Moving Object Detection and Segmentation using Background Subtraction by Kalman Filter

Objectives: Object tracking and detection are significant and demanding tasks in the area of computer vision such as video surveillance, vehicle navigation, and autonomous robot navigation. Methods/Statistical Analysis: This paper presents the moving object tracking using Kalman filter and reference of background generation. Kalman filter is based on two types of filters: cell Kalman filter and relation Kalman filters. The process entails separating an object into different sub-regions and discovering the relational information between sub-regions of the moving objects. Findings: In this paper, the precise and real-time method for moving object detection and tracking is based on reference background subtraction and use threshold value dynamically to achieve a more inclusive moving target. This method can effectively eliminate the impact of luminescence changes. Due to deployment of Kalman filter this fast algorithm is very straightforward to use to detect moving object in improved way and it has also a broad applicability. This technique is very authentic and typically used in video surveillance applications. Application/Improvements: This technique is very legitimate and typically used in video surveillance applications. The Kalman filtering algorithm upgrades the model and enlarges the dimensionality of the moving system state.