Drowsiness detection method and associated device

• US 9,113,837 B2
• Filed: 10/29/2012
• Issued: 08/25/2015
• Est. Priority Date: 01/06/2012
• Status: Active Grant

• Current Assignee: Wistron Corporation
• Original Assignee: Wistron Corporation
• Inventors: Li, Chia-Hsien, Chang, Yao-Tsung, Lin, Pai-Yang
• Primary Examiner(s): PHAM, QUANG
• Application Number:
  US13/662,968
• Publication Number:
  US 20130176129A1
• Time in Patent Office: 1,030 Days
• Field of Search: 340/575
• US Class Current: 1/1
• CPC Class Codes: A61B 2503/22 Motor vehicles operators, e.g. drivers, pilots, captains
• CPC Class Codes: A61B 5/1103 Detecting eye twinkling
• CPC Class Codes: A61B 5/18 for vehicle drivers or machine operators
• CPC Class Codes: A61B 5/377 using evoked responses
• 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: B60R 21/01552 detecting position of specific human body parts, e.g. face, eyes or hands
• CPC Class Codes: G06V 20/597 Recognising the driver's state or behaviour, e.g. attention or drowsiness
• CPC Class Codes: G08B 21/06 indicating a condition of sleep, e.g. anti-dozing alarms
• CPC Class Codes: G08B 23/00 Alarms responsive to unspecified undesired or abnormal conditions

Abstract

A drowsiness detection method for detecting drowsiness of a detection target is provided. The method includes a data collecting process including steps of: detecting physiological symptom information of the detection target; determining whether the detection target satisfies a predetermined drowsiness threshold condition according to the physiological symptom information; when the detection target satisfies the predetermined drowsiness threshold condition, triggering a stimulation event on the detection target; determining whether a response event is received from the detection target in response to the stimulation event; when the response event is not received, determining whether the stimulation event satisfies a threshold strength condition; and recording the physiological symptom information as a corresponding drowsiness detection condition of the detection target when the stimulation event satisfies the threshold strength condition.

Claims

1. A drowsiness detection method, for performing a drowsiness detection operation on a detection target, the drowsiness detection method comprising a data collecting process, the data collecting process comprising:

a) detecting, by a detection/control circuit, first brainwave symptom information of the detection target;

b) determining, by the detection/control circuit, whether the detection target satisfies a predetermined drowsiness threshold condition according to the first brainwave symptom information, and performing Step (c) if a result is affirmative;

c) triggering, by a simulation triggering circuit connected to the detection/control circuit, a stimulation event on the detection target in response to a driving signal provided by the detection/control circuit when the detection target satisfies the predetermined drowsiness threshold condition;

d) within a predetermined period after step (c), determining, by a response circuit connected to the detection/control circuit, whether a response event, triggered by the detection target in response to the stimulation event, is detected, wherein in response to the response event the response circuit provides a response signal to the detection/control circuit and the response event is a man-machine-interface response event;

e) determining, by the detection/control circuit, whether the stimulation event satisfies a threshold strength condition when the response event is not detected;

f) recording, by the detection/control circuit, the first brainwave symptom information as a drowsiness detection condition corresponding to the detection target when the stimulation event satisfies the threshold strength condition; and

g) increasing, by the detection/control circuit, a strength of the stimulation event when the stimulation event does not satisfy the threshold strength condition, and iterating step (c);

wherein the data collecting process further comprises adjusting the threshold strength condition and iterating steps (a) to (g) to obtain a plurality of the drowsiness detection conditions corresponding to different threshold strength conditions of the detection target.

2. The drowsiness detection method according to claim 1, further comprising a detection condition training process performed by the detection/control circuit, the detection condition training process comprising:

a′

) establishing an artificial neural network; and

b′

) entering the drowsiness detection conditions of the detection target into the artificial neural network to identify and store a first adaptive mental state determining condition corresponding to the detection target.

