IMPROVED ESTIMATION OF FRACTION IMAGES USING PARTIAL IMAGE-RESTORATION

Mixed pixels are unavoidable in remote sensing imagery, but signature mixing models can be effective in improving the usefulness of the imag e data. Several spectral unmixing algorithms have been developed for m ultispectral and hyperspectral imagery; however, the issue of unmixing error caused by sensor system point spread function (PSF) degradation has never been addressed. We have studied this source of unmixing err or and investigated the application of image restoration before spectr al unmixing to improve the unmixing results. In this paper we show tha t the characteristics of our problem makes the design of a restoration filter differ from that for visual enhancement. Simulation results sh ow that, depending on the scene complexity, unmixing error can be redu ced by up to 72% using the proposed restoration scheme. We conclude th at image restoration is useful in increasing the accuracy of spectral unmixing.