No-reference image quality assessment for images degraded by color quantization in HSV space

Abstract

Image quality assessment is a very important and challenging task for many image processing applications. The task of quality assessment of an image can be done either with the help of a reference image of the same scene or blindly without any reference image. No-reference image quality assessment algorithms specific to a particular type of distortion are very popular for different image processing applications. Color quantization is a technique to reduce the number of unique colors in the image, but excessive color quantization can reduce the visual quality of images. In this paper, we propose a no-reference image quality measure specific to quality assessment color quantized images and color quantized images with dither. The results are validated using a subset of the standard TID2013 image quality dataset for validating it in accordance with the human visual system.