湖泊水环境高空间遥感影像污染点源信息提取研究

传统的污染点源信息提取技术存在权重因子各指标取值范围广、提取精准度不准确、植被覆盖率检测效果差等问题,为此,采用高空间遥感影像对湖泊水环境污染点源信息提取进行研究.针对信息的选择和预处理,需要先获取污染点源信息,并利用多项式对遥感影像图中的畸形几何进行修正;通过利用TM影像技术分析波段组合和单独使用波段的分类效果;根据影像中不同的隶属度,构建水体规则,分析水体差异指数,由此提高信息提取的精度,进而完成污染点源信息的提取.通过选取验证区域,进行实验,实验结果表明,使用该技术能够缩小权重因子各指标取值范围,并提高污染点源信息提取的精准度与植被覆盖率检测的效率,实用性较强.

Information Extraction of Pollution Point Source of High Spatial Remote Sensing Image in Lake Water Environment

Traditional point source pollution information extraction technology has insufficiencies wide weighting factor range, inaccurate extraction, poor vegetation coverage testing results, therefore, high space remote sensing image of lake water environ-ment point source pollution information extraction were studied. For the selection and pretreatment of information, it is necessary to obtain the pollution point source information, and use the polynomial to modify the abnormal geometry in the remote sensing im-age. This study uses TM image technology to analyze band combination and single use band classification effect. According to the different membership degree of the image, the water body rules are constructed to analyze the water difference index, thus impro-ving the accuracy of information extraction, and then the extraction of pollution point source information. Area, by choosing vali-dation experiments, the experimental results show that using this technology to reduce the weighting factor of each index value range, and improve the precision of point source pollution information extraction and vegetation coverage testing efficiency, strong practicability.