An algorithm for image quality assessment

The paper presents a novel algorithm for image quality assessment. First, a simple model of the human visual system, consisting of a nonlinear function and a 2D filter, processes the input images. This filter has one user-defined parameter, whose value depends on the reference image. In the next step, the average value of locally computed correlation coefficients between the two processed images is found. This criterion is closely related to the way in which a human observer assesses image quality. In the last step, the image quality measure is computed as the average value of locally computed correlation coefficients, adjusted by the average correlation coefficient between the reference image and the error image. In this way, the proposed measure differentiates between the random and signal-dependant distortion, which have different effects on a human observer. The performance of the proposed quality measure is illustrated by examples involving images with different types of degradation.