A Background Modeling and Foreground Detection System using Scaling Coefficients for Video-based Surveillance Model

Abstract

In image processing, all the hassle of videotape is active research in the current era when background modeling and foregrounding are the phases of video processing used to identify objects that move under difficult conditions. It requires efficient methods to manage dynamic background and illumination changes, as well as algorithms that must meet real-time and low-memory needs. This article describes a foreground detection method using background modeling and scale factors determined by a new color model called red-light blue. They are used to compare two-key frames to find pixels with reduced brightness. In this regard, our system offers three consecutive operations, that is, in step 1: it overlaps the background with pixels. Step 2: Check the background pixels. Step 3: After validating Phases 1 and 2, it validates the background. After background calculation, foreground objects can be found using scaling factors and additional criteria.