Retina inspired no-reference image quality assessment for blur and noise

A blind image quality assessment technique with no-training is proposed in this paper. The proposed technique considers two important types of distortions viz. noise and blur for quality estimation of an image. The technique is motivated by two significant phenomena of perception in the retina of an eye. First being the center-surround retinal receptive field and second, existence of multiple spatial frequency channels. Center-surround retinal receptive field in the proposed technique is modeled with the help of Difference of Gaussians (DoG). In retina, multiple spatial frequencies have been found and due to this, signals generated from center and surround fields exist at different frequencies. In order to mimic center-surround receptive field at multiple frequencies, we compute multiple DoG images at multiple standard deviation values generated for different frequencies. Further, two significant features based on entropy and edges are extracted from the obtained DoG images and are subsequently used to compute the quality of the image. The proposed technique does not require any training with distorted or pristine images; or subjective human score to predict quality of the image. We evaluate the proposed technique on LIVE, CSIQ and TID08 databases and observe that the obtained results match very well with human subjective opinions. The proposed technique outperforms the latest no-training, no-reference (NR) based image quality assessment techniques.