In this study, we evaluate various techniques used for land cover mapping. All these techniques perform only if all the radiometric problems (drift of the sensor, atmospheric effects, etc.)  are corrected for remote sensing imagery. NDVI shows an excellent linearity as a function of the rate of vegetation cover.  NDVI performs better than all other indices in regions with relatively good vegetation cover, while TSAVI1 shows the sensitivity to the optical proprieties of bare soil subjacent to vegetation cover and is  suitable for dry (semi arid) regions. This analysis shows that 45.93% of area is under vegetation and 54.07% under non-vegetation. Both supervised and unsupervised classification approaches were tried to identify landuse categories in the district. The level of accuracy in supervised classifier was 94.67% compared to unsupervised classifier (78.07%).