Overview / Usage

According to the CDC motor vehicle safety division, one in five car accidents is caused by a distracted driver. Sadly, this translates to 425,000 people injured and 3,000 people killed by distracted driving every year. Vehicle operators behind the wheel need to avoid unsafe behaviors, such as talking to co-passengers, using mobile phones, and eating.

BONANG.ai hopes to improve these alarming statistics, and enhance insurance telematics, by turning smartphones into AI dashboard cameras to automatically detect drivers engaging in distracted behaviors. Are they driving attentively, wearing their seatbelt, or taking a selfie with their friends in the backseat?

Methodology / Approach

BONANG.ai platform turns a smartphone into a vehicle Dash Camera Systems powered by Intel OpenVINO Toolkit Inference Engine, to automatically detect drivers engaging in distracted behaviors. BONANG.ai is accelerated by Action Recognition algorithms, that classifies actions that are being performed on input video stream obtained from a smartphone using Intel WebRTC and Open Visual Cloud building blocks (SMART CITY TRAFFIC MANAGEMENT SAMPLE)

1. Activity Recognition for driver monitoring scenario. It recognizes 400 general actions, also including safe driving, talking to the phone and many more.
2. Unsupervised learning to detect traffic related anomalies such as lane violation, illegal U-turns, wrong-direction driving.
3. Supervised learning models for Person Detection network followed by the Person Attributes Recognition and Person Re-identification networks applied on top of the detection results.
4. Deployment of the Trained Models is done using Intel OpenVINO Toolkit

Technologies Used

Intel Xeon

Intel OpenVINO Toolkit

Open Visual Cloud

Intel IoT DevCloud

WebRTC

AWS

​

Collaborators