector)模型,基于优化后的模型进行京津冀地区尾矿库的自动提取。实验结果表明:京津冀地区检测出尾矿库2 696座,召回率达到93.3%。说明采用深度学习目标检测的方法提取尾矿库,取得了较好的效果,所提出的尾矿库提取方法可应用于全国及全球尾矿库的提取。

关键词: 遥感深度学习目标检测尾矿库京津冀    
Abstract:

The accidents of the tailing ponds in China are frequent, the damage caused by dam breaking is extremely serious. The current quantity and distribution of tailings pond is necessary for preventing tailings pond accidents and carrying out emergency work in tailings pond. The traditional survey method is mainly based on ground investigations, which is difficult to achieve large-scale high-frequency monitoring. A tailing pond detection method based on deep learning detection was proposed in this paper, which can quickly identify the locations of the tailing ponds and obtain their geographical distribution. The suitable training samples are produced based on the study of the characteristics of the tailing ponds on the remote sensing image. SSD (Single Shot Multibox Detector) model is adjusted according to the samples characteristics during the model training. The extraction of the tailing ponds in the Beijing-Tianjin-Hebei Region is realized based on optimized model. The experiment result shows that there are 2 696 tailing ponds which were detected in the Beijing-Tianjin-Hebei Region,the recall reaches 93.3%.This paper realized the extract the tailings pond with the method of deep learning target detection, and has achieve good results which can provides method for the national and global extraction of tailing ponds.

Key words: Remote sensing    Deep learning    Target detection    Tailings pond    Beijing-Tianjin-Hebei
收稿日期: 2019-11-02 出版日期: 2021-05-24
ZTFLH:  P237  
基金资助: 国家自然科学基金项目(42071407)
通讯作者: 陈正超     E-mail: 1104411435@qq.com;chenzc@radi.ac.cn
作者简介: 李庆(1994-),男,安徽池州人,硕士研究生,主要从事摄影测量与遥感技术方面的研究。E?mail:1104411435@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
李庆
陈俊杰
李庆亭
李柏鹏
卢凯旋
昝露洋
陈正超

引用本文:

李庆,陈俊杰,李庆亭,李柏鹏,卢凯旋,昝露洋,陈正超. 基于SSD模型的京津冀地区尾矿库检测[J]. 遥感技术与应用, 2021, 36(2): 293-303.

Qing Li,Junjie Chen,Qingting Li,Baipeng Li,Kaixuan Lu,Luyang Zan,Zhengchao Chen. Detection of Tailings Pond in Beijing—Tianjin—Hebei Region based on SSD Model. Remote Sensing Technology and Application, 2021, 36(2): 293-303.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.2.0293        http://www.rsta.ac.cn/CN/Y2021/V36/I2/293

