遥感影像小波特征与城市大气污染相关性对比分析  

Comparison of the wavelet characteristics of remote sensing images and urban air pollution

作　　者：薛岩松[1] 俞乐[1] 徐鹏玮 黄昭权[1] 

机构地区：[1]浙江大学空间信息技术研究所,杭州310027 [2]浙江省环保护科学设计研究院,杭州310007

出　　处：《环境科学学报》2010年第3期565-571,共7页Acta Scientiae Circumstantiae

基　　金：浙江省科技厅平台项目(No10127003)~~

摘　　要：PM10质量浓度是表征城市空气质量最为重要的污染指标之一.快速了解其浓度和大范围的分布状况,有利于控制PM10质量浓度,并最终提高空气质量.卫星遥感可在瞬间获取大区域的地表和大气信息,在污染监测上具有广泛的应用前景.本文在对覆盖杭州市2004年至2007年的MODIS数据进行了小波分析的基础上,建立了小波系数与同期PM10质量浓度实测数据的回归模型,并进一步比较了不同空气质量条件下的相应图像特征.初步实验表明,MODIS遥感影像特征与PM10浓度存在相关关系.PM10 concentration is one of the most important quantitative indicators to indicate air quality in urban areas. Efficient collection of concentration and distribution information of PM10 at a large scale benefits control of PM10 and ultimately will help improve air quality. Satellite remote sensing, which provides a means to obtain atmosphere and surface conditions in a large scale, is a good prospect for pollution monitoring. In this paper, a satellite remote sensing product, namely MODIS images captured from 2004 to 2007 over Hangzhou, were decomposed by wavelet transformation. The resultant wavelet coefficients were analyzed with the in situ PM10 concentration to establish a regression model. Further comparisons of image features in different air quality conditions were conducted. The initial experimental result shows that image features of MODIS are correlated with PM10 concentration.

关 键 词：遥感 PM10 小波分析 城市大气污染 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置] X513[自动化与计算机技术—控制科学与工程]