Real-Time Firearm Detection from CCTV Footage

Tech Stack: Python, OpenCV, Machine Learning, YOLOv5
Github URL: <a href="https://github.com/dhvani-k/Firearm_Detection_In_CCTVs_Using_YOLOv5" target="_blank">Project Link

This project aims to detect firearms in real-time from CCTV videos, highlighting them within bounding boxes for immediate attention.

Trained YOLO V3 object detection model with 3000+ images identifying firearms in, videos or live footage from CCTV with 92% accuracy.

Improved accuracy of detection by 10% by suppressing redundant bounding boxes and eliminating false positive detections.