煤炭资源开发为我国经济发展做出了巨大贡献，但同时也给矿区生态环境造成严重破坏。为此，实时、准确、有效地监测煤矿开采引起的地表环境问题尤为重要。本文以宁夏石嘴山市惠农采煤沉陷区地表环境为研究对象，首先，基于收集资料和实地调查，分析了研究区的地质环境条件和煤炭开采概况，查明了存在的主要矿山地质环境问题，并对五时相不同源遥感数据进行预处理研究；其次，结合实地调查成果、实测光谱曲线对比分析、水质分析及影像特征分析，建立了采煤沉陷区地表环境要素解译标志，进行了基于面向对象和高光谱遥感的信息提取方法研究；最后，对修正后的不同时相遥感影像解译成果进行叠加分析，实现了采煤沉陷区地表环境时空演化监测。主要研究成果如下：
（1）采用PCA、NNDiffuse Pan Sharpening、Gram-Schmidt和Brovey Sharpening四种融合算法对研究区QuickBird、WorldView-2及GF-2遥感影像进行融合处理研究，结果表明：QuickBird遥感数据采用Gram Schmidt算法的融合效果最优，WorldView-2和GF-2遥感数据采用PCA算法的融合效果最优；对GF-5高光谱数据进行坏波段剔除、波长信息写入、辐射校正、几何校正等预处理，得到了符合精度要求的基础数据；
（2）通过野外踏勘、取样测试及光谱曲线测量，分析了各类环境要素的野外特征、光谱特征和遥感影像特征，建立了荒地、煤矸石、煤堆、耕地、其他建筑物、居民地、水体和植被覆盖区等采煤沉陷区典型地物的解译标志；
（3）基于面向对象的采煤沉陷区地表环境信息提取方法的研究表明，随机森林分类的总体精度（73.8%）和Kappa系数（平均0.657）略优于CART决策树分类（72.8%，0.641）；结合实地调查对四时相随机森林分类结果进行修正，分类精度有很大提高，得到的地表环境要素解译图总体平均精度为89.3%，平均Kappa系数为0.872；
（4）基于高光谱数据的光谱角匹配信息提取法研究得到：惠农采煤沉陷区植被以乔木、灌木为主，草类为辅，矿坑水主要分布在矿区塌陷坑附近，面积0.392km2；其他地表水主要为人工蓄水池；裸土分布较广，占研究区总面积21.5%；煤矸石、松散砂石和粉煤灰主要分布在煤矿建筑周围，面积较小；
（5）对研究区四时相地表环境要素解译图进行叠加分析得到：1979年到2003年，耕地和植被覆盖区面积大量减少，向荒地和居民地转化较明显，主要是城市化进程中的基础设施建设和水土流失导致；煤堆、水体、煤矸石等面积增长迅速，主要由荒地转化而来，说明该时期采煤活动严重破坏了采煤沉陷区的地表环境；2003年至2018年间，植被覆盖面积增长最快，煤矸石、荒地面积减少，经过10多年的矿山地质环境恢复治理，采煤沉陷区的地表环境得到改善，基本消除了地面塌陷和地裂缝等地质灾害隐患。遥感监测结果与矿山地质环境恢复治理前后的地表环境变化一致。

