Results and Disscussions
1.3. Digital Image Analysis
Images of LANDSAT ETM+ (December, 2001), LISS III (February, 2004) and ASTER (June, 2004) were classified individually with separate training set. Qualitative and quantitative evaluation of the classification results of three different imageries will be presented and discussed below.
Supervised classification was performed using Maximum Likelihood Classifier, Minimum Distance to Mean and Parallelepiped classifier for LANDSAT (of spatial resolution of 30m), LISS III (23.5m) and ASTER images (15m) to classify the image into eight landuse classes namely healthy tea, moderately affected tea, affected tea, river, riverbed, settlements, scrubs and barren land. Vegetation classes like scrubs, settlements, etc. were found to be mixing with tea class. The classes were then validated from the field visits. Maximum Likelihood Classification (MLC) gave the best results as compared to Minimum Distance to Mean (MDM) and Parallelepiped Classification.