A Research For Extraction And Analysis Of Land Cover Information Of Yongij

The remote sensing images have been widely used in the research of usage and change of land. How to improve the efficiency of information extraction results and make the remote sens-ing technology practicality are becoming an urgent problem in the application. So far there are several extraction methods proposed, including two major parts:artificial visual interpretation and automatic extraction. The former one cost a lot of manpower and sources, so it is replacing by the later one. Surface coverage information extraction technology is getting more automated and efficient. Compare with the method based on the pixels, Object-oriented method does not have the problems of redundancy of data. In the research areas first we do some analysis accord-ing to the surface coverage data; then we make clear categories; At last, we figure out the real changes of different kinds of landscapes.This paper uses three periods of Landsat TM multi-spectral remote sensing image as the main data sources which combined with the region's DEM data and the present situation of land using data in 2008. Then extracting useful in-formation from the land coverage data based on the ENVI EX software,and using the classification precision of remote sensing images in 2008 to verify this method. After that we can use this method to extract information from 2002 and 2014 TM images. We can analyze the changes of land cover and their categories using the information extracted from the data we have. Through obtaining the specific flews of different categories we can predict the influence factors of the changes and the future trends.This paper proposed a useful method to obtain the land cover information in the study of usage and change of landscape. It has good reference value for classification of remote sensing data.