Recognition of land use on open-pit coal mining area based on DeepLabv3+ and GF-2 high-resolution images

Authors

ZHANG Chengye, College of Geosciences and Surveying Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China; State Key Laboratory of Coal Resources and Safe Mining, China University of Mining and Technology (Beijing), Beijing 100083, ChinaFollow
LI Feiyue, College of Geosciences and Surveying Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China
LI Jun, College of Geosciences and Surveying Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China; State Key Laboratory of Coal Resources and Safe Mining, China University of Mining and Technology (Beijing), Beijing 100083, ChinaFollow
XING Jianghe, College of Geosciences and Surveying Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China
YANG Jinzhong, China Aero Geophysical Survey and Remote Sensing Center for Natural Resources, Beijing 100083, China
GUO Junting, State Key Laboratory of Water Resource Protection and Utilization in Coal Mining, Beijing 102209, China
DU Shouhang, College of Geosciences and Surveying Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China

Abstract

A highly efficient means is provided by remote sensing and deep learning to keep tracking of land use in open-pit coal mining area. Based on the high–resolution images from the domestic GF-2 satellite, a DeepLabv3+ model was utilized to achieve recognition of land use on open-pit coal mining area. In addition, a comparison was made among Deeplabv3+, U-Net, FCN, Random Forest, Support Vector Machine, and Maximum Likelihood Method. Firstly, samples data from high-resolution images were produced and sensitivity tests were conducted to determine the optimal cutting size and mode of the sample. Then, the deep neural network model (DeepLabv3+) was trained for conducting experiments of recognition of land use. Finally, the recognition results of different methods were compared. The results show that the optimal cutting size of the sample on the open-pit coal mining of the study area is 512 pixel×512 pixel. The optimal cutting mode of the sample is random cropping. The overall accuracy and Kappa coefficient of the DeepLabv3+ for recognition of land use on open-pit coal mining area are 80.10% and 0.73, respectively, which are better than the recognition accuracy of the U-Net, FCN, Random Forest, Support Vector Machine and Maximum Likelihood Method. The DeepLabv3+ is kept in the same order of magnitude as the above five methods. The feasibility of DeepLabv3+ model and GF-2 images in recognition of land use on open-pit coal mining area is verified, which is important for monitoring and restoration of eco-environment on open-pit coal mining area.

Keywords

open-pit coal mining area, land use, high resolution image, deep learning, neural network, GF-2 satellite, automatic recognition, recognition accuracy

DOI

10.12363/issn.1001-1986.22.01.0029

Recommended Citation

ZHANG Chengye, LI Feiyue, LI Jun, et al. (2022) "Recognition of land use on open-pit coal mining area based on DeepLabv3+ and GF-2 high-resolution images," Coal Geology & Exploration: Vol. 50: Iss. 6, Article 11.
DOI: 10.12363/issn.1001-1986.22.01.0029
Available at: https://cge.researchcommons.org/journal/vol50/iss6/11

Reference

[1] 刘磊,郭二民,李忠华,等. 加强“十四五”露天煤矿开采环境管理的建议[J]. 中国煤炭,2021,47(10):61−66. LIU Lei,GUO Ermin,LI Zhonghua,et al. Suggestions on strengthening the environmental management of open−pit coal mines during the 14th Five–Year Plan[J]. China Coal,2021,47(10):61−66.

[2] 王猛,马如英,代旭光,等. 煤矿区碳排放的确认和低碳绿色发展途径研究[J]. 煤田地质与勘探,2021,49(5):63−69. WANG Meng,MA Ruying,DAI Xuguang,et al. Confirmation of carbon emissions in coal mining areas and research on low–carbon green development path[J]. Coal Geology & Exploration,2021,49(5):63−69.

[3] 董书宁,刘再斌,程建远,等. 煤炭智能开采地质保障技术及展望[J]. 煤田地质与勘探,2021,49(1):21−31. DONG Shuning,LlU Zaibin,CHENG Jianyuan,et al. Technologies and prospect of geological guarantee for intelligent coal mining[J]. Coal Geology & Exploration,2021,49(1):21−31.

[4] 王振庆,周艺,王世新,等. IEU–Net高分辨率遥感影像房屋建筑物提取[J]. 遥感学报,2021,25(11):2245−2254. WANG Zhenqing,ZHOU Yi,WANG Shixin,et al. House building extraction from high–resolution remote sensing images based on IEU–Net[J]. National Remote Sensing Bulletin,2021,25(11):2245−2254.

