AUTOMATIC DETECTION AND TRACKING OF PEOPLE IN VIDEO SURVEILLANCE
Guru Selvi.D 1, Vijay Anandh.T 2
1 PG Student, Department of ECE, National Engineering College, Kovilpatti, Tamil Nadu, .India.
2 Assistant Professor, Dept of ECE, National Engineering College, Kovilpatti, Tamil Nadu, .India.
Object detection and tracking are necessary in many computer vision applications such as surveillance, vehicle navigation, and autonomous robot navigation. In this paper video segmentation technique that accurately segments the moving objects using Gaussian Mixture model is introduced. In a video footage, the moving human beings are segmented with the help of GMM, then detected silhouettes can be segmented into head, torso, and leg parts of a human in order to reduce confusion that may occur between different individuals in case of global descriptors. This work proposes the spatial segmentation method for head, torso and leg segmentation. As a novel implementation, this work also genesis a 2D-trajectory plotting for analyzing the detected human’s movement over a time period. This is possible to calculate and plot trajectories by calculating the centroid of the silhouettes. This proposed work can coordinate the multiple human detection and able to plot their individual trajectory over a horizon.
