Fall Detection Application Using Deep Learning

Project Code :TCMAPY2129

Objective

The objective of the AI-Powered Fall Detection System is to develop an efficient, real-time solution for detecting falls using deep learning models, specifically YOLOv11 and YOLOv12. The project aims to provide a non-intrusive, accurate, and scalable system for fall detection without the need for wearable devices. It focuses on integrating computer vision techniques with AI to detect falls from images and live video streams. The system’s key objectives include ensuring high accuracy, providing actionable fall alerts, enabling real-time processing, offering user-friendly access through a web interface using flask framework, and facilitating easy monitoring through historical detection tracking on images.

Abstract

The "Fall Detection Application Using Deep Learning" project aims to design an automated system that identifies fall events using advanced image processing and deep learning algorithms. The system utilizes a dataset from Roboflow, containing two classes: "fall" and "no-fall." For accurate detection, the project leverages YOLO11n and YOLO12n models, which have shown promising performance. The application features modules such as image and live detection, along with a user-friendly interface for login, registration, and history viewing. Built with Flask for the backend and HTML, CSS, and JavaScript for the frontend, the system ensures easy navigation and real-time fall detection. The project's core objective is to provide a reliable solution for fall detection in controlled environments. The developed system can potentially be extended for various applications where detecting falls is crucial for safety monitoring.

Keywords: Fall Detection, YOLO, Deep Learning, Flask, Image Detection, Real-time Detection, Python, SQLite, Fall Dataset, Application

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.