Smart Road Accident and Fire Detection & Emergency Alert System

We developed a real-time Computer Vision system that can automatically detect bottles, identify whether a bottle has a cap, and recognize its color using a live webcam feed. This project combines YOLOv8 object detection with OpenCV image processing to create a practical AI-based inspection system for industrial automation and quality control.

This project develops an Arduino-based distance alarm system using an HC-SR04 ultrasonic sensor. The system continuously measures the distance of nearby objects. Green, yellow, and red LEDs indicate safe, warning, and danger zones, while a buzzer sounds when an object is too close. The project demonstrates how sensors, LEDs, and buzzers can work together through Arduino programming.

The AI-Enabled IoT System for Smart Surveillance and Security is a smart four-layer door-locking project designed to improve safety in homes, offices, laboratories, and other restricted areas. The system combines RFID card verification, password entry, fingerprint scanning, and face recognition before giving access. An ESP32 microcontroller controls the hardware, while Firebase stores and updates data in real time. A Flutter mobile application allows users to check the lock status, manage users, update passwords, and control the system remotely. This project brings together IoT, artificial intelligence, biometrics, cloud technology, and mobile development in one practical security solution.

GLUCOTWIN is an AI-driven healthcare application developed to support diabetes management in children aged 3 to 12 years. The system helps guardians by predicting insulin dosage based on important health factors such as blood pressure, body temperature, BMI, blood sugar condition, meal timing, and carbohydrate intake. The application is built using Flutter and Firebase, while a trained machine learning model processes health inputs and returns insulin predictions in real time. To improve safety and reliability, doctors can review the predicted results and provide medical recommendations. The system offers a structured, user-friendly, and intelligent platform for better child diabetes care

This project presents a smart Sign Language Recognition and Translation System designed to reduce the communication gap between sign language users and non-sign language users. The system supports two-way communication by converting text and speech into sign language and translating sign language into text and speech. It uses machine learning and computer vision techniques to recognize hand signs and gestures in real time. YOLOv8 is used for sign alphabet detection, while MediaPipe is used for hand landmark tracking and gesture analysis. The system also stores gesture patterns and hand angle data in JSON format for sentence-level recognition. A Flask-based web interface is used to provide an interactive and user-friendly experience. This project offers a practical solution for real-time communication, accessibility, and learning support.

NeuroLearn – AI Powered E-Learning Platform