[5] 刘戈,姜小光,唐伯惠. 特征优选与卷积神经网络在农作物精细分类中的应用研究[J]. 地球信息科学学报,2021,23(6):1071−1081. LIU Ge,JIANG Xiaoguang,TANG Bohui. Application of feature optimization and convolutional neural network in crop classification[J]. Journal of Geo−information Science,2021,23(6):1071−1081.

[6] 石军南,李和顺,刘晓农,等. 面向对象分类方法在森林采伐遥感监测中的应用[J]. 中南林业科技大学学报,2010,30(11):6−10. SHI Junnan,LI Heshun,LIU Xiaonong,et al. Application of classification object–oriented to forest cutting monitoring based on remote sensing[J]. Journal of Central South University of Forestry & Technology,2010,30(11):6−10.

[7] 李军,彭苏萍,张成业,等. 矿区生态环境定量遥感监测评价技术框架与应用[J]. 矿业科学学报,2022,7(1):9−25. LI Jun,PENG Suping,ZHANG Chengye,et al. Quantitative remote sensing–based monitoring and evaluation of the ecological environment in mining areas:Technology framework and application[J]. Journal of Mining Science and Technology,2022,7(1):9−25.

[8] 张成业，李军，雷少刚，等. 矿区生态环境定量遥感监测研究进展与展望[J/OL]. 金属矿山，2021：1–38 [2021-12-15]. http://kns.cnki.net/kcms/detail/34.1055.TD.20210915.1042.002.html.

ZHANG Chengye，LI Jun，LEI Shaogang，et al. Progress and prospect of the quantitative remote sensing for monitoring the eco–environment in mining area[J/OL]. Metal Mine，2021：1–38 [2021-12-15]. http://kns.cnki.net/kcms/detail/34.1055.TD.20210915.1042.002.html.

[9] 李镇,张岩,杨松,等. QuickBird影像目视解译法提取切沟形态参数的精度分析[J]. 农业工程学报,2014,30(20):179−186. LI Zhen,ZHANG Yan,YANG Song,et al. Error assessment of extracting morphological parameters of bank gullies by manual visual interpretation based on QuickBird imagery[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE),2014,30(20):179−186.

[10] 殷亚秋,杨金中,汪洁,等. 长江经济带废弃矿山占损土地遥感调查与生态修复对策[J]. 国土资源遥感,2020,32(2):170−176. YIN Yaqiu,YANG Jinzhong,WANG Jie,et al. Remote sensing survey of land occupied and damaged by abandoned mines along the Yangtze River Economic Belt and research on ecological remediation countermeasures[J]. Remote Sensing for Land and Resources,2020,32(2):170−176.

[11] 姚维岭,荆青青,周英杰,等. 基于遥感动态监测的山东省矿山地质环境恢复治理典型模式分析[J]. 矿产勘查,2015,6(5):627−634. YAO Weiling,JING Qingqing,ZHOU Yingjie,et al. Analysis on typical pattern of mine geological environment restoration and governance based on remote sensing dynamic monitoring in Shandong[J]. Mineral Exploration,2015,6(5):627−634.

[12] 汪洁,刘小杨,杨金中,等. 基于国产高空间分辨率卫星数据的浙江省矿山环境恢复治理典型模式分析[J]. 国土资源遥感,2020,32(3):216−221. WANG Jie,LIU Xiaoyang,YANG Jinzhong,et al. Typical model analysis of mine geological environment restoration and management in Zhejiang Province based on domestic high−resolution satellite data[J]. Remote Sensing for Land & Resources,2020,32(3):216−221.

[13] 李治,杨晓梅,孟樊,等. 物候特征辅助下的随机森林宏观尺度土地覆盖分类方法研究[J]. 遥感信息,2013,28(6):48−55. LI Zhi,YANG Xiaomei,MENG Fan,et al. LULC classification based on Random Forest with the aid of phenological features[J]. Remote Sensing Information,2013,28(6):48−55.

[14] 肖云华,王清,陈剑平,等. 基于粗糙集和支持向量机的融合算法在岩体质量评价中的应用[J]. 煤田地质与勘探,2008,36(6):49−53. XIAO Yunhua,WANG Qing,CHEN Jianping,et al. Application of data fusion in evaluation of engineering quality of rock mass based on rough sets and support vector machine[J]. Coal Geology & Exploration,2008,36(6):49−53.