The exploitation of coal resources has made a huge contribution to China's economic development, but it has also caused serious damage to the ecological environment in mining areas. Therefore, it is particularly important to monitor the surface environmental problems caused by coal mining in a real-time, accurate and effective way. In this paper, the surface environment of Huinong coal mining subsidence area in Shizuishan City, Ningxia is taken as the research object. Firstly, based on the collected data and field investigation, the geological environment conditions and the general situation of coal mining in the study area are analyzed, the main mine geological environment problems are clarified. And the five-phase remote sensing data from different sources are preprocessed. Secondly, combining the results of field survey, comparative analysis of measured spectral curves, water quality analysis and image feature analysis, the interpretation signs of surface environmental elements in the coal mining subsidence areas were established and research on two information extraction methods based on object-oriented and hyperspectral remote sensing was carried out. Finally, the corrected interpretation results of remote sensing images with different phases are superimposed and analyzed to realize the monitoring of spatial and temporal evolution of the surface environment in the coal mining subsidence area. The results of the study are as follows:
(1)PCA, NNDiffuse Pan Sharpening, Gram-Schmidt and Brovey Sharpening are used to fuse the remote sensing images of QuickBird, WorldView-2 and GF-2 in the study area. It is shown that: The fusion result of QuickBird is the best by use of GS method while the fusion results of WorldView-2 and Gaofen-2 are the best by use of PCA method. In addition, the GF-5 hyperspectral data were preprocessed by bad band elimination, wavelength information writing, radiation correction and geometric correction, and the basic data meeting the accuracy requirements were obtained.
(2)Through the field investigation, sampling and spectral curve measurement, the field characteristics, spectral characteristics and remote sensing image characteristics of various environmental elements were analyzed, and the interpretation signs of typical features in coal mining subsidence areas, such as badlands, coal gangue, coal heaps, cultivated land, other construction, residential areas, vegetation cover areas, and water bodies were established .
(3)The research on the extraction method of surface environment information in coal mining subsidence area based on object-oriented method shows that the overall accuracy (73.8%) and kappa coefficient(0.657)of random forest classification are slightly better than those of CART classification(72.8%,0.641).Combined with the field survey to correct the four-temporal random forest classification results, the classification accuracy was greatly improved, and the overall average accuracy of the obtained surface environmental elements interpretation map was 89.3%, with an average kappa coefficient of 0.872.
(4)The spectral angle matching information extraction method based on hyperspectral data obtained that: the vegetation in the Huinong coal mining subsidence area is mainly trees and shrubs, supplemented by grasses; mine pit water is mainly distributed near the mine collapse pit, with an area of 0.392km2, other surface water is mainly artificial reservoirs; bare soil is widely distributed, accounting for 21.5% of the total area of the study area, while coal gangue, loose gravel and coal fly ash, which is mainly distributed around the coal mine buildings, with a small area.
(5)The overlay analysis of four-phase interpretation map of surface environmental factors in the study area can be concluded : From 1979 to 2003, the area of cultivated land and vegetation cover decreased sharply, and the conversion to badlands and residential areas was particularly obvious, which was mainly caused by the infrastructure construction and soil erosion in the process of urbanization. Coal heaps, water bodies, and coal gangue exhibited a trend of rapid growth, mainly converted from badlands, indicating that Mining activities have caused severe damage to the environment of the mining area. From 2003 to 2018, the vegetation-covered area exhibited the fastest growth rate, while the area of coal gangue and badlands significantly reduced. After more than 10 years of mine geological environment renovation, the surface environment of the coal mining subsidence area greatly improved, and the hidden dangers of geological disasters such as ground subsidence and ground fissures have been basically eliminated.The surface environment before and after the renovation of the mine geological environment in the study area is consistent with the results of remote sensing monitoring.

[1] 赵国君,赵祺彬,申文金.废弃矿山地质环境治理模式探讨[J].资源与产业,2018, 20(5): 77-82.

[2] 中华人民共和国自然资源部.中国矿产资源报告[M].北京:地质出版社, 2021.

[3] 康红普,徐刚,王彪谋,等.我国煤炭开采与岩层控制技术发展40a及展望[J].采矿与岩层控制工程学报,2019,1(2):7-39.

[4] Kuehn F, King T V V, Hoerig B,et al.Remote Sensing for Site Characterization [M]. Springer, 2000: 10-20.

[5] Culshaw M G, Nathanail C P,Leeks G J L, et al. The role of web-based environmental information in urban planning-the environmental information system for planners[J].The Science of the total environment, 2006, 360(1): 233-245.

[6] Charou E, Stefouli M, Dimitrakopoulos D, et al.Using Remote Sensing to Assess Impact of Mining Activities on Land and Water Resources[J]. Mine water and the environment, 2010, 29(1): 45-52.

