Welcome to DRIVE AI

Developed by :

Implementing next gen Vehicle Security & Safety using ML & Computer Vision

Deployed on HEROKU for easy access via mobile phones

Drive AI provides a 3 fold AI facility for your vehicle

  • Distance Estimation & Classification - Back Camera Object Detection
  • Mood Recognizer using Facial Expressions - Music Player
  • Drowsiness Detection - Alarm Notification!

Part 1 : Stay Alert When You Drive!

How to resolve Drowsiness Detection?

  • Capture facial landmarks and features
  • Mathematically estimate the Eye Aspect Ratio in real time frames
  • Monitor the variation in EAR over a select number of frames and buzz the alarm when set condition is met

Demo Run on Drowsiness Detection - Part 1

Demo Run on Drowsiness Detection - Part 2

Demo Run on Drowsiness Detection - Part 3 : Web App Deployed

Part 2 : We Analyse Your Expressions & Auto-Play Mood Based Music!

How Drive AI detects mood based on expression to play music :

  • Train a dataset of human faces that depict 6 classes of expressions
  • Gather fine and broad features using CNN archicture inspired by ResNet50 Architecture
  • Based on predicted class, fetch music from database and play it

Part 3 : Be Aware Of Objects In Surroundings, Track and Estimate Distance In Real Time

How to estimate distance and classify objects ?

  • Using a combination of regression and classification to create bounding boxes + labels
  • YOLO-tiny model is trained on a dataset to accomplish this task
  • Pinhole Camera Principle helps in estimating distance through triangular similarity

Demo Run on Object Detection - Locally Executed

Why is Drive AI an essential tool for vehicles in 2019?

  • A simplified yet optimal safety tool that utilises fast ML/AI algorithms
  • Requires no explicit interaction from user end
  • Mood based music player improves overall UX and adds to the functionality

The USP - Convenience, Functionality & State-Of-The-Art Technology

We aim to develop a local yet efficient software solution that can create impact in the community and help propogate the novel uses of ML/AI

UI For The Deployed Application : Main Menu

Tech - Stack used by Drive AI :

Front End :

  • Angular 7
  • Bootstrap
  • HTML5
  • SCSS
  • Typescript

Back End :

  • Flask RESTful
  • Python
  • OpenCV
  • Numpy, Scipy, Pandas, Sklearn
  • Keras, Tensorflow
  • MongoDB Atlas

IDEs & Other Resources :

  • Jupyter Notebook
  • VS Code
  • Google Colaboratory
  • Sublime Text
  • Kaggle

Roadmap For The Future :

  • Completing the optimised Progressive Web App (PWA) to enhance feasibiity and testing of the product
  • Integrating the 3 features with portable hardware : Raspberry Pi & USB cameras on the dash
  • Making the project open source and available to the community for further development

Video Demo for Object Detection : On Campus