Multiple human tracking using Blob analysis andKalman filtering

Conference: Third International Conference on Current Trends in Engineering Science and Technology
Author(s): Jagadeesh B, Chandrashekar M Patil Year: 2017
Grenze ID: 02.ICCTEST.2017.1.151 Page: 866-873

Abstract

Computer Vision research field is gaining more importance with wide applications in video surveillance, video retrieval and analysis. Video surveillance provides continuous monitoring enhancing security and control. The proposed algorithm detects the moving object; classify the moving object as human and keep track of the human. Human tracking plays an integral role in many fields of human–machine interaction. Even though there is increase of progress in this field, visual tracking still remains a challenging task. Proposed tracking algorithm uses a combination of Blob extraction and Kalman filtering. Precise tracking of moving human beings, identification and estimation of their future location in an unknown scene are the main objectives of the proposed work. The input video is split into individual frames. Blob extraction is used for modeling followed by Kalman filtering where positions of humans are detected and tracked, followed by validation.

<< BACK

ICCTEST - 2017