Weapon Detection System

[venkat-2811]

Related Issue

NOTE: Kindly write in the following format -

Closes #51

Description

Weapon Detection Using YOLOv3

Project Overview

This project utilizes the YOLOv3 (You Only Look Once) algorithm for real-time weapon detection. The code supports two modes of operation:

1. Webcam-based weapon detection: Uses the webcam to detect weapons in real-time.
2. Video file-based weapon detection: Uses a pre-recorded video file to detect weapons.

Features

• Real-time detection of weapons using either a webcam or a video file.
• Display of bounding boxes and labels for detected objects.
• Customizable detection thresholds.

Technologies and Libraries Used

• Python (version 3.x)
• OpenCV: Library for computer vision tasks.
• NumPy: Library for numerical computations.
• YOLOv3: Pre-trained deep learning model for object detection.

Type of PR

• [ ] Bug fix
• [ ] Feature enhancement
• [ ] Documentation update
• [X] Other (specify): New Model for Weapon Detection

Screenshots / videos (if applicable)

The following image demonstrates the output of the YOLOv3 model, showing bounding boxes and labels for detected objects:

Checklist:

• [X] I have performed a self-review of my code
• [X] I have read and followed the Contribution Guidelines.
• [X] I have tested the changes thoroughly before submitting this pull request.
• [X] I have provided relevant issue numbers, screenshots, and videos after making the changes.
• [X] I have commented my code, particularly in hard-to-understand areas.

Additional context:

Customization

You can customize the detection thresholds and other parameters within the script to suit your specific requirements. Detailed comments within the code will guide you through making these adjustments.

Contact

For further assistance or inquiries, please reach out via the repository's contact information.

DESCRIPTION

This project is a real-time weapon detection system utilizing OpenCV and YOLO (You Only Look Once) object detection framework. The system is designed to detect various weapons such as knives, guns, and bombs using a pre-trained YOLOv3 model. The implementation involves loading the YOLO model with custom-trained weights and configuration files, capturing live video feed from the camera, and processing each frame to detect and highlight weapons. The detection results are displayed on the screen with bounding boxes and labels around the detected weapons. This tool is potentially useful for enhancing security measures in public spaces by providing an automated method for weapon detection.

[Aryan-Chharia]

@venkat-2811 dont upload the zip file. See the rest of the projects in the repo and upload like them

[venkat-2811]

Okay sir , my bad. will raise a correct one.