Land cover/land use mapping using different wavelet packet transforms for LISS IV imagery

Texture in high resolution satellite images requires substantial amendment in the conventional segmentation algorithms. This study examined and evaluated the use of wavelet packet transforms for urban texture analysis and image classification in high spatial resolution LISS IV imagery, which provide more details of the urban areas. This paper analyzes the performance of the combination of Wavelet Statistical Features (WSFs) and Wavelet Co-occurrence Features for the classification of LISS IV images. Per pixel classification accuracy is improved in this work by varying the window size. Four indices (user's accuracy, producer's accuracy, overall accuracy and kappa coefficient) are used to assess the accuracy of the classified data. Experimental results show that a multi-band and multi-level wavelet packet approach can be used to drastically increase the classification accuracy.