• M. Kamarasan

Over the past decade, the video content analysis is attracted by many scholars. In this paper, we present Multiresolution based Full Range Auto-Regressive model (FRAR) for video content analysis. The video signal is assumed to be a linear combination of a spatial-temporal independent component which is temporally approximated by FRAR parameters. The FRAR model yields a linear approximation to the temporal evolution of a stationary stochastic process. The 2D wavelet transform is incorporated to decompose the video frame into various subbands which represent the spatial-temporal multiresolution features. The spatial-temporal features are extracted from each subbands which is efficiently resemble the Human Visual System (HVS) characteristics. The FRAR model describes each spatial-temporal feature vector as a linear combination of the previous vectors within a reasonable time interval. Shot boundaries are well detected based on the FRAR prediction errors, and then at least one keyframe is extracted from each shot for further analysis. The experimental result shows that the proposed method obtains higher precision when compared to that of existing methods.

FRAR, wavelet transform, multiresolution, spatial-temporal, keyframe, subbands.

"Video content Analysis based on Full-Range Autoregressive Model using Multi-resolution Features", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 6, page no.134-143, June 2018, Available :http://www.jetir.org/papers/JETIR1806817.pdf

"Video content Analysis based on Full-Range Autoregressive Model using Multi-resolution Features", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 6, page no. pp134-143, June 2018, Available at : http://www.jetir.org/papers/JETIR1806817.pdf