Identification of human actor in various scenarios by applying background modeling

摘要

The identification of human actors in various scenarios with dynamic backgrounds has increased significance in numerous video-based applications. Lot of problems relevant to the detection of actors in scenarios with a dynamic background has to be addressed. This work proposes an approach to identify human actions with dynamic backgrounds and illumination variations. To accurately detect the foreground, the proposed method applies background modeling. The proposed method was assessed with approaches such as FrmDiff, MoG and ACMMM03. The assessment was performed on renowned datasets such as CDnet,WALLFLOWER and I2R. The results of assessment demonstrate the accuracy and the enhancement of the proposed method over the other methods taken into account.