STATE DETECTING DEVICE, STATE DETECTING METHOD, AND NON-TRANSITORY COMPUTER-READABLE MEDIUM

• US 20120105234A1
• Filed: 03/23/2010
• Published: 05/03/2012
• Est. Priority Date: 07/09/2009
• Status: Active Grant

• Current Assignee: Aisin Seiki Co., Ltd. (Aisin Corp.)
• Original Assignee: Aisin Seiki Co., Ltd. (Aisin Corp.)
• Inventors: Oguri, Koji, Kishimoto, Yoshifumi
• Granted Patent: US 8,786,448 B2
• Time in Patent Office: Days
• US Class Current: 340/576
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• 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 5/7264 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: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/24317 Piecewise classification, i.e. whereby each classification requires several discriminant rules
• 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: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Disclosed is a state detecting device that detects characteristic quantities regarding the driver that are represented by biological information such as the eyelid opening degree and frequency of heartbeat of the driver, identifies the group to which the state of the driver belongs between a Group 1 and a Group 2 into which multiple classes defined using the activity level of the driver as an indicator are sorted based on the detected characteristic quantities, outputs information including the identification results as output values, and detects the class to which the current state of the driver belongs based on loss values calculated from the output values. From the detected class, to what degree the activity level of the driver is impaired can be detected along with the level.

Claims

1. A state detecting device, comprising:

a characteristic quantity detector for detecting information regarding a monitored subject as characteristic quantities;

an identification device for receiving said characteristic quantities as input values, identifying a group to which a state of said monitored subject belongs between a first group and a second group into which multiple classes defined using the activity level of said monitored subject as an indicator are sorted based on said characteristic quantities, and outputting information including the identification results as output values;

a loss value calculator for calculating loss values from the output values output from said identification device; and

a class detector for detecting said class to which the state of said monitored subject belongs based on said loss values.

2. The state detecting device according to claim 1, wherein:

said multiple classes are sorted into said first group and second group in multiple patterns.

3. The state detecting device according to claim 1 or 2, wherein:

said loss value calculator defines said loss values using an exponential function.

4. The state detecting device according to claim 1, wherein:

provided that the number of said classes is p and the number of said identification device is G,

each of said identification device outputs a sign indicating that the activity level of said monitored subject belongs to the first group consisting of the classes corresponding to the elements of a value “

1”

in each row of a code table W presented by W ε

{1, 0, −

1}<sup>p·

G</sup> or to the second group consisting of the classes corresponding to the elements of a value “

−

1” and

an absolute value indicating the credibility.

5. The state detecting device according to claim 4, wherein:

said loss value calculator obtains the loss defined by an exponential function for each of said identification device and each of said classes based on said sign and credibility output from said identification device and said code table W, and further obtains the sum of losses corresponding to the same class.

6. The state detecting device according to claim 5, wherein:

said class detector detects the class of which said sum of losses is the smallest as the class to which the state of said monitored subject belongs.

7. The state detecting device according to claim 1, wherein:

the information regarding said monitored subject is biological information of said monitored subject.

8. The state detecting device according to claim 7, wherein:

the biological information of said monitored subject includes at least one of information on the sight line of said monitored subject and information on the heartbeat of said monitored subject.

9. The state detecting device according to claim 1, wherein:

the characteristic quantity detector supplies the detected characteristic quantities to said identification device.

10. A state detecting method, comprising:

an identification process to receive characteristic quantities that are indicators of a state of a monitored subject as input values, identify a group to which the state of said monitored subject belongs between a first group and a second group into which multiple classes defined using the activity level of said monitored subject as an indicator are sorted based on said characteristic quantities, and output information including the identification results as output values;

a loss value calculation process to calculate loss values from the output values output in said identification process; and

a class detection process to detect said class to which the state of said monitored subject belongs based on said loss values.

11. A non-transitory computer-readable medium stored a program that allows a computer to execute the following steps:

an identification step of receiving characteristic quantities that are indicators of a state of a monitored subject as input values, identifying a group to which the state of said monitored subject belongs between a first group and a second group into which multiple classes defined using the activity level of said monitored subject as an indicator are sorted based on said characteristic quantities, and outputting information including the identification results as output values;

a loss value calculation step of calculating loss values from the output values output in said identification step; and

a class detection step of detecting said class to which the state of said monitored subject belongs based on said loss values.