PrisonSecure: A Smart Surveillance System for Prisons using Deep Learning

DOI

10.2991/978-94-6463-866-0_82How to use a DOI?

Keywords

Smart Surveillance; Prison Security; Weapon Detection; Face Recognition; Deep Learning; Real time Monitoring; Computer Vision

Abstract

Prison security needs fast and accurate systems to detect weapons and identify people in real time. Depending only on human guards is not always reliable, as they can miss threats or react slowly. In this paper, a smart surveillance system that uses deep learning to detect weapons and recognize prisoner faces from live CCTV (Closed-Circuit Television) footage. The system uses YOLOv11(You Only Look Once) to detect weapons like handguns and knives. To reduce false alarms from objects like forks or scissors, hard negative mining is applied. For face recognition, a ResNet based face detector, extract features using FaceNet, and classify faces using a Support Vector Machine (SVM). PrisonSecure performs well even in challenging conditions such as low light, crowded areas, and different face angles. The system achieved precision of 0.93, recall of 0.93, and 95.79% mAP@0.5 for weapon detection, showing high reliability. For face recognition, it reached 93.75% accuracy with very few incorrect matches. An integrated alert system provides real-time audio warnings when weapons are detected, helping security staff respond quickly to threats. Overall, PRISONSECURE shows how combining deep learning for weapon detection and face recognition can greatly improve prison surveillance, making it smarter, faster, and more dependable.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - B. Athul Krishna
AU  - R. P. Aakash
AU  - Joshua Jose
AU  - S. Anubha Pearline
PY  - 2025
DA  - 2025/10/31
TI  - PrisonSecure: A Smart Surveillance System for Prisons using Deep Learning
BT  - Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)
PB  - Atlantis Press
SP  - 1015
EP  - 1033
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-866-0_82
DO  - 10.2991/978-94-6463-866-0_82
ID  - Krishna2025
ER  -