WO2011004641A1 - 状態検出装置、状態検出方法及びプログラム - Google Patents
状態検出装置、状態検出方法及びプログラム Download PDFInfo
- Publication number
- WO2011004641A1 WO2011004641A1 PCT/JP2010/055010 JP2010055010W WO2011004641A1 WO 2011004641 A1 WO2011004641 A1 WO 2011004641A1 JP 2010055010 W JP2010055010 W JP 2010055010W WO 2011004641 A1 WO2011004641 A1 WO 2011004641A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- class
- group
- state
- monitored person
- value
- Prior art date
Links
- 238000000034 method Methods 0.000 title description 9
- 230000000694 effects Effects 0.000 claims abstract description 21
- 238000001514 detection method Methods 0.000 claims description 62
- 238000004364 calculation method Methods 0.000 claims description 7
- 210000000744 eyelid Anatomy 0.000 abstract description 8
- 238000002474 experimental method Methods 0.000 description 8
- 206010041349 Somnolence Diseases 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 230000035484 reaction time Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000001121 heart beat frequency Effects 0.000 description 3
- 208000032140 Sleepiness Diseases 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000037007 arousal Effects 0.000 description 2
- 230000037321 sleepiness Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/06—Alarms for ensuring the safety of persons indicating a condition of sleep, e.g. anti-dozing alarms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/0245—Detecting, measuring or recording pulse rate or heart rate by using sensing means generating electric signals, i.e. ECG signals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/18—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24317—Piecewise classification, i.e. whereby each classification requires several discriminant rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/597—Recognising the driver's state or behaviour, e.g. attention or drowsiness
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02405—Determining heart rate variability
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1103—Detecting eye twinkling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Definitions
- the present invention relates to a state detection device, a state detection method, and a program.
- Patent Literature 1 Patent Literature 1, Patent Literature 2, and Non-Patent Literature 1 disclose state detection devices that evaluate driver activity.
- the state detection device disclosed in Patent Document 1 calculates a standard deviation of the amount of movement of a vehicle that moves in a direction orthogonal to the traveling direction.
- the state detection device detects the driver's arousal level based on the result of comparing the calculated minimum and maximum values of the standard deviation with a threshold value.
- the state detection device disclosed in Patent Document 2 continuously detects the driver's line-of-sight direction for a predetermined period. This state detection device generates line-of-sight distribution information indicating the distribution in the detected line-of-sight direction. The state detection device determines whether or not the driver is driving the vehicle casually from the directionality of the line of sight indicated by the generated line-of-sight distribution information.
- Non-Patent Document 1 detects a heart rate that varies depending on a driver's mental load as a feature amount. This state detection device detects a driver's driving load by performing feature value matching using AdaBoost.
- the driver's driving tendency and the behavior pattern according to the driver's arousal level differ depending on the driver.
- the state detection devices disclosed in Patent Documents 1 and 2 uniformly detect the state of the driver. For this reason, the state detection devices disclosed in Patent Documents 1 and 2 cannot accurately detect the individual state of the driver.
- the state detection device disclosed in Non-Patent Document 1 detects the state of each driver based on the result of learning the driving tendency of each driver.
- the state detection device disclosed in Non-Patent Document 1 only categorizes the driver's state into two types of states, normal operation and non-normal operation. For this reason, the state detection device disclosed in Non-Patent Document 1 cannot detect the degree of burden on the driver during non-normal driving.
- the present invention has been made under the above-described circumstances, and an object thereof is to detect the state of each person being monitored according to the degree.
- the state detection apparatus of the present invention includes a feature amount detection unit that detects information about a monitored person as a feature amount, the feature amount as an input value, and the target amount is based on the feature amount.
- a feature amount detection unit that detects information about a monitored person as a feature amount, the feature amount as an input value, and the target amount is based on the feature amount.
- the group to which the state of the monitored person belongs is identified, and information including the identified result is included.
- the plurality of classes may be classified into the first group and the second group in a plurality of patterns.
- the loss value calculation means may define the loss value by an exponential function.
- each identification means indicates that the activity level of the monitored person is represented by W ⁇ ⁇ 1, 0, ⁇ 1 ⁇ p ⁇ G.
- W the number of classes
- each identification means indicates that the activity level of the monitored person is represented by W ⁇ ⁇ 1, 0, ⁇ 1 ⁇ p ⁇ G.
- the loss value calculation means calculates the loss defined by an exponential function for each identification means and each class based on the code, the reliability, and the code table W output from each identification means. Further, the sum of losses corresponding to the same class may be obtained.
- the class detection means may detect the class with the smallest sum of losses as the class to which the state of the monitored person belongs.
- the information related to the monitored person may be biological information of the monitored person.
- the biological information of the monitored person may include the line-of-sight information of the monitored person.
- the biological information of the monitored person may include heartbeat information of the monitored person.
- the state detection method of the present invention uses, as an input value, a feature amount that is an index indicating the state of the monitored person, and uses the activity level of the monitored person as an index based on the feature amount.