图1  京津冀地区审图号:GS(2019)1822
图2  京津冀区域尾矿库提取流程图
图3  SSD模型结构图
图4  4种类型的尾矿库
图5  尾矿库多样性
图6  尾矿库统计图
图7  不同空间分辨率遥感图像上的尾矿库
空间分辨率精确度
1米0.54
2米0.81
4米0.73
表1  不同空间分辨率图像下的预测精确度
图8  网络训练结果
置信度阈值真实个数预测个数正检个数TP误检个数FP漏检个数FN精确率召回率F1分数
0.19111 753900853110.510.990.68
0.21 054865189460.820.950.88
0.3896808881030.900.890.89
0.4810759511520.940.830.88
0.5738709292020.960.830.89
0.6686669172420.980.730.84
表2  预测结果统计表
图9  检测结果审图号:GS(2019)1822
图10  不同分辨率图像上的检测结果
图11  模型误检地物
1 Wang Tao,Hou Kepeng,Guo Zhenshi,et al.Application of Analytic Hierarchy Process to Tailings Pond Safety Operation Analysis[J].Rock and Soil Mechanics,2008,29(Sup.1):680-687.
1 王涛,侯克鹏,郭振世,张成良.层次分析法(AHP)在尾矿库安全运行分析中的应用[J].岩土力学,2008,29():680-686.
2 Tian Wenqi,Xie Xuyang.Tailings Pond Situtaion in China and Safety Countmeasure Suggestion[J].China Mine Engineering,2009,38(6):42-49.
2 田文旗,谢旭阳.我国尾矿库现状及安全对策的建议[J].中国矿山工程,2009,38(6):42-49.
3 Tan Qulin,Shao Yun.Application of Remote Sensing Technology to Environment Pollution Monitoring[J].Remote Sensing Technology and Application,2000,15(4):246-251.
3 谭衢霖,邵芸.遥感技术在环境污染监测中的应用[J].遥感技术与应用,2000,15(4):246-251.
4 Hao Lina,Zhang Zhi,He Wenxi,et al.Tailings Reservoir Recognition Factors of the High Resolution Remote Sensing Image in Southeastern Hubei[J].Remote Sensing for Land and Resources,2012,24(3):154-158.
4 郝利娜,张志,何文熹,等.鄂东南尾矿库高分辨率遥感图像识别因子研究[J].国土资源遥感,2012,24(3):154-158.
5 Qiang Jianhua.Remote Sensing Investigation and Environment Impact Analysis of Tailing Dams in Shaanxi Province[J].Metal Mine,2013,42(10):112-115.
5 强建华.陕西省尾矿库遥感调查与环境影响分析[J].金属矿山,2013,42(10):112-115.
6 Xiao Rulin,Lv Jie,Fu Zhuo,et al.The Application Remote Sensing in the Environment Risk Monitoring of Tailings Pond in Zhangjiakou City,China[J].Remote Sensing Technology and Application,2014,29(1):100-105.
6 肖如林,吕杰,付卓,等.张家口市尾矿库环境风险遥感监测应用研究[J].遥感技术与应用,2014,29(1):100-105.
7 Wang Chuangye,Du Xiaoya,Cai Zonglei.Extraction and Interpretation on Remote Sensing Information of Tailings and Solid Waster[J].Coal Technology,2017,36(7):17-19.
7 王创业,杜晓娅,蔡宗磊.尾矿及固体废弃物遥感信息提取与解译[J].煤炭技术,2017,36(7):17-19.
8 Zhou Yingjie,Wang Xiaohong,Yao Weiling,et al.Remote Sensing Investigation and Environment Impact Analysis of Tailing Ponds in Shandong Province[J].Geological Survey of China,20174,(4):88-92.周英杰,王晓红,姚维岭,等.山东省尾矿库遥感调查与环境影响分析[J].中国地质调查,2017,4(4):88-92.
9 Hinton GE, Osindero S, Teh YW. A Fast Learning Algorithm for Deep Belief Nets[J]. Neural Computation, 2006, 18(7): 1527-1554.
10 Ren S, He K, Girshick R. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.
11 Chai Zifeng.Method About Sliding-Windows-based Weakly Labeled Object Detection[D].Harbin:Harbin Institute of Technology,2016.
11 柴子峰.基于滑动窗口的弱标记物体检测方法研究[D].哈尔滨:哈尔滨工业大学,2016.
12 Joseph R, Santosh D, Ross G, et al. You Only Look Once: Unified, Real-time Object Detection[J]. CVPR2016.doi: 10.1109/CVPR.2016.91.
doi: 10.1109/CVPR.2016.91
13 Liu W, Anguelov D, Erhan D, et al. SSD:Single Shot Multibox Detector[C]∥ European Conference on Computer Vision. Springer, Cham,2016:21-37.
14 Girshick R.Fast R-CNN[C]∥ The IEEE International Conference on Computer Vision(ICCV),2015:1440-1448.
15 Zhang F,Du B,Zhang L,et al.Weakly Supervised Searning based on Coupled Convolutional Neural Networks for Aircraft Detection[J]. IEEE Transactions on Geoscience and Remote Sensing,2016,54(9):5553-5563. doi:10.1109/TGRS.2016. 2569141.
doi: 10.1109/TGRS.2016. 2569141
16 Tang J,Deng C,Huang G,et al.Compressed-domain Ship Detection on Spaceborne Optical Image Using Deep Neural Network and Extreme Learning Machine[J].IEEE Transactions on Geoscience and Remote Sensing,2015,53(3):1174-1185.doi:10.1109/TGRS.2014.2335751.
