A Statistical Method for the Performance Evaluation of the Gray Compression of SAR Images
Xiao Wang · Radar & Ecm · 2003
Two statistical parameters are proposed in this paper, which characterize the errors of the histograms brought by different gray compression algorithms for Synthetic Aperture Radar (SAR) images. Firstly we transform the histogram dynamic range of the original images and calculate the relative mean and relative variance. Then we compare these values with the mean and variance of the histograms of the compression images and obtain the relative mean error and relative variance error. Simulation results are in accordance with manual evaluation and indicate the proposed method can be used for the evaluation of the performances of SAR images gray compression algorithms.