Video quality assessment using objective parameters based on image segmentation

This work presents a methodology for video quality assessment using objective parameters based on image segmentation. Natural scenes are segmented into plane, edge and texture regions, and a set of objective parameters are assigned to each of these contexts. A perception-based model that predicts subjective ratings is defined by computing the relationship between objective measures and results of subjective assessment tests, applied to a set of natural scenes and MPEG-2 video codecs. In this model, the relationship between each objective parameter and the subjective impairment level is approximated by a logistic curve, resulting in an estimated impairment level for each parameter. The final result is achieved through a linear combination of estimated impairment levels, where the weight of each impairment level is proportional to its statistical reliability. The results presented show that the use of region-based objective measurements provides more accurate predictions compared to predictions based on global parameters.