[7] Malaviya S, Munsi M, Oinam G, et al. Landscape approach for quantifying land use land cover change (1972-2006) and habitat diversity in a mining area in Central India (Bokaro, Jharkhand)[J]. Environmental Monitoring and Assessment, 2010, 170(1-4): 215-229.

[8] Li X, He Z, Jiang L,et al.Research on Remote Sensing Dynamic Monitoring of Ecological Resource Environment Based on GIS[J].Wireless Personal Communications,2018,102(4): 2941-2953.

[9] 吴德礼,朱申红,王铮.国内外矿山尾矿综合利用现状与思考[J].青岛建筑工程学院学报, 2001(4): 84-87.

[10] Mularz S. Satellite and Airborne Remote Sensing Data for Monitoring of an Open-cast Mine[J]. International Archives of Photogrammetry and Remote Sensing, 1998, 32: 395-402.

[11] Cloutis E A. Review Article Hyperspectral geological remote sensing: evaluation of analytical techniques[J]. International journal of remote sensing, 1996, 17(12): 2215-2242.

[12] Ferrier G. Application of Imaging Spectrometer Data in Identifying Environmental Pollution Caused by Mining at Rodaquilar, Spain[J]. Remote sensing of environment, 1999, 68(2): 125-137.

[13] Roy P, Guha A, Kumar K V. An approach of surface coal fire detection from ASTER and Landsat-8 thermal data: Jharia coal field, India[J]. International Journal of Applied Earth Observation and Geoinformation, 2015, 39: 120-127.

[14] Mishra R K, Bahuguna P P, Singh V K. Detection of coal mine fire in Jharia Coal Field using Landsat-7 ETM+ data[J]. International Journal of Coal Geology, 2011, 86(1): 73-78.

[15] Zou X,Su W,Chen Y, et al. Estimation of Fractional Vegetation Cover in Opencast Coal Mine Dump Area Using Landsat Tm Data[J]. Sensor Letters, 2010, 8(1): 81-88.

[16] Mansor S B, Cracknell A P, Shilin B V, et al. Monitoring of underground coal fires using thermal infrared data[J]. International journal of remote sensing, 1994, 15(8): 1675-1685.

[17] Prakash A, Gens R,Vekerdy, Z. Monitoring coal fires using multi-temporal night-time thermal images in a coalfield in north-west China[J].International Journal of Remote Sensing, 1999, 20(14), 2883-2888.

[18] Xin-Zhi L, Ping W, Yan-Bin Z. Application of SPOT 5 data fusion on investigating the ecological environment of mining area[C]. IEEE, 2009.

[19] Harbi H, Madani A. Utilization of SPOT 5 data for mapping gold mineralized diorite–tonalite intrusion, Bulghah gold mine area, Saudi Arabia[J]. Arabian Journal of Geosciences, 2014, 7(9): 3829-3838.

[20] Mezned N,Mechrgui N,Abdeljaouad S.Enhanced Mapping and Monitoring of Mine Tailings Based on Landsat Etm+ and Spot 5 Fusion in the North of Tunisia[C].Geoscience and Remote Sensing Symposium,2014: 4800-4803.

[21] 陈述彭,周上盖.腾冲航空遥感试验回顾[J].遥感信息,1986(2):11-12.

[22] 管海晏.遥感技术在煤田地质工作中的应用研究[J].地质学报,1989,(1):36-49.

[23] 康高峰.应用遥感技术调查研究煤层自燃灾害[J].国土资源遥感, 1992(4): 34-40.

[24] 雷利卿,岳燕珍,孙九林,等.遥感技术在矿区环境污染监测中的应用研究[J].环境保护,2002(2):33-36.

[25] 陈旭.遥感解译分析矿山开发对生态环境的影响[J].资源调查与环境,2004(1):13-17.

[26] 丁爱华,鲁明星,吕冰.基于遥感技术的采煤塌陷区时空演变研究[J].安徽农业科学,2010,38(7): 3723-3725.

