基于TM影像的人工开发地类遥感信息提取

Artificial Exploitation of Remote Sensing Information Extraction Based on TM Image

人工开发与人类活动密集相关,人工开发用地的研究对于未来研究城镇及耕地扩张具有重要意义。“同谱异物”这一现象给遥感解译等相关研究工作带来许多困难。本文基于中分辨率遥感影像Landsat5 TM数据,综合地物时相特征及其光谱特征,构造比值裸露指数与相关指数结合,利用最大似然监督分类法对菲律宾地区的人工开发用地进行提取。结果表明,采用比值裸露指数分类的总体精度达87.3%,相较于NDBI指数分类的精度有所提高。

Artificial development is closely related to intensive human activities,and the study of artificial development land is of great significance for the future study of urban and cultivated land expansion.The phenomenon of"same spectrum with different objects"brings many difficulties to remote sensing interpretation and other related research work.Based on the Landsat5 TM data of the medium-resolution remote sensing image,this paper integrated the time-phase characteristics and spectral characteristics of ground features,combined the Ratio Barren Index and relevant index,and extracted the artificial development land in the Philippines by the maximum likelihood supervised classification method.The results show that the overall accuracy of the ratio building index classification is 87.3%,which is higher than the accuracy of NDBI index classification.