Fuzzy logic and temporal information applied to video quality assessment

  • Authors:

  • Carlos Danilo Miranda Regis;José Vinícius De Miranda Cardoso;Ítalo De Pontes Oliveira;Marcelo Sampaio De Alencar

  • Affiliations:

  • Institute for Advanced Studies in Communications, Federal University of Campina Grande and Federal Insitute of Education, Science and Technology of Paraíba, Campina Grande, Paraíba, Braz ...;Institute for Advanced Studies in Communications, Federal University of Campina Grande, Campina Grande, Paraíba, Brazil;Federal Insitute of Education, Science and Technology of Paraíba, Campina Grande, Paraíba, Brazil;Institute for Advanced Studies in Communications, Federal University of Campina Grande, Campina Grande, Paraíba, Brazil

  • Venue:
  • Journal of Mobile Multimedia
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Video Quality Assessment (VQA) plays an important role for video communications systems and services, mainly to determine, accurately, the ratio between the provided quality and the resource demand. The objective VQA is a fast and viable methodology to determine the video quality for video service providers, although it presents an unsatisfactory correlation with the scores of quality given by the Human Visual System (HVS). The authors propose a novel full reference objective video quality metric considering spatial and temporal analysis. The spatial analysis used an algorithm, based on fuzzy logic, to classify the regions in three components. Temporal analysis was performed by means of the perceptual weighted structural similarity index (PW-SSIM) between the frames that contained the differences of pixels in the same spatial position and in subsequent frames. To validate the proposed VQA algorithm, the correlation coeffcients between the objective measures and the subjective scores provided by the LIVE Video Quality Database were computed, considering the following distortions: H.264 and MPEG-2 encoding and transmission of H.264 bit-streams over IP and wireless networks. The results demonstrate that the proposed algorithm is a competitive alternative when compared with the classical objective algorithms such as MOVIE.