Driver's fatigue detection system and method

• US 9,402,577 B2
• Filed: 12/23/2014
• Issued: 08/02/2016
• Est. Priority Date: 12/23/2014
• Status: Active Grant

• Current Assignee: Automotive Research & Testing Center
• Original Assignee: Automotive Research & Testing Center
• Inventors: Feng, Yen-Cheng, Ko, Ming-Kuan
• Primary Examiner(s): PHAM, TOAN NGOC
• Application Number:
  US14/580,629
• Publication Number:
  US 20160174890A1
• Time in Patent Office: 588 Days
• Field of Search: 340/576, 340/573.1, 340/575, 180/272
• US Class Current: 1/1
• CPC Class Codes: A61B 5/0077 Devices for viewing the surface of the body, e.g. camera, magnifying lens
• CPC Class Codes: A61B 5/18 for vehicle drivers or machine operators
• CPC Class Codes: A61B 5/6893 Cars
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7282 Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A61B 5/746 Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms alarms per se G08B
• CPC Class Codes: B60W 2040/0827 due to sleepiness
• CPC Class Codes: B60W 40/08 related to drivers or passengers
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

The present invention discloses a driver'"'"'s fatigue detection system and method. The system comprises a vital sign detection device generating at least one personal vital sign value; a storage device storing a linear statistic equation; a processor electrically connected with the vital sign detection device and the storage device; and a display device electrically connected with the processor. The processor receives the personal vital sign value, retrieves the linear statistic equation, substitutes the personal vital sign value into the linear statistic equation to generate a predictive vehicle deviation value, and determines whether the predictive vehicle deviation value is over a preset vehicle deviation value. If yes, the processor generates an alert signal to the display device to present an alert image. The present invention determines whether the stability of the driver is decreasing according to his physiological status and reminds the driver to avoid a traffic accident beforehand.

Claims

1. A driver'"'"'s fatigue detection method, comprising

Step (A):

a detection device inputting a plurality of reference vital sign values and a plurality of reference vehicle deviation values to a processor, and said processor processing said reference vital sign values and said reference vehicle deviation values statistically to generate a linear statistic equation;

Step (B);

said detection device inputting at least one personal vital sign value to said processor, and said processor substituting said personal vital sign value into said linear statistic equation to generate a predictive vehicle deviation value; and

Step (C);

said processor determining whether said predictive vehicle deviation value is over a preset vehicle deviation value;

if yes, said processor generating an alert signal;

if no, returning to said Step (B).

2. The driver'"'"'s fatigue detection method according to claim 1, wherein in said Step (A), a plurality of reference driver'"'"'s image values is also input to said processor in addition to said reference vital sign values and said reference vehicle deviation values, and wherein if one said reference driver'"'"'s image value indicates a driver is splitting his attention and if one said reference vehicle deviation value detected at an identical time point indicates that a vehicle driven by said driver is deviating from a lane, said reference driver'"'"'s image value and said reference vehicle deviation value are filtered out.

3. The driver'"'"'s fatigue detection method according to claim 1, wherein in said Step (B), a personal driver'"'"'s image value is also input to said processor in addition said personal vital sign value;

said processor also substitutes said personal driver'"'"'s image value into said linear statistic equation to generate said predictive vehicle deviation value.

4. The driver'"'"'s fatigue detection method according to claim 3, wherein in said linear statistic equation, said predictive vehicle deviation value equals to said personal vital sign value multiplied by a slope plus said personal driver'"'"'s image value and plus a constant.

5. The driver'"'"'s fatigue detection method according to claim 1, wherein in said Step (C), said preset vehicle deviation value is one of a low-level vehicle-deviation alert value and a medium-level vehicle-deviation alert value;

if said predictive vehicle deviation value is below said low-level vehicle-deviation alert value, said processor generates a low-level alert signal;

if said predictive vehicle deviation value is between said low-level vehicle-deviation alert value and said medium-level vehicle-deviation alert value, said processor generates a medium-level alert signal;

if said predictive vehicle deviation value is over said medium-level vehicle-deviation value, said processor generates a high-level alert signal.

6. The driver'"'"'s fatigue detection method according to claim 1, wherein in said Step (B), said processor receives a personal vehicle deviation value and said personal vital sign value, and uses said personal vehicle deviation value and said personal vital sign value to update said linear statistic equation.

7. A driver'"'"'s fatigue detection system comprising

a vital sign detection device generating at least one personal vital sign value;

a storage device storing a linear statistic equation;

a processor electrically connected with said vital sign detection device and said storage device, receiving said personal vital sign value, retrieving said linear statistic equation, substituting said personal vital sign value into said linear statistic equation to generate a predictive vehicle deviation value, determining whether said predictive vehicle deviation value is over a preset vehicle deviation value, and generating an alert signal if said predictive vehicle deviation value is over said preset vehicle deviation value; and

a display device electrically connected with said processor, receiving said alert signal, presenting an alert image according to said alert signal.

8. The driver'"'"'s fatigue detection system according to claim 7, wherein said display device includes an audio element that is electrically connected with said processor, receives said alert signal, and generates an alert sound according to said alert signal.

9. The driver'"'"'s fatigue detection system according to claim 7, wherein a driver'"'"'s image detection device is electrically connected with said processor and generates a personal driver'"'"'s image values to said processor;

said processor also substitutes said personal driver'"'"'s image value into said linear statistic equation to generate said predictive vehicle deviation value.

10. The driver'"'"'s fatigue detection system according to claim 9, wherein in said linear statistic equation, said predictive vehicle deviation equals to said personal vital sign value multiplied by a slope plus said personal driver'"'"'s image value and plus a constant.

11. The driver'"'"'s fatigue detection system according to claim 7, wherein said preset vehicle deviation value is one of a low-level vehicle-deviation alert value and a medium-level vehicle-deviation alert value;

if said predictive vehicle deviation value is below said low-level vehicle-deviation alert value, said processor generates a low-level alert signal to said display device to enable said display device to present a low-level alert image;

if said predictive vehicle deviation value is between said low-level vehicle-deviation alert value and said medium-level vehicle-deviation alert value, said processor generates a medium-level alert signal to said display device to enable said display device to present a medium-level alert image;

if said predictive vehicle deviation value is over said medium-level vehicle-deviation value, said processor generates a high-level alert signal to said display device to enable said display device to present a high-level alert image.

12. The driver'"'"'s fatigue detection system according to claim 7 further comprising a vehicle deviation detection device electrically connected with said processor, wherein said vehicle deviation detection device generates a plurality of reference vehicle deviation values, and said vital sign detection device generates a plurality of reference vital sign values, and wherein said processor uses said reference vital sign values and said reference vehicle deviation values to generate said linear statistic equation.

13. The driver'"'"'s fatigue detection system according to claim 12, wherein said vehicle deviation detection device also generates at least one personal vehicle deviation value to said processor;

said processor uses said personal vehicle deviation value and said personal vital sign value to update said linear statistic equation.