作者: G. Saravana , Ankur Pandit , R Rout , Satish Kumar
DOI:
关键词:
摘要: This paper describes the object/region based approach for satellite image classification to extract snow cover area in high mountainous region. The objective is develop an application classify imagery using a Fuzzy-based supervised method spectral information of input image. procedure first divided into several groups (in space) on selected training datasets Gaussian function as measure similarity. result was then segmented regions fuzzyfication process. process also incorporates edge avoid intermixing sub-pixels remotely sensed images. After that extracted has been classified rule. Several were samples region classification. Each compared and assigned its closest class. implemented MATLAB software tested USA’s MODIS TERRA imagery. significantly reduces mixed pixel problem which suffered by most methods.