An Improved Motion Detection Algorithm Based on ViBe

Yun-xiao MA, Zhi WENG, Yu-fan JIANG, Jia-wei FENG, Su ZHANG, Shu-sheng QIU

Abstract


ViBe (Visual Background Extractor) algorithm is a kind of fast and accurate background subtraction algorithm, which is universally applied in video monitoring. The only drawback is the moving target in the first frame would introduce an artifact called ghost. In order to solve this problem, an eliminating ghost algorithm was designed in this paper. The algorithm depended on the foreground and the adjacent background area histogram and similarity measure. Because dynamic background video was easily influenced by background noise and interference, morphological image processing method was used in this paper. Experiments results show that the improved algorithm can quickly solve the ghost, foreground noise and holes problem.

Keywords


Motion detection, Vibe algorithm, Ghost suppression, Morphological image processing


DOI
10.12783/dtetr/ameme2017/16244

Refbacks

  • There are currently no refbacks.