- the program of the present invention uses a feature amount, which is an index indicating the state of the monitored person, as an input value to the computer, and indicates the activity level of the monitored person based on the feature amount.
- the state of each person being monitored can be detected according to the degree.
- FIG. 1 is a block diagram of a state detection device according to a first embodiment. It is a block diagram of an identification unit. It is a figure for demonstrating operation
- the state detection device 10 acquires biological information of a driver who drives a car.
- the state detection device 10 detects the driving state of each driver together with the degree based on the acquired biological information.
- the state detection device 10 acquires the driver's biometric information, detects the feature amount of the driver, and detects the feature amount, and drives based on the feature amount detected by the biometric information acquisition device 20. And an identification device 30 for detecting the driving state of the person together with the degree of the driving state.
- the biometric information acquisition device 20 acquires, as the driver's biometric information, for example, information related to the opening of the heel resulting from sleepiness and information related to the heartbeat frequency.
- the biological information acquisition apparatus 20 acquires information on the eyelid opening by analyzing the eyelid image.
- the biological information acquisition device 20 detects, as the feature amount a1, a value obtained by normalizing the driver's eyelid opening in the range of 0 to 1 using the acquired information on the eyelid opening.
- the biological information acquisition device 20 acquires information on the heartbeat frequency by analyzing the RR interval.
- the biometric information acquisition device 20 detects the frequency in the low frequency region of the driver's heartbeat interval variation as the feature amount a2 from the power spectrum of the heartbeat variation using the acquired information on the heartbeat frequency, and in the high frequency region. Is detected as a feature quantity a3.
- the biological information acquisition device 20 outputs information x (a1, a2, a3) regarding the detected feature amount to the identification device 30.
- information x (a1, a2, a3) relating to the feature amount is hereinafter simply referred to as input information x.
- the identification device 30 detects a class to which the driving state of the driver belongs, for example, among three classes (class 1, class 2, and class 3) classified using the driver's activity as an index.
- the identification device 30 includes an identification unit 40 and a class detection unit 50.
- the driver's activity can be defined by the length of time (reaction time) until the driver takes action when the driver takes action in response to an external input.
- the driver's reaction time tends to become longer due to drowsiness and reduced attention.
- the driver's reaction time is measured in various situations.
- the driver's state when the driver's reaction time is short is the normal state, and this is class 1.
- the state of the driver when the driver's reaction time is long is class 3.
- a state between two classes (class 1 and class 3) is defined as class 2.
- the driver's state is divided into three classes.
- the identification unit 40 includes an encoder 41, six identifiers 42 1 to 42 6, and a decoder 43.
- the encoder 41 performs an encoding process using a code table W represented by the following equation (1).
- p is the number of classifiers
- G is the number of classes.
- Each row of the code table W includes both “1” and “ ⁇ 1”.
- the number of classes to which the operating state belongs is three.
- the number of discriminators is six. Therefore, the code table W shown by the general formula of the following formula (1) is composed of 18 elements arranged in a matrix of 6 rows and 3 columns, as shown by the following formula (2).
- the three elements in each row of the code table W correspond to class 1, class 2, and class 3, respectively, from left to right.
- the three classes are classified into a first group consisting of classes corresponding to elements having a value of “1” and a second group consisting of classes corresponding to elements having a value of “ ⁇ 1”.
- the number of classes in each group may be one or plural (two).
- the class corresponding to the element whose value is 0 is excluded from the classification (not classified into the first group or the second group).
- the three elements in the first row of the code table W mean that class 1 is classified into the first group, and class 2 and class 3 are classified into the second group.
- the three elements in the second row of the code table W mean that class 2 is classified into the first group, and class 1 and class 3 are classified into the second group. Further, the three elements in the third row of the code table W mean that class 3 is classified into the first group, and class 1 and class 2 are classified into the second group.
- the three elements in the fourth row of the code table W mean that class 1 is classified into the first group and class 2 is classified into the second group.
- the three elements in the fifth row of the code table W mean that class 1 is classified into the first group and class 3 is classified into the second group.
- the three elements in the sixth row of the code table W mean that class 2 is classified into the first group and class 3 is classified into the second group.
- the encoder 41 associates the code 1 [1, ⁇ 1, ⁇ 1] including the three elements in the first row of the code table W with the input information x. I do.
- the encoder 41 outputs the input information x that code 1 is associated to the identifier 42 1.
- the encoder 41 includes code 2 [-1, 1, -1], code 3 [-1, -1, 1 and 3 elements each corresponding to the second to sixth lines of the code table W. ], Code 4 [1, -1, 0], code 5 [1, 0, -1], code 6 [0, 1, -1] and the input information x are respectively associated.
- the encoder 41 outputs the corresponding code 2 to the input information x code 6 is associated, to the discriminators 42 2 to 42 6 corresponding to each code 2-6, respectively.
