BACKGROUND SUBTRACTION ALGORITHM BASED ABANDONED OBJECT DETECTION

Abstract

Abandoned object detection is a basic necessity in numerous video observation settings. Two background short term and long term are built and refreshed persistently continuously. Change detection Mean-ratio and Log Ratio operators can be utilized. To accomplish higher exactness combination. Movement based criteria should be connected for static and moving object detection. Proposed strategy can distinguish little deserted objects inside low quality observation recordings, and it is additionally strong to the fluctuating illuminations and dynamic background.