[15] 刘涛,孙忠林,孙林. 基于最大似然法的遥感图像分类技术研究[J]. 福建电脑,2010,26(1):7−8. LIU Tao,SUN Zhonglin,SUN Lin. Research on remote sensing image classification based on maximum likelihood method[J]. Journal of Fujian Computer,2010,26(1):7−8.

[16] 高文龙,苏腾飞,张圣微,等. 矿区地物分类及土地利用/覆盖变化动态监测:以海流兔流域为例[J]. 国土资源遥感,2020,32(3):232−239. GAO Wenlong,SU Tengfei,ZHANG Shengwei,et al. Classification of objects and LUCC dynamic monitoring in mining area:A case study of Hailiutu watershed[J]. Remote Sensing for Land & Resources,2020,32(3):232−239.

[17] 董佰山,杨耘,崔琴芳. 基于高分二号的矿区土地利用精细分类与分析[J]. 甘肃科学学报,2019,31(1):81−85. DONG Baishan,YANG Yun,CUI Qinfang. Detailed classification and analysis based on separator No.2 land use in mining area[J]. Journal of Gansu Sciences,2019,31(1):81−85.

[18] 尹宝才,王文通,王立春. 深度学习研究综述[J]. 北京工业大学学报,2015,41(1):48−59. YIN Baocai,WANG Wentong,WANG Lichun. Review of deep learning[J]. Journal of Beijing University of Technology,2015,41(1):48−59.

[19] 冯海林,胡明越,杨垠晖,等. 基于树木整体图像和集成迁移学习的树种识别[J]. 农业机械学报,2019,50(8):235−242. FENG Hailin,HU Mingyue,YANG Yinhui,et al. Tree species recognition based on overall tree image and ensemble of transfer learning[J]. Transactions of the Chinese Society for Agricultural Machinery,2019,50(8):235−242.

[20] CHEN Yangyang,MING Dongping,LYU Xianwei. Superpixel based land cover classification of VHR satellite image combining multi−scale CNN and scale parameter estimation[J]. Earth Science Informatics,2019,12(3):341−363.

[21] MARTINI M,MAZZIA V,KHALIQ A,et al. Domain−adversarial training of self−attention−based networks for land cover classification using multi−temporal sentinel−2 satellite imagery[J]. Remote Sensing,2021,13(13):2564.

[22] 陈文超,刘达伟,魏新建,等. 基于地震资料有效信息约束的深度网络无监督噪声压制方法[J]. 煤田地质与勘探,2021,49(1):249−256. CHEN Wenchao,LlU Dawei,WEI Xinjian,et al. Unsupervised noise suppression method for depth network seismic data based on prior information constraint[J]. Coal Geology & Exploration,2021,49(1):249−256.

[23] 张成业,邢江河,李军,等. 基于U–Net网络和GF–6影像的尾矿库空间范围识别[J]. 自然资源遥感,2021,33(4):252−257. ZHANG Chengye,XING Jianghe,LI Jun,et al. Recognition of the spatial scopes of tailing ponds based on U–Net and GF–6 images[J]. Remote Sensing for Natural Resources,2021,33(4):252−257.

[24] 胡乃勋,陈涛,甄娜,等. 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. HU Naixun,CHEN Tao,ZHEN Na,et al. Object–oriented open pit extraction based on convolutional neural network[J]. Remote Sensing Technology and Application,2021,36(2):265−274.

[25] BALANIUK R,ISUPOVA O,REECE S. Mining and tailings dam detection in satellite imagery using deep learning[J]. Sensors,2020,20(23):6936.

[26] 李宏韬,李娜,鲜明睿,等. 基于MODIS的神东矿区植被覆盖时空变化分析[J]. 林业建设,2021(5):47−52. LI Hongtao,LI Na,XIAN Mingrui,et al. Analysis on spatio–temporal variations of vegetation cover in Shendong mining area base on MODIS[J]. Forestry Construction,2021(5):47−52.

[27] 刘英，魏嘉莉，岳辉，等. 神东矿区土壤侵蚀时空特征及驱动力分析[J/OL]. 测绘科学，2021：1–14 [2021-12-15]. http://kns.cnki.net/kcms/detail/11.4415.P.20211103.1421.002.html.

