Gadget Distraction Solution Behind the Wheels (DoDD 3.0 Release)

Overview

As part of my 2019 Science fair project recommendation, I proposed ‘retina scanning solution’ and alerting the driver when they get distracted behind the wheels. In 2020, my Objective was to design an Engineering Prototype – a device with a camera, which will be placed above the steering wheel to track the retina and warn drivers when they get distracted by gadgets. For this, I used Raspberry pi 4 B, and coded in Python3 using OpenCV libraries, along with couple of Haars Classifiers XML, containing training data on face recognition & eye landmarks. I named this prototype as “Detection of Distracted Driving” or DoDD 1.0. In 2020, I recommended future work to be ‘SMS alerts if distracted ≥ 3 times’, to family member of the driver. Apart from this, I increased the accuracy of With & Without Glasses and Infrared (IR) Night mode (NV) filter from DoDD 1.0. In 2022, I implemented SMS alerts if distracted > 3times and added a bonus feature called “Drowsiness detection”, DoDD 2.0. In 2023, I implemented Emotion detection, primarily “Anger Detection”, to help prevent road rage, and released DoDD 3.0. Testing Procedure for evaluating DoDD 3.0’s feature of Emotion detection, I tested with With Glasses, Without Glasses and Sunglasses. They were compared with the expected results, along with sounds and display on the LCD screen. For good driving, “Good Driver” text with expected on LCD. Result In Without Glasses, the success rate was 78% compared to With Glasses, which was 67%, and for With Sunglasses, it was only 22%. The Classifier couldn’t decipher the images with the glasses and sunglasses clearly. Therefore, for Anger detection testing needs to be carried out with more machine learning datasets involving glasses and sunglasses, and then test it again. Conclusion is overall the accuracy level for identifying Distracted driving & Anger Detection, was higher for Without Glasses compared to With Glasses & With Sunglasses scenarios. DoDD’s 3.0 development was more challenging with multiple dependencies on python libraries. Though it was challenging, the crux of the project was - it thought me how to bring awareness on ROAD RAGE and DISTRACTIONS in today’s world. I feel like this topic has many more avenues to explore and research on. Which will help save more lives, as ANGER and MENTAL HEALTH has been one of the fatal crises of our society today.

Awards (1)

• Category Award: HM

Competition history

• CSEF 2023 Behavioral & Social Sciences · Entry S0427

Resources

• CSEF published award list ↗
• CSEF published award list ↗
• CSEF published award list ↗
• CSEF public showcase (Internet Archive capture) ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Behavioural and Social Sciences projects
• CSEF Behavioural and Social Sciences projects
• Every CSEF 2023 project