Block-based codebook model with oriented-gradient feature for real-time foreground detection

Jiu Xu, Ning Jiang, Satoshi Goto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

In this paper, a novel approach is proposed to achieve the foreground objects detection in video surveillance system using codebook method. The block-based background model upgrades the pixel-based codebook model to block level which can utilize the dependency and find relationships between neighbouring pixels, thus improving the processing speed and reducing memory during model construction and foreground detection. Moreover, by adding the orientation and magnitude of the block gradient, the codebook model contains not only information of color, but also the texture feature. The texture information can further reduce noises and refine more entire foreground regions. Experimental results prove that our method has better performance compared with the standard codebook and some other former algorithms.

Original languageEnglish
Title of host publicationMMSP 2011 - IEEE International Workshop on Multimedia Signal Processing
DOIs
Publication statusPublished - 2011
Event3rd IEEE International Workshop on Multimedia Signal Processing, MMSP 2011 - Hangzhou
Duration: 2011 Nov 172011 Nov 19

Other

Other3rd IEEE International Workshop on Multimedia Signal Processing, MMSP 2011
CityHangzhou
Period11/11/1711/11/19

ASJC Scopus subject areas

  • Signal Processing

Fingerprint Dive into the research topics of 'Block-based codebook model with oriented-gradient feature for real-time foreground detection'. Together they form a unique fingerprint.

Cite this