Morphological De-noising Method in Differential Image for Change Detection

Guoqing Li, Bo Liu, Xiujie JiangJournal of Convergence Information Technology 2012 · DOI

Abstract

The differential image plays an important role in applications of video processing. In this article, the features of noise distribution in the differential image are discussed. Different with previous work which focused on the intensity distribution of differential image with pixels absolute value, the isolation of the noises in spatial distribution and the fluctuation of the value signs are considered to be important features to distinguish the noises and useful signals for change detection in differential image. With objects moving in the continuous frames, the significant regions whose pixels value are non-zero show different features from noise whose pixels value are non-zero too. Based on these different features, a simple method composed by basic morphology operations, closing and opening operations, is proposed to de-noise in differential image for change detection. Results of experiments show that the method is satisfactory in de-noising and low in computational complexity.
BibTeX
@article{li2012morphological,
  author = {Guoqing Li and Bo Liu and Xiujie Jiang},
  title = {{Morphological De-noising Method in Differential Image for Change Detection}},
  journal = {Journal of Convergence Information Technology},
  year = {2012},
  doi = {10.4156/jcit.vol7.issue8.11},
}

Loading related content...

Figure