Activity Flow - Human Activity Recognition

Project Name:

Enhancing HAR Model Accuracy through Multimodal Sensor Data Integration

Technologies:

Machine Learning, Android Development - Java, Backend - Flask, Frontend - React JS, JavaScript, JSX, Bootstrap

Overview:

A human activity recognition app is a software tool that uses sensors such as accelerometers, gyroscopes, and magnetometers to monitor and track the activities of the user. The app provides precise information about the activity performed by the user throughout the day with 90% accuracy. The app can recognize a wide range of activities such as walking, running, cycling, sitting, standing, and even sleeping.

The app works by collecting data from the sensors and using machine learning algorithms to analyze and classify the activity being performed. The algorithms are trained on a large dataset of labeled activity data, which allows the app to accurately recognize and classify activities with high accuracy.

The app provides real-time feedback to the user about their activity level and can also provide alerts and reminders to encourage them to stay active throughout the day. The app can also track the user's progress over time, allowing them to set goals and monitor their progress towards achieving them.

Lookup:

• Home Page

• Login Page

• Dashboard Page

• Profile Page

• About Page

• Projectmates Page

• Download App Page

Installation Instructions:

Prerequisites:

• Python 3.7 or higher
• Android Studio
• Flask
• React JS

Backend Setup:

1. Clone the repository

git clone https://github.com/jacktherock/ActivityFlow.git

2. Navigate to the backend directory

cd ActivityFlow/har_backend

3. Create a virtual environment

python -m virtualenv venv

4. Activate the virtual environment

venv\Scripts\activate

5. Install the required Python packages

pip install -r requirements.txt

6. Set the necessary environment variables

export FLASK_APP=route.py
export FLASK_DEBUG=1

7. Run Flask app

python run_flask.py

Frontend Setup:

1. Navigate to the frontend directory

cd ActivityFlow/har_frontend

2. Install the required dependencies

npm install

Research Paper:

2023 IJNRD | Volume 8, Issue 5 May 2023 | ISSN: 2456-4184 | IJNRD.ORG

Abhijeet Sonawane, "Enhancing HAR Model Accuracy through Multimodal Sensor Data Integration", 2023 IJNRD, Volume 8, Issue 5 May 2023, ISSN: 2456-4184

Contributors:

• Pratik Chopane
• Nikhil Chinchore
• Harshada Shelke

Thank You !