VIDEO SEGMENTATION USING A NOVEL LBP DESCRIPTOR

Abstract
Video segmentation is the basis for content-based video retrieval, object recognition, object tracking, and video compression. This paper proposes a kind of novel and easy spatial-temporal LBP coding method, using the spatial-temporal 2 × 2 × 2 neighborhood clique to encode the changes in a video. Based on the coding method, a scheme of video segmentation is developed. Compared to the traditional segmentation method, its distinguished advantage is that it does not need to construct the background model and is simple in computation. Experimental results indicate that this new algorithm can give satisfying segmentation results.

Authors
Zhongkun He, Yongping Zhang, Dechun Zheng
Ningbo University of Technology, China

Keywords
Video Segmentation, Local Binary Patterns, Change Detections, Morphological Filters
Published By :
ICTACT
Published In :
ICTACT Journal on Image and Video Processing
( Volume: 5 , Issue: 1 )
Date of Publication :
August 2014

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.