A Spatial Domain Feature Based Approach for No Reference Image Quality Assessment of JPEG Compressed Images

This is the world of technological advancement; where emphasis is more on data transfer and that too with higher speeds. This leads to higher bandwidth requirement. To resolve the bandwidth requirement problem various compression techniques are employed. But with this compression and data transfer new problem arises that is nothing but distortions. And at the same time it is also important to assess the quality of these images being processed and transferred. This paper presents a novel approach for image quality assessment without any reference. This paper focuses on JPEG compressed images which mainly suffers blocking and blurring artifacts. A spatial domain approach is employed along with histogram to quantify the distortion. Results are found to be correlating well with the subjective image quality assessment and various standard quality assessment parameters. Results are also found to be comparable to that of full reference quality assessment technique.