基于贝叶斯定理的黑臭水体遥感监测方法研究

【摘要】 黑臭水体是水环境受到严重有机污染产生的极端现象,近年来成为环境治理中的热点问题。以河北廊坊市为研究区,系统采样获取水体各项理化指标,结合同期过境Sentinel-2影像进行黑臭水体识别工作。选取黑臭水体光谱指数(BOI)为提取特征,基于贝叶斯定理计算理论阈值,在过程中修正了实验样本与研究区客观情况之间的差异,通过K-S检验估计黑臭水体和清洁水体的BOI分布函数,计算理论阈值并进行黑臭水体提取。通过加入2类水体的基础概率,降低了因样本差异、主观经验而带来的影响,使结果更加符合研究区的客观实际,有助于开展区域工程化应用。更多还原

【Abstract】 Black and odorous water is an extreme phenomenon of serious organic pollution in the water environment. In recent years,it has become a hot issue in environmental monitoring and governance. This paper takes Langfang,Hebei Province as the research area,and the system obtains various physical and chemical indicators of the water body,and then combines the Sentinel-2 images of the transit during the same period to perform remote sensing identification of black and odorous water bodies. The black and odorous water spectral index(BOI)is selected as the extraction feature,and the theoretical threshold is calculated based on Bayes’ theorem. In the process,the difference between the experimental sample and the objective situation in the study area is corrected,and the BOI distribution function of black and odorous water bodies and clean water bodies is estimated by K-S test,the theoretical threshold is calculated and black and odorous water bodies are extracted. It reduces the impacts of sample differences and subjective experience by adding the basic probability of the two types of water bodies,making the results more in line with the objective reality of the study area,and helping to develop regional engineering applications.更多还原