3. The drowsiness detection method according to claim 2, further comprising a drowsiness monitoring process performed by the detection/control circuit, the drowsiness monitoring process comprising:

a″

) detecting second brainwave symptom information of the detection target;

b″

) determining whether the second brainwave symptom information satisfies the first adaptive mental state determining condition; and

c″

) determining the detection target as in a first mental state when the second brainwave symptom information satisfies the first adaptive mental state determining condition.

4. The drowsiness detection method according to claim 3, wherein the drowsiness monitoring process further comprises:

d″

) iterating step (a″

) when the second brainwave symptom information does not satisfy the first adaptive mental state determining condition.

5. The drowsiness detection method according to claim 3, wherein the detection condition training process further comprising the detection/control circuit identifies and stores a second adaptive mental state determining condition corresponding to the detection target based on the drowsiness detection conditions corresponding to the detection target entered into the artificial neural network.

6. The drowsiness detection method according to claim 5, wherein the drowsiness monitoring process further comprises:

d″

) determining whether the second brainwave symptom information satisfies the second adaptive mental state determining condition when the second brainwave symptom information does not satisfy the first adaptive mental state determining condition; and

e″

) determining the detection object as in a second mental state when the second brainwave symptom information satisfies the second adaptive mental state determining condition.

7. A drowsiness detection device, for performing a drowsiness detection operation on a detection target, comprising:

a detection/control circuit, for detecting first brainwave symptom information of the detection target, determining whether the detection target satisfies a predetermined drowsiness threshold condition according to the first brainwave symptom information, and further providing a driving signal when the detection target satisfies the predetermined drowsiness threshold condition;

a stimulation triggering circuit, for triggering a stimulation event on the detection target in response to the driving signal; and

a response circuit, for detecting a response event triggered by the detection target in response to the stimulation event, and accordingly providing a response signal to the detection/control circuit, wherein the response event is a man-machine-interface response event;

wherein, the detection/control circuit further detects whether the detection target correspondingly triggers the response event according to the response signal within a predetermined period from the simulation event, determines whether the stimulation event satisfies a threshold strength condition when the response event is not detected, and records the first brainwave symptom information as a drowsiness detection condition corresponding to the detection target when the stimulation event satisfies the threshold strength condition;

wherein the detection/control circuit further increases a strength of the stimulation event when the stimulation event does not satisfy the threshold strength condition, and triggers the stimulation event with the increased strength on the detection target to detect the response event corresponding to the stimulation event with the increased strength on the detection target; and

wherein the detection/control circuit further adjusts the threshold strength condition and drives the drowsiness detection device to obtain a plurality of the drowsiness detection conditions corresponding to different threshold strength conditions.

8. The drowsiness detection device according to claim 7, wherein the detection/control circuit further establishes an artificial neural network, and enters the drowsiness detection conditions of the detection target into the artificial neural network to identify and store a first adaptive mental state determining condition corresponding to the detection target.

9. The drowsiness detection device according to claim 8, wherein the detection/control circuit further detects a second brainwave symptom information of the detection target, determines whether the second brainwave symptom information satisfies the first adaptive mental state determining condition, and determines the detection target as in a first mental state when the second brainwave symptom information satisfies the first adaptive mental state determining condition.

10. The drowsiness detection device according to claim 9, wherein the detection/control circuit again detects the second brainwave symptom information when the second brainwave symptom information does not satisfy the first adaptive mental state determining condition.

11. The drowsiness detection device according to claim 9, wherein the detection/control circuit further identifies and stores a second adaptive mental state determining condition corresponding to the detection target based on the drowsiness detection conditions of the detection target entered into the artificial neural network.

12. The drowsiness detection device according to claim 11, wherein the detection/control circuit determines whether the second brainwave symptom information satisfies the second adaptive mental state determining condition when the second brainwave symptom information does not satisfy the first adaptive mental state determining condition, and determines the detection object as in a second mental state when the second brainwave symptom information satisfies the second adaptive mental state determining condition.