Research On No-reference Image Quality Assessment Based On HVS

The main evaluation methods in image quality evaluation system will be divided into two categories: subjective evaluation and objective evaluation. Subjective evaluation is most accurate but also time wasting and effort consuming, limited to abundant measurements and real time applications; objective evaluation can be classified into full-reference,reduced-reference and no-reference according to the availability of the original image. In many cases the original image can not be acquired which limits the full-reference and reduced-reference methods. Images are supplied to the humans, therefore studying no reference image quality evaluation methods based on HVS has important practical significance. This paper summarizes the image quality assessment methods,the no-reference image quality measurement is stressed and then two important no-reference methods based on HVS are introduced.Firstly, analyzes the current research status and the future trend of image quality evaluation methods. Then describes two evaluation methods and three designing principles.Finally, introduces the characteristics of human vision system, puts forward two kinds of no Two algorithms based on HVS is proposed in this paper to evaluate no-reference image quality assessment:1.No-reference quality assessment based on HVS for blurred image. Firstly calculates the local variance to extract notable regions of the image, then use Reblur algorithm to obtain the evaluation result. Experimental results show that the algorithm is superior to many existing classical algorithms on algorithm performance.2.No-reference quality assessment based on HVS for JPEG compressed image. This paper calculates three eigenvalues of the parts extracted from the image for blockness and blur effect. Experimental results show that the algorithm is high computational efficiency and meets the three design principles.The algorithms proposed in this paper can be applied to image processing system and guide the hardware aspects, which has great theoretical importance and practical value.