Driver State Monitoring
Introduction
Driver State Monitoring (DSM) Engine detects a user's face, tracks their gaze, and determines whether their eyes are closed or blinking. This technology is based on video image data that tracks information regarding face position, head direction, facial features, gaze direction/angle, and the degree to which the eyes are closed.
The engine provides a function to register, recognize, or unregister a user's face based on tracking information. It can also detect if the camera is blocked.
DSM Engine provided by LG AI Platform supports the following features:
Feature | Description |
---|---|
Detecting the presence of a face |
Determines if a user’s face is present in the video |
Tracking facial features |
Tracks a user's facial features in the video, such as eyes, nose, mouth, and facial lines |
Tracking facial position and head direction |
Uses 3D tracking to determine facial position and head direction in the video |
Tracking a user's gaze |
Tracks a user's gaze and determines gaze direction |
Determining whether a user's eyes are closed |
Determines the degree to which a user’s eyes are closed |
Providing a function to register, recognize, and unregister a user’s face |
Allows you to register or unregister a user's face through face recognition |
Determining the degree to which the camera is blocked |
Determines the degree to which the camera is blocked by an obstacle in front of the user |
Engine Structure
DSM Engine receives video input recorded by the camera or saved in storage media, analyzes the video, and detects relevant information before sending the result to an application.
It can receive video images and engine settings as input data, and uses them for image analysis.
Examples of Use
DSM Engine is used in automobiles that require video-based services.
- Drowsiness detection
This tracks a driver’s facial features, gaze direction, and blinking pattern to judge if the driver is falling asleep.
- Driver inattention detection
This tracks a driver’s facial features, gaze direction, and blinking pattern to judge if the driver is inattentive.
- User authentication
This recognizes a driver's facial features to judge whether the driver is a registered user.