Land cover land use mapping using optimized wavelet packet transform features

Abstract

Today land use and land cover mapping is of great significance in scientific scholarly research planning and management Regional land use pattern reflects the character of interaction between man and environment and the influence of distance and resources based on mankind s basic economic activities Remotely sensed satellite images provide a synoptic overview of the whole area in a very short time span This leads to quick and truthful representation of the real world in the best possible manner It provides an insight to coordinate relationship among transportation residential industrial and recreational land uses besides providing broad scale inventories of natural resources and monitoring environmental issues including land reclamation mangrove restoration disaster relief water quality and planning economic development The accuracy of land use and land cover mapping largely depends on the resolution of the remotely sensed images However the resolution of remote sensing images increases every day raising the level of detail and the heterogeneity of the scenes Higher spatial resolution of satellite images adds a complexity for automated extraction of image semantics Increases in spatial resolution extend variation of objects in its spectral structural and textural characteristics Existing techniques used to process lower resolution satellite images do not generalize well to higher resolutions due to the changed appearance of features of interest In this work land use and land cover mapping is done for benchmark Ahmedabad and Madurai newline newline newline