Advanced Surveillance System Using Internet of Things
Year: 2025
Type: article
Abstract: As security concerns increase across industries such as banking, education, military, and commercial sectors, traditional surveillance systems struggle to effectively distinguish between authorized personnel and intruders. This project aims to address these challenges by developing an Advanced Surveillance Detection System powered by deep learning algorithms and IoT technologies. The system utilizes Convolutional Neural Networks (CNNs) for uniform recognition, distinguishing between authorized individuals (e.g., soldiers, staff) and unauthorized persons, even if they wear similar uniforms. In addition, people counting mechanisms track the number of individuals in the surveillance area, ensuring any anomalies—such as unexpected increases in personnel—trigger real-time alerts. The system seamlessly integrates with IoT devices like buzzers to provide immediate notifications to the control platform. By automating surveillance, the project enhances operational security across multiple environments, enabling both active patrol and after-hours monitoring. This solution offers a scalable, efficient, and reliable approach to modern security management by combining deep learning and IoT to reduce vulnerabilities and ensure proactive threat mitigation. Keywords: Surveillance System, Deep Learning, CNN, IoT, Uniform Recognition, People Counting, Real-Time Alerts, Security Management, Anomaly Detection
DOI: 10.55041/ijsrem41745
Author: Aniket K Katare
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4407845265