Real-Time Recognition and Tracing of Moving Objects in Video Images using Background Subtraction, Kalman Filter and Particle Filter

Improving the reliability and performance of target detection and tracking in multi-field environments is one of the key areas for moving target detection and tracking. At present, the detection and tracking of moving targets are one of the difficult research hotspots in the field of image processing. Due to the limited detection capabilities of object detectors such as vehicles, they often fail to alert in the event of an accident, especially in open areas. Aiming at the problem of target detection and tracking in a large space environment, several target detections and tracking algorithms combined with multiple image features are proposed. This paper discusses several methods for detecting moving targets. The algorithm is based on a variety of visual features of the target in the image, extracting a small number of features to detect the existence of the target. Kalman filtering and particle filtering are tools for accurately estimating data values. Simulation results show the effectiveness of these methods.