A novel framework for background subtraction and foreground detection

作者：

Highlights：

• A new data structure, Mino Vector (MV), is designed to serve the whole framework.

• Concepts of dynamic nature (DN) and dynamic rank (DR) are defined for the first time.

• A normal distribution is chosen as the pilot estimate for the Varying KDE.

• The threshold of each pixel is adaptively controlled by each pixel's own DR.

• A novel update mechanism called Tetris update scheme (TUS) is proposed.

摘要

•A new data structure, Mino Vector (MV), is designed to serve the whole framework.•Concepts of dynamic nature (DN) and dynamic rank (DR) are defined for the first time.•A normal distribution is chosen as the pilot estimate for the Varying KDE.•The threshold of each pixel is adaptively controlled by each pixel's own DR.•A novel update mechanism called Tetris update scheme (TUS) is proposed.

论文关键词：Background modeling,Mino vector,Dynamic nature,KDE,Denoising,Tetris update scheme

论文评审过程：Received 21 November 2017, Revised 19 June 2018, Accepted 1 July 2018, Available online 3 July 2018, Version of Record 7 July 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.07.006