基于统计学理论的内陆水质遥感反演进展  

Advances in Remote Sensing of Inland Water Quality Based on Statistics

作　　者：熊育久[1] 曾爽[1] 吴秀芹[1] 邱国玉[1] 王勤学 赵少华[1] 尹婧[1] 

机构地区：[1]北京师范大学资源学院,地表过程与资源生态国家重点实验室,北京100875 [2]日本国立环境研究所,日本筑波305-8506

出　　处：《遥感信息》2008年第3期92-98,共7页Remote Sensing Information

基　　金：国家重点基础研究发展计划(2006CB400505)

摘　　要：遥感具有多尺度、多时相、覆盖面广、节省人力物力等特点,可快速反演大面积水体的整体水质状况,对其进行周期性监测与评价,成为水资源管理的热点。然而,由于光在水中传播时与各种不同物质相互作用的机理极其复杂,尤其是水中各种物质对光的吸收、后向散射大不相同,水质的遥感反演成为当前遥感应用领域的难点。本文在重点综述国外在基于统计学理论基础上的内陆水质遥感反演研究后,总结出具有代表性的反演模型,并结合水质反演的原理,指出基于统计学理论的内陆水质遥感反演存在的问题,旨在促进国内内陆水质的遥感反演研究、为相关领域的研究人员提供咨询和借鉴。With the advantages of multi-scale, multi-temporal, wide coverage, low cost and time saving, remote sensing could be used to retrieve and assess water quality periodically from a synoptic view for numerous water bodies simultaneously. It becomes a hot study domain in water resource management. The mechanism between light and water interactions, however, is very complicated because different matters and their concentration have different absorption and backscattering characters, which makes water quality retrieving from remote sensing very difficult. In this paper, we reviewed the progresses in inland water quality assessment by remote sensing based on statistics and summarized several representative algorithms for certain water parameters. Thereafter, the shortcomings of these algorithms were analyzed from the principle of remote sensing of inland water quality, aiming to give references to the related researchers and thus to promote the researches on water quality monitoring.

关 键 词：遥感 内陆水体 水质反演 回归模型 相关分析 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]