Blind quality assessment for screen content images by orientation selectivity mechanism

Currently, most objective image quality assessment (IQA) methods designed for screen content images (SCIs) require reference information, and existing blind IQA metrics cannot obtain consistent results with subjective scores. In this study, we propose a novel blind IQA method for SCIs based on orientation selectivity mechanism. First, we extract orientation features to compute the visual distortion of degraded SCIs by orientation selectivity mechanism. The statistical orientation features are further obtained by the histogram of orientation features. Second, the statistical structure features are calculated as the complementary information of orientation features for quality prediction of degraded SCIs. Finally, we employ support vector regression (SVR) as the mapping function from these extracted statistical features to quality scores. Experimental results show that the proposed method can obtain better performance of visual quality prediction for SCIs than other existing related methods.