No Reference Color Image Contrast and Quality Measures

No-reference (NR) image quality assessment is essential in evaluating the performance of image enhancement and retrieval algorithms. Much effort has been made in recent years to develop objective NR grayscale and color image quality metrics that correlate with perceived quality evaluations. Unfortunately, only limited success has been achieved and most existing NR quality assessment is feasible only when prior knowledge about the types of image distortion is available. This paper present: a) a new NR contrast based grayscale image contrast measure: Root Mean Enhancement (RME); b) a NR color RME contrast measure CRME which explores the three dimensional contrast relationships of the RGB color channels; c) a NR color quality measure Color Quality Enhancement (CQE) which is based on the linear combination of colorfulness, sharpness and contrast. Computer simulations demonstrate that each measure has its own advantages: the CRME measure is fast and suitable for real time processing of low contrast images; the CQE measure can be used for a wider variety of distorted images. The effectiveness of the presented measures is demonstrated by using the TID2008 database. Experimental results also show strong correlations between the presented measures and Mean Opinion Score (MOS)(1).