Research on water body information extraction and monitoring in high water table mining areas based on Google Earth Engine
Anya Zhong, Zhen Wang, Yulong Gen
China University of Mining and Technology - Beijing
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
The extensive and intensive exploitation of coal resources has led to a particularly prominent issue of water accumulation in high groundwater table mining areas, significantly impacting the surrounding ecological environment and directly threatening the red line of cultivated land and regional food security. To provide a scientific basis for the ecological restoration of water accumulation areas in coal mining subsidence, a study on the extraction of water body information in high groundwater level subsidence areas is conducted. The spectral characteristics of land types within mining subsidence areas were analyzed through the application of the Google Earth Engine (GEE) big data cloud platform and Landsat series imagery. This study addressed technical bottlenecks in applying traditional water indices in mining areas, such as spectral interference from coal slag, under-detection of small water bodies, and misclassification of agricultural fields. An Improved Normalized Difference Water Index (INDWI) was proposed based on the analysis of spectral characteristics of surface objects, in conjunction with the OTSU algorithm. The effectiveness of water body extraction using INDWI was compared with that of Normalized Difference Water Index (NDWI), Enhanced Water Index (EWI), and Modified Normalized Difference Water Index (MNDWI). The results indicated that: (1) The INDWI demonstrated the highest overall accuracy, surpassing 89%, and a Kappa coefficient exceeding 80%. The extraction of water body information in mining areas was significantly superior to that achieved by the other three prevalent water indices. (2) The extraction results of the MNDWI and INDWI water Index generally aligned with the actual conditions. The boundaries of water bodies extracted using MNDWI in mining subsidence areas were somewhat ambiguous, leading to the misidentification of small water accumulation pits and misclassification of certain agricultural fields. In contrast, the extraction results of INDWI exhibited better alignment with the imagery, with no significant identification errors observed. (3) Through the comparison of three typical areas, it was concluded that the clarity of the water body boundary lines extracted by INDWI was higher, with relatively fewer internal noise points, and the soil ridges and bridges within the water bodies were distinctly visible, aligning with the actual situation. The research findings offer a foundation for the formulation of land reclamation and ecological restoration plans in coal mining subsidence areas.
逐年被引趋势
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
计算机 / AIRemote-Sensing Image Classification
Wetland Management and Conservation · Remote Sensing and Land Use
参考文献 44
此处列出前 3 条
引用本文 2
按被引量排序,此处列出前 3 条