[27] 毕晓佳.遥感技术监测采煤塌陷区的应用研究[J].煤炭技术,2015,34(4):111-112.

[28] 李峰,梁汉东,米晓楠,等. Landsat8卫星影像的多子区决策树土地覆被分类方法（英文）[J].红外与激光工程,2015, 44(7): 2224-2230.

[29] 刘凤梅,王少军,张志,等.铁铜多金属矿开采矿区及其固体废弃物信息提取方法研究[J].金属矿山, 2010(5):168-170.

[30] 刘琼,聂洪峰,吕杰堂,等.GIS在矿产资源开发状况遥感动态监测中的应用[J].国土资源遥感,2005(1):61-65.

[31] 雷兵,甘宇航,李兰,等.矿区环境动态监测系统建设研究[J].遥感信息, 2014(4):103-106,111.

[32] 况顺达,赵震海.SPOT5在矿山监测中的应用[J].地质与勘探,2005(3):79-82.

[33] 褚进海,彭鹏,李郑,等.矿山遥感监测中QuickBird数据融合方法研究[J].国土资源遥感,2009(3):107-109.

[34] 张焜,马世斌,刘丽萍.基于SPOT5数据的盐湖矿产开发及矿山环境遥感监测[J].国土资源遥感,2012(3):146-153.

[35] 王俊芳,曾新超.高分辨率影像在矿山环境遥感监测中的应用研究[J].测绘,2015,38(5):220-223.

[36] 安志宏,聂洪峰,王昊,等.ZY-102C星数据在矿山遥感监测中的应用研究与分析[J].国土资源遥感,2015, 27(2):174-182.

[37] 汪洁,殷亚秋,于航,等.基于RS和GIS的浙江省矿山地质环境遥感监测[J].国土资源遥感, 2020,32(1): 232-236.

[38] 周英杰,李思佳,魏红艳,等.“高分一号02、03、04星”卫星数据在矿山开发环境遥感监测中的初步应用评价[J].测绘与空间地理信息,2020,43(3):73-75.

[39] 尚慧.宁夏矿山地质环境评价与动态监测分析[D].西安:长安大学,2013.

[40] 孙冰.基于GIS的宁夏石嘴山市地质灾害危险性评价[D].西安:长安大学, 2013.

[41] 陈越平.宁夏石嘴山采煤沉陷区治理及景观设计研究[D].西安:西安建筑科技大学,2015.

[42] 潘腾.高分二号卫星的技术特点[J].中国航天, 2015(1): 3-9.

[43] 潘腾,关晖,贺玮.“高分二号”卫星遥感技术[J].航天返回与遥感,2015, 36(4):16-24.

[44] 曹若臣,荆林海,向中林,等.Hyperion高光谱遥感在新疆西南果勒地区蚀变信息提取的应用[J].测绘与空间地理信息,2018,41(9): 75-78.

[45] Song W, Song W,Gu H, et al.Progress in the Remote Sensing Monitoring of the Ecological Environment in Mining Areas[J].International Journal of Environmental Research and Public Health,2020,17(6):1846-1863.

[46] Xu J,Zhao H, Yin P,et al. Remote sensing classification method of vegetation dynamics based on time series Landsat image: a case of opencast mining area in China[J]. EURASIP Journal on Image and Video Processing,2018, 2018(1):113-123.

[47] 朱述龙,钱曾波.遥感影像镶嵌时拼接缝的消除方法[J].遥感学报,2002(3):183-187.

[48] 陈春祥.多源多时相遥感影像镶嵌方法探讨[J].微计算机信息,2009,25(21):241-243.

[49] 张安定.遥感原理与应用题解[M].北京：科学出版社,2017:241.

[50] 李宏杰,戴福初,许领,等.地质灾害调查中ETM+与SPOT5 Pan影像融合与评价[J].国土资源遥感,2008(1):43-45.

[51] 刘锟,付晶莹,李飞.高分一号卫星4种融合方法评价[J].遥感技术与应用,2015,30(5):980-986.

