Author : Pallavi M and Shivakumar B R

Classification of Low-Resolution Remotely Sensed data using suitable classification methods depends on the quality of the data. For low resolution images, it is always difficult if not impossible to differentiate classes as the number of classes is increased above 8. Hence, as a prerequisite to image classification, Pan Sharpening or merging the low resolution image with a high resolution image of the same area considered, can improve the quality of the data. The method used for Pansharpening in this paper is Wavelet Resolution Merge. Further, image enhancement techniques can be applied to data for making the considered dataset more interpretable. In this paper, Histogram equalization is employed on the data after Pansharpening process to study the improvements. Image parameters such as Standard Deviation and Mean are considered for decision making. It has been found that, by Pansharpening and Histogram Equalization, the quality of the input data is improved, which can further yield better classification results.

Keywords: Pansharpening, Resolution Merge, Wavelet Transform, Histogram Equalization, Image Fusion, Remote Sensing

Article published in International Journal of Current Engineering and Technology, Vol.5, No.2 (April-2015)