Abstract: The use of Unmanned Aerial Vehicles (UAVs) has brought drastic security issues in areas considered sensitive like defense zones, industrial installations, and restricted locations. This paper is a double-layered intelligent surveillance system with AI-based drone detection coupled with IoT – enabled ground object monitoring. The air detection module uses a Convolutional Neural Network (CNN) to analyze live video streams and detect unauthorized flying objects like drones and the ground detection module uses a Node MCU- driven ultrasonic sensor with a servo motor for 180° scanning. Both modules use Fire based cloud services to synchronized at a in real-time and send instant mobile notifications to authorized personnel. The system provides end-to-end aerial and ground-level monitoring, providing scalability, cost-effectiveness, and quick response, and can thus be deployed to high- security contexts including borders, airports, and military bases.

Keywords: Drone images, Node MCU, NN, Ultrasonic sensor, Drone Detection / IoT Surveillance YOLOv3, Real-Time Monitoring, Intrusion Detection, Surveillance System, Video Frame Analysis, Sensor Fusion, Node MCU, Firebase, Inertial Sensors, Object Recognition, Obstacle Avoidance, Cloud Alerting.


Downloads: PDF | DOI: 10.17148/IJARCCE.2022.11832

How to Cite:

[1] K Kishor Kumar, "Convolutional Neural Networks for Classification of Aerial Images," International Journal of Advanced Research in Computer and Communication Engineering (IJARCCE), DOI: 10.17148/IJARCCE.2022.11832

Open chat
Chat with IJARCCE