Abstract: Quality assessment is important for digital visual signal in the field of image and video processing. Screen content images are used in multi-device communication applications. Process evaluation, implementation, and optimization quality assessment becomes important. Screen content images (SCIs) shows different statistical properties in textual and pictorial regions, and the human visual system (HVS) also behaves differently while viewing the textual and pictorial regions in terms of the extent of visual field. Quality assessment of distorted screen content images subjectively and objectively with the help of screen image quality assessment database (SIQAD). The subjective quality scores shows, which part of the image (text or picture) have greater quality to that of overall visual quality. The single stimulus methodology is used to obtain three kinds of subjective scores i.e. the entire, textual, and pictorial regions, respectively. Analyzing the subjective data, we recommend a weighting strategy for correlation among these three subjective scores. An objective metric is used to measure the visual quality of distorted screen content images by considering visual difference of text and picture regions. The proposed quality assessment method gives better predicts to the perceptual quality of screen images, also leads to an effective way to optimization screen content coding schemes.
Keywords: Screen content image, image segmentation, quality assessment, subjective and objective quality assessment.