doi: 10.1109/TGRS.2014.2335751
17 Li Shumin,Feng Quanlong,Liang Qichun,et al.Aircraft in Domestic High Resolution Remote Sensing Images Using Deep-learning[J].Remote Sensing Technology and Application,2018,33(6):1095-1102.
17 李淑敏,冯权泷,梁其椿,等.基于深度学习的国产高分遥感影像飞机目标自动检测[J].遥感技术与应用,2018,33(6):1095-1102.
18 Huang Jin,Fu Fa,Zhao Na.Research on the Application of Deep Learning in Remote Sensing Images Objection Detection[J].Wireless Internet Technology,2019,16(1):11-13.
18 黄金,付发,赵娜.深度学习在遥感图像目标检测中的应用研究[J].无线互联科技,2019,16(1):11-13.
19 Fang Mengliang,Huang Gang.A Ship Detection Method in Optical Remote Sensing Images[J].Computer Technology and Development,2019(8):1-6.
19 方梦梁,黄刚.一种光学遥感图像船舶目标检测技术[J].计算机技术与发展,2019(8):1-6.
20 Tang Cong,Ling Yongshun,Zheng Kedong,et al.Object Detection Method of Multi-View SSD based on Deep Learning[J].Infrared and Laser Engineering,2018,47(1):302-310.
20 唐聪,凌永顺,郑科栋,等.基于深度学习的多视窗SSD目标检测方法[J].红外与激光工程,2018,47(1):302-310.
21 Zhu Minchao,Feng Tao,Zhang Yu.Remote Sensing Image Multi-target Detection Method based on FD-SSD[J].Computer Applications and Software,2019,36(1):232-238.
21 朱敏超,冯涛,张钰.基于FD-SSD的遥感图像多目标检测方法[J].计算机应用与软件,2019,36(1):232-238.
22 Ge Rongfeng,Xu Kaipeng,Chi Yanyan,et al.Eco-environmental Impact of Mineral Resources Exploitation in Beijing-Tianjin-Hebei Region[J].Chinese Journal of Environmental Management,2017,9(3):46-51.
22 葛荣凤,许开鹏,迟妍妍,等.京津冀地区矿产资源开发的生态环境影响研究[J].中国资源管理,2017,9(3):46-51.
23 Lv Jie.Remote Sensing Research and Application of Tailings Pond—A Case Study of Tailings Pond in Hebei Province[D].Beijing: ChinaUniversity of Geosciences,2014.
23 吕杰.尾矿库遥感监测技术研究与应用—以河北省尾矿库为例[D].北京:中国地质大学,2014.
24 Zhang Shifu. Discussion on the Method of High-resolution Images Acquisition based on Google Earth[J].Electronic Te-chnology, 2013(12): 16-18.
24 张世富. 基于Google Earth的高分辨率影像获取方法探讨[J]. 电子技术, 2013(12):16-18.
25 Lu Kaixuan,Li Guoqing,Chen Zhengchao,et al.Extraction of Steel Plants Based on Optimized SSD Network Incorporating Negative Sample’s Multi Channels[J].Journal of University of Chinese Academy of Sciences, 2020,37(3):352-359.
25 卢凯旋, 李国清, 陈正超, 等. 基于负样本多通道优化SSD网络的钢铁厂提取[J]. 中国科学院大学学报,2020,37(3):352-359.
26 Yan Kai,Shen Ting,Chen Zhengchao,et al.Automatic Extraction of Tailing Pond based on SSD of Deep Learning[J].Journal of University of Chinese Academy of Sciences,2020,37(3):360-367.
26 闫凯,沈汀,陈正超,等.基于深度学习的SSD模型尾矿库自动提取[J].中国科学院大学学报,2020,37(3):360-367.
[1] 陈敏,潘佳威,李江杰,徐璐,刘加敏,韩健,陈奕云. 结合VGGNet与Mask R-CNN的高分辨率遥感影像建设用地检测[J]. 遥感技术与应用, 2021, 36(2): 256-264.
[2] 胡乃勋,陈涛,甄娜,牛瑞卿. 基于卷积神经网络的面向对象露天采场提取[J]. 遥感技术与应用, 2021, 36(2): 265-274.
[3] 陈妮,应丰,王静,李健. 基于U-Net的高分辨率遥感图像土地利用信息提取[J]. 遥感技术与应用, 2021, 36(2): 285-292.
[4] 林娜,冯丽蓉,张小青. 基于优化Faster-RCNN的遥感影像飞机检测[J]. 遥感技术与应用, 2021, 36(2): 275-284.
[5] 金亚秋. 空间微波遥感研究与应用—丛书述评[J]. 遥感技术与应用, 2021, 36(1): 1-10.
[6] 李佳,辛晓洲,彭志晴,李小军. 地表蒸散发遥感产品比较与分析[J]. 遥感技术与应用, 2021, 36(1): 103-120.
[7] 刘良云,白雁,孙睿,牛振国. “全球生态系统碳循环关键参数立体观测与反演”项目概述与研究进展[J]. 遥感技术与应用, 2021, 36(1): 11-24.
[8] 熊育久,冯房观,方奕舟,邱国玉,赵少华,姚云军. 蒸散发遥感反演产品应用关键问题浅议[J]. 遥感技术与应用, 2021, 36(1): 121-131.
[9] 张茂,张霞,胡光成,王楠. 遥感干旱指数在洛川苹果干旱监测中的适用性分析[J]. 遥感技术与应用, 2021, 36(1): 187-197.
[10] 曹霸,凌成星. 基于GF-1遥感数据的若尔盖高寒沼泽湿地地上生物量与土壤有机碳密度估算[J]. 遥感技术与应用, 2021, 36(1): 229-236.
[11] 王昌博,李爱农,张晓荣,南希,边金虎,Kamran Muhammad. 基于遥感和GIS的中巴经济走廊多发展情景生态风险综合评价[J]. 遥感技术与应用, 2021, 36(1): 65-78.
[12] 蒋玲梅,崔慧珍,王功雪,杨建卫,王健,潘方博,苏旭,方西瑶. 积雪、土壤冻融与土壤水分遥感监测研究进展[J]. 遥感技术与应用, 2020, 35(6): 1237-1262.
[13] 岳珊娜,车涛,戴礼云,肖林,邓婕. 北半球及典型区雪深时空分布与变化特征[J]. 遥感技术与应用, 2020, 35(6): 1263-1272.
[14] 徐瑾昊,冯敏,王建邦,冉有华,祁元,杨联安,李新. 基于高分遥感数据和深度学习的石冰川自动提取研究[J]. 遥感技术与应用, 2020, 35(6): 1329-1336.
[15] 张永宏,曹海啸,阚希. 基于FY-4A/AGRI时空特征融合的新疆地区积雪判识[J]. 遥感技术与应用, 2020, 35(6): 1337-1347.