A New Quality Assessment for Geometrically Distorted Images Using Importance Map

Abstract

The use of quality measures based on human perception is a matter of increasing interest in multimedia applications. Most of the previous methods dealt with displacement field causing distortion and have not given much consideration to HVS based quality assessment. Gabor filter based quality metric, though it used the HVS concept it dealt only with the low level processing of edges and bars. To improve the performance of the Gabor based metric it is necessary to apply higher level perceptual factors. In fact, HVS consists of two major components: an early vision model and a visual attention model which indicates regions of interest of a scene through the use of Importance Maps. The attention model takes into account several factors which are known to influence visual attention and eye movements, and these factors must be considered to include higher level visual processing in human vision, to obtain a more accurate indication of picture quality. In this paper, using the importance map the main features in the image are noted and the quality is assessed.