Tracking Missing Person in Large Crowd Gathering Using Intelligent Video Surveillance.
Adnan Nadeem1, Muhammad Ashraf2, Nauman Qadeer3
1Faculty of Computer and Information System, Islamic University of Madinah, Madinah 42351, Saudi Arabia.
Sensors (Basel, Switzerland)
|July 27, 2022
Summary
This study introduces an intelligent system for tracking missing persons in crowded areas using face recognition in low-resolution videos. The mechanism effectively locates individuals in challenging, unconstrained environments, improving public safety.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Public Safety
Background:
- Locating missing persons in large crowds via video face recognition is difficult due to dynamic factors.
- Existing solutions often require high-resolution images, limiting their applicability in real-world scenarios.
Purpose of the Study:
- To develop an intelligent mechanism for tracking missing persons in unconstrained, large gathering scenarios.
- To address the challenge of face recognition in low-resolution video footage.
Main Methods:
- A four-phase approach: online reporting with spatio-temporal features, geo-fencing for search space reduction, enhanced face detection using fused algorithms (Viola Jones, LBP, CART, HAAR), and face recognition.
- Utilized a dataset of 2208 low-resolution images from a large crowd gathering.
Main Results:
- The proposed mechanism demonstrated good performance in tracking missing persons within a challenging, low-resolution, large-crowd environment.
- The fusion of detection algorithms optimized face region localization.
Conclusions:
- The intelligent tracking mechanism is effective for locating missing individuals in complex, unconstrained settings.
- This system offers a viable solution for enhancing safety and security in large public gatherings.
Keywords:
Viola Jones cascades fusionfaces detectionfacial recognitionintelligent video surveillancelarge crowd gatheringmissing persons trackingspatio-temporal featuresunconstrained environment
