No Reference Image Quality Assessment: Feature Extraction Approach

In our day to day life we are accompanied with digital visual information. Various distortions are introduced during the sharing, transmission or storing of this digital information. Image quality assessment refers to the judgment of the image quality as various image processing applications depends on this information. Image quality can be measured by two ways, subjective and objective method. Subjective method where human judges the quality of image using mean opinion score (MOS) method. This method is time consuming, expensive and cannot be implemented in system where real time valuation of image quality is needed while Objective method can be implemented without human involvement and predict image quality that is reliable with human subjective observation. When quality of test images are done without the information of distortion present in that image then it is called no reference image quality assessment or blind image quality assessment. The objective of this paper is to overview the different methodologies employed for quality assessment of an image and briefly explains the no reference image quality assessment method using a feature extraction approach and analyses its performance.