Accelerated image quality estimation based on the Monte Carlo approach for various types of image distortions

In the paper the application of the Monte Carlo method for the fast estimation of image quality is investigated. Taking into account the variety of possible distortions present in the images, probably the largest publicly available database of 1700 images with 17 types of distortions has been used for the practical verification of the proposed approach. The analysis has been performed for some of the "state-of-the-art" image quality metrics as an alternative to the classical ones together with the comparison to well known Peak Signal to Noise Ratio.