【作者】 李新芝；

【导师】 王萍；

【作者基本信息】 山东科技大学 ， 摄影测量与遥感， 2010， 硕士

【摘要】 由于煤炭开采而引起的矿区生态环境污染和破环问题不断涌现,且呈现日益增长的趋势,矿区环境污染灾害的调查和治理已成为矿区可持续发展亟待解决的重要课题。植被信息是指示生态环境的重要参数,矿区植被信息的有效提取与监测是矿区生态环境监测和治理的重要组成部分。本文以肥城矿区为研究区,利用多源遥感数据,根据植被的波谱特征和遥感图像特征,从空间和时间序列上动态分析了矿区植被生长变化状况及其与周围景观因素的关系,从而间接地反映出矿区扩展及生态环境状况,为矿区环境污染监测提供技术支持。本文主要进行了以下几方面的研究：1.针对矿区生态环境监测中的实际需求,本文充分利用SPOT-5数据2.5m分辨率的全色波段,运用小波变换、主成分分析等融合方法,分析了其在矿区生态环境要素调查中的应用,并将其作为矿区植被信息提取分析的基础数据。2.为有效提取与监测矿区植被信息,综合运用缨帽变换、植被与土壤相关性分析、支持向量机分类等方法提取了矿区植被信息,并制作了植被等级分布图,确定了不同污染程度的植被覆盖面积；运用光谱变化向量分析方法对矿区2000年-2009年的生态环境的变化信息进行了监测,确定了植被信息的动态变化。3.针对矿区植被与生态环境要素间相关分析的研究相对不足,本文反演了矿区地表温度,确定了植被覆盖度,从而分析了植被覆盖度与地表温度、大气污染及不同下垫面类型的关系,为决策部门制定相关生态措施提供了科学依据。4.以综合指数法为基础,通过计算生物丰度指数、植被覆盖指数、水网密度指数、土地退化指数、环境质量指数综合分析了矿区生态环境状况,给出了各分矿区的生态环境状况指数等级,这不仅丰富和完善了生态环境影响评价的实践应用案例,还为矿区节能减排提供重要的技术支撑。5.通过野外调查,结合航拍数据对监测的矿区植被信息进行精度评价,根据分析结果及矿区生态环境状况指数提出了相关措施。实验结果表明,综合利用3S技术,采用数据融合、缨帽变换、支持向量机分类、光谱变化向量分析等方法,结合矿区实际获取的数据,能有效地提取矿区植被信息,提高矿区生态环境监测水平,在同类型的矿区生态环境遥感动态监测方面具有很大的推广应用价值。更多还原

【Abstract】 The environmental pollution and ecological problems appeared continuously with production of coal mining, and these were becoming more and more serious. It had become an important subject research to investigate and manage the environmental pollution and disaster for sustainable development in mining area. The vegetation information is an important parameter that indicates the state of ecological environment. So the effective extraction and monitor vegetation information of mine is an important component of monitoring and managing the ecological environment. In this paper, taking Feicheng mining as the study area, using multi-source remote sensing data, according to the characteristics of vegetation spectral and remote sensing image, the status of vegetation growth and the relationship with the surrounding landscape were investigated based on dynamic analysis from the space and time-series. The results reflect indirectly the mining expansion and ecological environment, and provide technical support for monitoring pollution of the mining area.In this paper, the studies had been made as following:1. In cognizance of the actual demand of monitoring the ecological environment, this paper makes full use of SPOT-5 panchromatic data that resolution at 2.5m, using the fusion method such as wavelet transform, principal component analysis, analyzing the elements of the ecological environment in mining areas and make it as the basic data that extract and analyze the vegetation information of mining.2. In order to extract and monitor the vegetation information of mining effectively, the vegetation information in mining areas was extracted comprehensive use of tasseled cap transformation, vegetation and soil correlation analysis, support vector machine classification, meanwhile the grade distribution figures of vegetation were made, and the pollution levels of different vegetation cover also were determined. The ecological environment of mine was monitored and the dynamic changes of vegetation information were determined in the period 2000-2009 using spectral change vector analysis.3. According to lack of quantitative analysis between the vegetation and ecological environment in mining area, in this paper the surface temperature of the mining area was inversed and the vegetation coverage was determined, so the relationship of vegetation coverage, surface temperature, air pollution and different types of underlying surface were quantitative described. The results would provide scientific basis for departments making relevant policy measures.4. Based the comprehensive index, the ecological environment of mining area were analyzed by calculating the biological abundance index, vegetation coverage index, water density index, land degradation index, environmental quality comprehensive index. It is not only enrich and improve the practical application cases of the ecological environment impact assessment, but also provides an important technical support energy saving for the mining.5. Through field surveys, combine the aerial data evaluate the accuracy of vegetation information. According to the results of quantitative analysis and the ecological environment index, the relevant measures were proposed.Experimental results show that the comprehensive utilization of 3S technology, using the methods of data fusion, tasseled cap transformation, support vector machine classification, spectral change vector analysis, combined with the actual acquisition of data mining, the vegetation information in the mining area could be effectively extracted, and the level of monitoring the ecological environment could be improved. It has great application value in dynamic monitoring of the ecological environment in the same type of mining.更多还原