[52] 肖昶,余晓敏,韩逸飞.高分二号卫星影像融合技术研究[J].地理空间信息,2018,16(6):13-16.

[53] 徐佳,关泽群,何秀凤,等.基于传感器光谱特性的全色与多光谱图像融合[J].遥感学报,2009,13(1):97-102.

[54] Wang X, Sun X, Fang Y. Genetic algorithm solution for multi-period two-echelon integrated competitive/uncompetitive facility location problem[J].Asia-Pacific journal of operational research,2008,25(1): 33-56.

[55] 周为峰,曹利,李小恕,等.沿海牡蛎养殖的WorldView-2影像融合方法评价[J].遥感技术与应用, 2018,33(1):103-109.

[56] 韩闪闪,李海涛,顾海燕.高分辨率遥感影像融合研究[J].测绘科学,2009,34(5):60-62.

[57] 尤畅,万里萍,蔡伟.浅谈地理国情普查影像制作、影像融合算法[J].测绘与空间地理信息, 2014,37(6):168-170.

[58] Ji L,Gallo K.An Agreement Coefficient for Image Comparison[J].Photogrammetric Engineering & Remote Sensing,2006,72(7):823-833.

[59] 赵瑞,崔希民,刘超.GF-5高光谱遥感影像的土壤有机质含量反演估算研究[J].中国环境科学,2020,40(8):3539-3545.

[60] 宫鹏,浦瑞良.高光谱遥感及其应用[M].北京:高等教育出版社,2000:200.

[61] 马娜,胡云锋,庄大方,等.基于最佳波段指数和J-M距离可分性的高光谱数据最佳波段组合选取研究—以环境小卫星高光谱数据在东莞市的应用为例[J].遥感技术与应用,2010,25(3): 358-365.

[62] 赵庆展,刘伟,尹小君,等.基于无人机多光谱影像特征的最佳波段组合研究[J].农业机械学报,2016,47(3):242-248.

[63] Drăguţ L, Csillik O,Eisank C,et al.Automated parameterisation for multi-scale image segmentation on multiple layers[J].ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 88: 119-127.

[64] Drǎguţ L, Tiede D, Levick S R. ESP:a tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data[J].International journal of geographical information science: IJGIS, 2010,24(6):859-871.

[65] Gerçek D,Toprak V,Strobl J.Object-based classification of landforms based on their local geometry and geomorphometric context[J].International journal of geographical information science:IJGIS,2011, 25(6):1011-1023.

[66] Benz U C,Hofmann P,Willhauck G,et al.Multi-resolution object-oriented fuzzy analysis of remote sensing data for GIS-ready information[J].ISPRS Journal of Photogrammetry and Remote Sensing, 2004,58(3-4): 239-258.

[67] 袁定波,刘成林,汪国斌.面向对象的矿区信息提取方法的应用与研究[J].遥感信息,2013,28(2): 110-115.

[68] 张俊,于庆国,朱晓东,等.面向对象的高分辨率影像特征选择研究[J]. 测绘科学,2011,36(3): 141-143.

[69] 王贺,陈劲松,余晓敏.面向对象分类特征优化选取方法及其应用[J].遥感学报,2013,17(4): 816-829.

[70] 白韬.面向对象的GF-2遥感影像多层次分类方法研究[D].长春:吉林大学,2020.

[71] 陈金丽.面向对象的最邻近算法研究与实现[D].北京:中国地质大学,2009.

[72] 陈鸣,胡慧萍,杨敏华.基于面向对象特征提取的BP神经网络分类—以武陵源地区为例[J].现代测绘,2017, 40(3):17-20.

[73] 霍光杰,胡乃勋,陈涛,等.融合支持向量机和面向对象方法的矿区土地利用信息提取[J].河南理工大学学报(自然科学版),2021,40(2):70-75.

[74] 牟昱璇,邬明权,牛铮,等.南方地区复杂条件下的耕地面积遥感提取方法[J].遥感技术与应用,2020,35(5):1127-1135.

