Using Entropy and 2-D Correlation Coefficient as Measuring Indices for Impulsive Noise Reduction Techniques

摘要

Several measuring indices have been used in the literature of image restoration to examine the performance of the proposed schemes. In this paper, entropy and two-dimensional correlation coefficient have been studied and used as performance measuring indices for image quality. An adaptive two-stage algorithm for restoration of images corrupted by impulse noise proposed in one of our previous works is used in the simulation experiments. The impulse noise used is also called salt and pepper noise which is typically introduced due to sharp and sudden disturbance in the image. Several images of different characteristics that have been subject to a wide range of noise densities are tested. Experimental results show the superiority of such indices in measuring the performance of the adaptive filtering scheme to reduce the noise corrupting the images.