Abstract: The present condition and its changing trend of wetland are very important to the making of policy upon the reclamation,exploitation,management and protection of wetland.In this paper,the coastal wetlands in the northern part of Jiangsu province were taken as study object and the technology about the extraction of wetland was explored by using multi-features and knowledge rules and multi-spectral Landsat 7 ETM+ images acquired on May 26,2002 and June 17,2007,in combination with the analysis upon the characteristics of wetlands and its presentation in remotely sensed imagery and the data of field investigation of the same period.At first,the extraction of wetland on the image acquired in 2002 was carried out based on knowledge rules,and then,the fusion of supervised and unsupervised classification was used to classify the image acquired in 2007.Both of the two methods obtained good results.At last,a change detection was performed on the two images using both comparison after classification and RB-NDVI-NDMI methods.The two results are in substantial agreement,but they have pros and cons.

Key words: coastal wetland of Jiangsu Province, extracting information, change detection, knowledge rules, multi-features