[75] 孙建伟,王超,王娜,等.基于CART决策树的ZY-3卫星遥感数据土地利用分类监测[J].华中师范大学学报(自然科学版),2016,50(6): 937-943.

[76] 徐乔,孟凡利,余绍淮.基于面向对象随机森林分类模型的滑坡遥感解译[J].中外公路,2019,39(3):30-34.

[77] 翟天林,金贵,邓祥征,等.基于多源遥感影像融合的武汉市土地利用分类方法研究[J].长江流域资源与环境,2016,25(10):1594-1602.

[78] 刘星雷,鲁铁定,龚循强.基于GF-2遥感影像的面向对象分类方法比较研究[J].江西科学,2019,37(6):913-916.

[79] 赵萍,傅云飞,郑刘根,等.基于分类回归树分析的遥感影像土地利用/覆被分类研究[J].遥感学报, 2005(6):708-716.

[80] 张俊,朱国龙,李妍.面向对象高分辨率影像信息提取中的尺度效应及最优尺度研究[J].测绘科学,2011, 36(2):107-109.

[81] Walston L J, Cantwell B L, Krummel J R.Quantifying spatiotemporal changes in a sagebrush ecosystem in relation to energy development[J]. Ecography, 2009, 32(6): 943-952.

[82] Dubovyk O, Menz G, Conrad C,et al. Spatio-temporal analyses of cropland degradation in the irrigated lowlands of Uzbekistan using remote-sensing and logistic regression modeling[J].Environmental Monitoring and Assessment,2013,185(6): 4775-4790.

[83] Moradi F,Kaboli H S, Lashkarara B. Projection of future land use/cover change in the Izeh-Pyon Plain of Iran using CA-Markov model[J]. Arabian Journal of Geosciences, 2020,13(19):998-1015.

[84] Masek J G, Vermote E F, Saleous N E, et al. A Landsat Surface Reflectance Dataset for North America,1990-2000[J]. IEEE Geoscience and Remote Sensing Letters,2006,3(1):68-72.

[85] 沈聪颖,甘淑,李新澳,等.基于GF-5高光谱特征分析的水体提取方法研究[J].全球定位系统,2021,46(5):117-122.

[86] 王润生,杨苏明,阎柏琨.成像光谱矿物识别方法与识别模型评述[J].国土资源遥感, 2007(1):1-9.

[87] Chein-I C. An information-theoretic approach to spectral variability, similarity,and discrimination for hyperspectral image analysis[J]. IEEE Transactions on Information Theory, 2000,46(5):1927-1932.

[88] Van der Meer F.The effectiveness of spectral similarity measures for the analysis of hyperspectral imagery[J]. International Journal of Applied Earth Observation and Geoinformation, 2006, 8(1): 3-17.

[89] 陶秋香.植被高光谱遥感分类方法研究[J].山东科技大学学报(自然科学版), 2007(5):61-65.

[90] 许宁,胡玉新,雷斌,等.基于改进光谱特征拟合算法的高光谱数据矿物信息提取[J].光谱学与光谱分析, 2011,31(6):1639-1643.

[91] 温兴平,胡光道,杨晓峰.基于光谱特征拟合的高光谱遥感影像植被覆盖度提取[J].地理与地理信息科学,2008(1): 27-30.

[92] 王晋年,张兵,刘建贵,等.以地物识别和分类为目标的高光谱数据挖掘[J].中国图象图形学报,1999(11): 59-66.

[93] 甘甫平,刘圣伟,周强.德兴铜矿矿山污染高光谱遥感直接识别研究[J].地球科学, 2004(1):119-126.

[94] 陈云海,穆亚超,颉耀文.基于Landsat遥感影像的黑河干流中游湿地信息提取[J].兰州大学学报(自然科学版),2016,52(5):587-592.

[95] 陈超,江涛,岳远平.监督分类和目视修改相结合在高分辨率遥感影像中的应用[J].国土资源信息化, 2009(5):37-40.