A New No-Reference Assessment Metric of Blocking Artifacts Based on HVS Masking Effect

In this article, a new algorithm-NABBHM based on HVS masking effect to assess blocking artifacts without reference is presented. We consider the brightness and complexity masking effect through defining two factors of masking to revise the original blocking artifacts value to make it more consistent with the human visual. Experiments show that NABBHM has strong feasibility and is highly fitting the subjective assessment value. Blocking artifacts is the main types of distortion of B-DCT, therefore whether accurately assess it or not is precondition of accurately assessing the quality of entire frame of video, in particular, for the restoration of the video using the same compression algorithm, the severity of the blocking artifacts can directly and greatly influence the quality of video. At present, variety of measurement methods for Blocking artifacts has been presented in Ref.(3-7).Ref.(6) gives a non reference measurement method, however the use of the visual model in which is too simple, not well reflect the human visual characteristics. The algorithm models given by Ref. (3), (7) need reference images. The visual model of Ref. (7) is too