Direct observation event triggering of drowsiness

1. A system for indicating an event, comprising:

an interface configured to:

receive a face tracking data for a driver of a vehicle;

receive a sensor data for the vehicle; and

a processor configured to:

merge the face tracking data and the sensor data to obtain a feature vector, comprising to:

evaluate a linear function or a nonlinear function of the face tracking data and the sensor data to obtain relevant features; and

assemble the relevant features into the feature vector;

determine a degree of drowsiness using a first threshold and a second threshold for the driver of the vehicle based at least in part on the feature vector;

in response to a determination that the degree of drowsiness is greater than a first threshold, indicate an event pertaining to the driver of the vehicle, comprising to:

trigger data capture;

in response to a determination that the degree of drowsiness is less than or equal to the first threshold:

omit indicating an event pertaining to the driver of the vehicle;

determine whether a driving time is equal to or exceeds a third threshold; and

in response to a determination that the driving time is equal to or exceeds the third threshold, warn the driver.

2. The system as in claim 1 , wherein data associated with the event is stored.

3. The system as in claim 2 , wherein the data comprises camera data.

4. The system as in claim 2 , wherein the data comprises sensor data.

5. The system as in claim 2 , wherein the data comprises audio data.

6. The system as in claim 2 , wherein the data comprises captured data.

7. The system as in claim 1 , wherein the data is transmitted.

8. The system as in claim 7 , wherein the data is transmitted to a vehicle data server.

9. The system as in claim 1 , wherein the degree of drowsiness is determined using a model.

10. The system as in claim 9 , wherein the model is one of the following: rule-based model, statistical model, neural network based model, or an automatically calibrating anomaly detection model.

11. The system as in claim 9 , wherein the model is trained using one or more of the following: driving simulator data, video data, sensor data, audio data, or captured data.

12. The system as in claim 9 , wherein the model is used to determine a drowsy driving behavior using only non-face-data sensors.

13. The system as in claim 9 , wherein the model is used to determine a real-time driver warning using only non-face-data sensors.

14. The system as in claim 1 , wherein sensor data comprises one or more of the following: steering data, acceleration data, braking data, application of gas data, deviations data, jerkiness of responses data, corrections data, or overcorrections data.

15. The system as in claim 1 , wherein the driver is warned using one or more of the following an audible warning, an illumination warning, or a haptic warning.

16. The system as in claim 1 , wherein the processor is further configured to warn one or more of the following: a dispatcher, a system administrator, or a supervisor.

17. The system as in claim 1 , wherein in the event that the degree of drowsiness is greater than the second threshold, indicate a warning.

18. The system as in claim 1 , wherein the third threshold is determined based on time of day.

19. A method for indicating an event, comprising:

receiving a face tracking data for a driver of a vehicle;

receiving a sensor data for the vehicle;

merging the face tracking data and the sensor data to obtain a feature vector, comprising:

evaluating a linear function or a nonlinear function of the face tracking data and the sensor data to obtain relevant features; and

assembling the relevant features into the feature vector;

determining, using a processor, a degree of drowsiness using a first threshold and a second threshold for the driver of the vehicle based at least in part on the feature vector;

in the event that the degree of drowsiness is greater than a first threshold, indicating an event pertaining to the driver of the vehicle, comprising:

triggering data capture;

in response to a determination that the degree of drowsiness is less than or equal to the first threshold:

omitting indicating an event pertaining to the driver of the vehicle;

determining whether a driving time is equal to or exceeds a third threshold; and

in response to a determination that the driving time is equal to or exceeds the third threshold, warning the driver.

20. A computer program product for indicating an event, the computer program product being embodied in a tangible computer readable storage medium and comprising computer instructions for:

receiving a face tracking data for a driver of a vehicle;

receiving a sensor data for the vehicle;

merging the face tracking data and the sensor data to obtain a feature vector, comprising:

evaluating a linear function or a nonlinear function of the face tracking data and the sensor data to obtain relevant features; and

assembling the relevant features into the feature vector;

determining a degree of drowsiness using a first threshold and a second threshold for the driver of the vehicle based at least in part on the feature vector;

in the event that the degree of drowsiness is greater than a first threshold, indicating an event pertaining to the driver of the vehicle, comprising:

triggering data capture;

in response to a determination that the degree of drowsiness is less than or equal to the first threshold:

omitting indicating an event pertaining to the driver of the vehicle;

determining whether a driving time is equal to or exceeds a third threshold; and

in response to a determination that the driving time is equal to or exceeds the third threshold, warning the driver.