基于混合核函数SVM分类模型的采矿塌陷区遥感监测
Remote Sensing Monitoring of Mining Sunk Areas Change Based on Mixed Kernel Function SVM Classfication Model
近年来，由于采矿区的过度开采，造成地表塌陷积水、土壤质地破坏、水环境污染、废弃物堆积等一系列问题。矿业城市的土地利用随着资源开采而受到严重的破坏，这种伴随着剧烈地表扰动的过程，使矿区四周的土地利用格局在短期内发生变化。本文以徐州沛县东北部的大屯矿区局部区域为例，选取LandsatTM/ETM+的25年间的时序数据，利用混合核函数的支持向量机分类方法对采矿塌陷区的土地利用/覆盖的时空变化情况进行研究，特别对塌陷区的积水区域进行了耕地资源与塌陷地积水等信息的自动提取及变化检测。从1987年至2012年的塌陷地积水变化可以分析获得矿区开采引起的土地利用遭到的破坏程度，随着煤矿的不断开采，塌陷面积日益趋大，并且塌陷区积水面积扩展方向就是地下煤矿开采的方向。这为后续进一步研究采矿区的土地利用/覆盖变化、地表沉降与耕地资源变化提供前提基础，为保护矿产资源，集约利用土地资源，以及煤矿区土地资源与矿产资源的协调开采与保护提供决策。
With the high-strength exploitation of mineral resources in the mining city, the environment faces a new challenge and a serious situation recently. During the exploitation of mineral resources, there was a lot of coal-mining depression land, the buildings and other public facilities were destroyed, especially the cultivated land resources. So the analysis of mineral resources mining and cultivated land resources change has attracted more and more attention.The goal of this paper is to promote the coordinate development of land use and coal resource exploitation. The mining area of Datun was chosen as the study area in Peixian, Jiangsu. The multi-temporal remote sensing images were selected as data source since 1987. On the basis of the study of theory and method, this paper conducts the research on the change detection of cultivated land resource.
林怡、叶勤、何丽玲、王鹏育、张玉关
矿山环境保护矿山地质、矿山测量环境科学技术现状
采矿塌陷区，混合核函数的SVM分类，变化检测
mining areaSupport Vector Machinechange Detection.
林怡,叶勤,何丽玲,王鹏育,张玉关.基于混合核函数SVM分类模型的采矿塌陷区遥感监测[EB/OL].(2014-09-04)[2025-08-16].http://www.paper.edu.cn/releasepaper/content/201409-43.点此复制
评论