LIU Ying，WEI Jiali，YUE Hui，et al. Analysis on temporal and spatial characteristics and driving factors of soil erosion in Shendong mining area[J/OL]. Science of Surveying and Mapping，2021：1–14 [2021-12-15]. http://kns.cnki.net/kcms/detail/11.4415.P.20211103.1421.002.html.

[28] 薛明,韦波,李景文,等. 优化BP神经网络的多特征融合遥感影像分类方法[J]. 测绘科学,2021,46(11):47−55. XUE Ming,WEI Bo,LI Jingwen,et al. Multi feature fusion remote sensing image classification method based on optimized BP neural network[J]. Science of Surveying and Mapping,2021,46(11):47−55.

[29] 王芳,杨武年,王建,等. 国产高分卫星像素级影像融合及其质量评价[J]. 测绘科学,2021,40(8):73−80. WANG Fang,YANG Wunian,WANG Jian,et al. Chinese high-resolution satellite pixel level image fusion and its quality evaluation[J]. Science of Surveying and Mapping,2021,40(8):73−80.

[30] CHEN Liangchieh,ZHU Yukun,PAPANDREOU G,et al. Encoder−Decoder with atrous separable convolution for semantic image segmentation[J]. In Proceedings of the 15th European Conference on Computer Vision (ECCV),2018(11211):833−851.

[31] 王华俊，葛小三. 一种轻量级的DeepLabv3+遥感影像建筑物提取方法[J/OL]. 自然资源遥感，2021：1−8 [2021-12-15]. http://kns.cnki.net/kcms/detail/10.1759.P.20211203.1351.008.html.

WANG Huajun，GE Xiaosan. Lightweight DeepLabv3 plus building extraction method from remote sensing image[J/OL]. Remote Sensing for Natural Resources，2021：1–8 [2021-12-15]. http://kns.cnki.net/kcms/detail/10.1759.P.20211203.1351.008.html.

[32] 李海涛,戴莉莉,顾海燕,等. 样本尺寸对遥感影像FCN训练模型的影响分析[J]. 测绘科学,2019,44(6):133−137. LI Haitao,DAI Lili,GU Haiyan,et al. Analysis of sample size influence on FCN training model in remote sensing image[J]. Science of Surveying and Mapping,2019,44(6):133−137.

[33] RONNEBERGER O，FISCHER P，BROX T. U–Net：Convolutional networks for biomedical image segmentation[C]// International Conference on Medical Image Computing and Computer–Assisted Intervention Cham：Springer，2015：234–241.

[34] SHELHAMER E,LONG J,DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(4):640−651.

[35] 邓军,雷昌奎,曹凯,等. 采空区煤自燃预测的随机森林方法[J]. 煤炭学报,2018,43(10):2800−2808. DENG Jun,LEI Changkui,CAO Kai,et al. Random forest method for predicting coal spontaneous combustion in gob[J]. Journal of China Coal Society,2018,43(10):2800−2808.

[36] BREIMAN L. Random forests[J]. Machine Language,2001,45(1):5−32.

[37] 张松兰. 支持向量机的算法及应用综述[J]. 江苏理工学院学报,2016,22(2):14−17. ZHANG Songlan. A survey of improved algorithm and application on support vector machine[J]. Journal of Jiangsu University of Technology,2016,22(2):14−17.

[38] 曾爱平,张嘉玮,任恩明,等. 基于VMD和SVM的煤厚预测方法研究[J]. 煤田地质与勘探,2021,49(6):243−250. ZENG Aiping,ZHANG Jiawei,REN Enming,et al. Research on the coal thickness prediction method based on VMD and SVM[J]. Coal Geology & Exploration,2021,49(6):243−250.

[39] ARAI K. Maximum likelihood classification based on classified result of boundary mixed pixels for high spatial resolution of satellite images[J]. International Journal of Advanced Computer Science and Applications (IJACSA),2020,11(9):24−30.

[40] ORIMOLOYE I R,OLOLADE O O. Spatial evaluation of land–use dynamics in gold mining area using remote sensing and GIS technology[J]. International Journal of Environmental Science and Technology,2020,17(11):4465−4480.

[41] 张开放,苏华友,窦勇. 一种基于混淆矩阵的多分类任务准确率评估新方法[J]. 计算机工程与科学,2021,43(11):1910−1919. ZHANG Kaifang,SU Huayou,DOU Yong. A new multi–classification task accuracy evaluation method based on confusion matrix[J]. Computer Engineering & Science,2021,43(11):1910−1919.