Landuse and landcover change detection is one of the most important applications of remote sensing study. In a mountainous region like Himalayas, landcover and landuse change detection study is always problematic and it is always important to select the best classification system to perform the change detection study. In the present study pixel and segment based approaches of classification have been used to compare the result. Due to hilly nature of the area, lots of shadow present in the image, which hinder the proper classification. For that one topographic normalization model has been developed using DEM to extract the information from shadowed area, and then the classifications are performed and post classification change detection study has been carried out.  In the present study the landcover and landuse classification is done with IRS 1C LISS-III images of pre and post Chamoli Earthquake. Some of the changes in landuse classes are correlated with the after effect of Chamoli Earthquake and one landslide map has been prepared for the area showing the locations of landslides occurred after Chamoli Earthquake.