Raksha alert: Intruder Detector

Raksha Alert is a comprehensive, unified home security platform that integrates multiple advanced technologies to provide real-time intruder detection and emergency response capabilities. By combining facial recognition, object detection, and voice-activated emergency alerts, it addresses the limitations of traditional security systems, ensuring a safer and smarter home environment.

• Facial Recognition and Facial Expression Recognition
• Hazardous Object Detection
• Voice-Activated Emergency Response
• Real-Time Alerts
• Personalized Chatbot

• Python: The main programming language used for backend development.
• OpenCV: Used for processing images and video streams in real-time.
• Deep Learning: Machine learning models for face recognition and emotion detection.
• Flask: A lightweight web framework for creating the alert system.
• YOLOv4: A state-of-the-art object detection model for detecting threats like weapons.
• Geocoder: For geolocation services.

• Webcam: For video input and real-time monitoring.
• Server: To process data and host the system.
• High-speed Internet: Necessary for real-time data transmission and communication with the cloud.

• Face_recognition: Pre-trained face recognition model.
• DeepFace: Pre-trained facial expression recognition model.
• YOLOv4: Object detection framework.
• Flask: Used to deploy the system as a web application.
• Geocoder: For geolocation tracking.