 |
Fleet | OEM | Research | DSS |
|
Driver State Sensor - OEM |
|
 |
The Seeing Machines DSS or driver state sensor is a camera based driver monitoring system which derives information about the driver’s attention to the road and the driver’s fatigue level. The system can be readily integrated with other assistance systems such as lane departure warning and automatic cruise control.
Driver Aware Vehicles
With the emerging ability of vehicles to sense their environment, the ability of the vehicle to sense driver awareness, intent and attention becomes increasingly important. Only the correlation of potentially hazardous conditions in the environment with the state of the driver allows to intelligently dispatch warnings to the driver and operate driver assistance and safety systems most effectively. The key to driver aware vehicles is a sensor which derives information about the condition of the driver.
|
|
Driver State Sensor
The driver state sensor is a computer vision based system utilizing a tiny camera embedded in the dash board of the vehicle to observe driver's face. The image processing algorithms measure the 3D head position and orientation and the eyelid closure level of the driver. The system operates completely automatically and without the requirement to calibrate the system for individual drivers. All a driver will notice are the appropriate warnings of the vehicles driver assistance systems and less nuisance alarms. |
Distraction and Drowsiness Detection
The DSS derives driver distraction through analysis of the head motion. In particular for cognitive demanding activities, the head orientation follows the eye gaze, and is a robust indicator for the direction of the driver's attention. It is also independent of sunglasses, eyeglasses and potential reflections obscuring the minute details required to measure eye gaze.
The drowsiness level of a driver is derived from the eyelid closure level. Different fatigue metrics can be employed, and the DSS at this stage directly provides the calculation of the widely accepted PERCLOS metric as a trend indicator and the instantaneous detection of micro sleeps.
Vehicle Integration
The goal is to create safer vehicles which take the state of the driver into account. Information about distraction is valuable for the modulation of warnings from lane departure warning systems, forward collision systems and pedestrian detection systems. When combined with a driver state sensor, these systems can deliver warnings promptly when distraction is detected, and delayed or suppressed when the driver is known to observe the road scene. Similar integration potential exists for lane change assistance. Drowsiness can be utilized, beyond immediate audio warnings, to deactivate systems which are potentially dangerous when used while driving drowsy, such as cruise control and lane keeping assistance systems.
Demonstrators and Serial Production
The DSS has been specifically developed to build demonstrators and run naturalistic driving studies to evaluate driver assistance system integration. While packaged in a rugged unit with 12V-24V power supply, the unit provides all the convenience of a PC, with USB storage options, a graphical user interface available from a laptop, and data access through LAN and serial ports.
Seeing Machines is pursuing the development of a serial automotive product. For more information on this option please contact Seeing Machines. |
|
 |
|
Detection of fatigue events.
|
|
|
|
Distinct traces (green) during distraction outside road center cluster (red)
|
|

|
|
|
How the DSS works
1. The DSS continuously runs searches for the driver’s face in the image and automatically generates a model of the face that takes the unique facial features of the driver into account. This process takes less than a second.
2. Once the model is generated, the DSS begins real-time 3D head pose tracking, providing metrics on subject distraction and attention. Head pose is tracked to +/- 90 degrees of rotation.
3. The system then finds and tracks the eyelid closure, monitoring the duration of blinks and providing real-time eyelid closure level measurements.
All data is output for logging or to be input to other driver assistance systems which benefit from information on the driver’s state of fatigue and attention.
|
 |
System components
The key components of the DSS are the rugged processing unit (12V-24V power supply), the driver sensor and the unobtrusive and safe infra-red LED illumination units. Auxiliary components include a speaker for audio warnings and a GPS sensor to detect vehicle motion and speed. |


|
 |
|