Objective color image quality assessment based on Sobel magnitude

Abstract

Image quality assessment algorithms aim to evaluate the perceptual quality of an image by assigning an evaluation score. By comparing the scores, the perceptual similarity or difference between two images can be assessed. In this paper, we present a new full-reference image quality metric method which was developed by combining Sobel magnitude and chrominance information in the YIQ color space. But differing from existing methods, our model incorporates color intensity adaptation to extract and enhance perceptually significant image features. The proposed metrics are tested on three well-known databases available in the literature (TID2013, TID2008, and CSIQ). Experimental results presented to confirm that the proposed metric is an effective and have low computational complexity.

Quality index evaluation of videos based on fuzzy interface system

“…Subjective IQA is measured by asking a number of expert observers to look at each distorted image and rate its quality to obtain the mean opinion score (MOS) or differential mean opinion score (DMOS). Although subjective IQA methods are the most natural, accurate and reliable way of measuring the image quality, they have several drawbacks, including inconvenience and computation time [6, 7], which make them impractical for real‐time use or for control of image quality in automated systems. However, subjective IQA methods are usually correlated and benchmarked against any new objective IQA method to reference its performance against human visual perception.…”