This project demonstrates a comprehensive solution for detecting weapons, such as knives and guns, in images, videos, and live webcam feeds using the YOLOv8 object detection model. The system provides real-time notifications and actionable guidance through an integrated web application.

• Object Detection: Utilizes YOLOv8 for real-time identification of weapons.
• Web Application: Built with Flask, providing an intuitive interface for uploading media and viewing results.
• Real-Time Notifications: Sends alerts to users via Telegram when a weapon is detected.
• AI-Powered Guidance: Integrates CharacterAI to offer contextual suggestions upon detection.
• Versatility: Supports static images, video files, and live webcam feeds.

• YOLOv8: A state-of-the-art object detection model.
• Flask: Lightweight web framework for building the application interface.
• Telegram API: For sending instant notifications.
• CharacterAI: Provides AI-driven recommendations to users.
• OpenCV: For image and video processing.
• TensorFlow & PyTorch: Deep learning frameworks for training and fine-tuning detection models.

1. Upload Media: Users can upload images or videos via the web interface or enable live webcam monitoring.
2. Detection: The YOLOv8 model identifies weapons and marks them with bounding boxes.
3. Notifications: Alerts are sent through Telegram with detection results and actionable suggestions.
4. Real-Time Monitoring: The system provides live feedback for webcam streams.

This project is based on the Weapons and Knives Detector with YOLOv8 model by João Assalim.
I have utilized this model as a foundation and made modifications to adapt it for my use case, including integrating real-time notifications, a web interface, and AI-driven recommendations.