Processing of Hyperspectral Data using Wavelet Transform

Abstract

Remote sensor technology has encouraged series of research work in the area of signal and image processing. This is because the application of remote sensor has made it possible to obtain different types of signals and images from different places all over the world. In most cases, data obtained from hyperspectral images are found to be too voluminous and noisy. This, to a certain extent affects the accuracy of the result obtained when such signals or images are further processed for some applications. Previous research works have not sufficiently addressed this fundamental problem. Therefore, this research work is out to make use of Wavelet Transform for processing signals obtained from hyperspectral images with a view to denoise and reduce the data dimensionality without losing part of its content. Having undergone the process of denoising, the quality of the image or signal is drastically improved in terms of its clarity and size. This produces a better result when such signal is used for some applications. The system was implemented using MatLab wavelet tool. Hence, the result obtained is found to be better than the previous ones. The result also produced an hyperspectral spectrum/signal that has been thoroughly denoised and dimensionally reduced to an acceptable size within a very short computational time.