- the discriminators 42 1 to 42 6 are binary discriminators that have been learned by AdaBoost and have the same configuration.
- the discriminator 42 i classifies the plurality of classes into the first group and the second group based on the code i supplied thereto.
- the discriminator 42 i is configured such that the driver state (active state) represented by the input information x (a1, a2, a3) is composed of a class to which the element “1” of the code i is assigned. Identify (discriminate) whether it belongs to the second group consisting of the class to which the element “ ⁇ 1” of the code i is assigned or not, and the reliability indicating the accuracy of the identification result and the identification result , Output to the decoder 43.
- a code 1 [1, ⁇ 1, ⁇ 1] is supplied to the discriminator 42 1 .
- the classifier 421 1 has the “driver state” represented by the input information x (a1, a2, a3) in the first group consisting of the class 1 (normal state) to which the element “1” is assigned. It is identified whether it is classified into a second group composed of class 2 (mandled state) and class 3 (mandested state) to which element “ ⁇ 1” is assigned. Thereafter, the classifier 42 1 outputs an output value h 1 (x) indicating the identification result and the reliability (likelihood).
- the output value h 1 (x) is composed of a numerical value with a sign ( ⁇ ).
- the sign of the output value h 1 (x) corresponds to the sign of the element constituting the code.
- the identifier 42 1 indicates that the driving state of the driver represented by the input information x is in the first group (class 1). Indicates that it has been identified as belonging.
- the sign of the output value h 1 (x) is ⁇ ( ⁇ 0)
- the identifier 42 1 indicates that the driving state of the driver represented by the input information x is the second group (class 2). Or, it indicates that it has been identified as belonging to class 3).
- the absolute value of the output value h 1 (x) indicates the reliability of the identification result identified by the classifier 42 1 .
- the identifier 42 1 the state of the driver input information x is representative belongs and identifies the second group (Class 2 or 3), its reliability is
- each of the discriminators 42 2 to 42 6 is composed of a class to which the element “1” is assigned to the “driver state” represented by the input information x in accordance with the corresponding codes 2 to 6, respectively. Whether it is classified into the first group or the second group composed of the class to which the element “ ⁇ 1” is assigned is identified.
- Each discriminator 42 2 to 42 6 outputs each output value h 2 (x) to h 6 (x) corresponding to the discrimination result.
- the identifier 42 2 the state of the driver input information x is representative, identified as belonging to the second group (Class 1 or 3), its reliability is
- the discriminator 42 3 the state of the driver, identified as belonging to the first group (class 3), the reliability is
- Discriminator 42 4 the state of the driver, belongs and identifies the second group (class 2), its reliability is
- Identification 42 5 the state of the driver, identified as belonging to the second group (class 3), the reliability
- Discriminator 42 6 the state of the driver, identified as belonging to the first group (class 2), its reliability is
- the decoder 43 performs a decoding process using the code table W represented by the above equation (2).
- each of the six elements in the first column of the code table W represents whether the class 1 is classified into the first group or the second group.
- Each of the six elements in the second column of the code table W represents whether the class 2 is classified into the first group or the second group.
- Each of the six elements in the third column of the code table W represents whether the class 3 is classified into the first group or the second group.
- the output values of the discriminators 42 1 to 42 6 represent the Euclidean distance from the discrimination plane and are equivalent to the reliability of the discrimination result. Therefore, when the first group is defined by the element having the value “1” and the second group is defined by the element having the value “ ⁇ 1” as described above, the discriminator 42 1 ⁇ 42 6 output value h n of (x) is, the more the code is larger its value is positive, the operation state of the driver indicating that more likely belonging to the first group. Further, the output values h n discriminator 42 1 ⁇ 42 6 (x) is, the more its sign is small its value is negative (enough absolute value is large), tend to the operating state of the driver belonging to the second group Represents being strong.
- the decoder 43 uses the output values of the discriminators 42 1 to 42 6 and the elements in the column direction of the code table W to calculate the loss values L 1 to L 3 corresponding to the classes 1 to 3, respectively. calculate.
- a function represented by the following equation (3) is used for calculating the loss values L 1 to L 3 .
- the loss value L 1 corresponding to class 1 is calculated as in the following equation (4).
- the decoder 43 calculates a loss value L 2 and a loss value L 3 corresponding to class 2 and class 3, respectively.
- the loss value L 2 is ⁇ 161667
- the loss value L 3 is 153546.
- the decoder 43 outputs the calculated loss values L 1 to L 3 to the class detection unit 50.
- the class detection unit 50 detects the class corresponding to the smallest loss value among the loss values L 1 to L 3 output from the decoder 43.
- the class detection unit 50 outputs the detected class to the outside as a class to which the driving state of the driver belongs. For example, as described above, when the loss value L 1 output from the decoder 43 is 7018, the loss value L 2 is ⁇ 161667, and the loss value L 3 is 153546, the class detection unit 50 determines the minimum loss value L 2. Class 2 corresponding to is output to the outside as a class to which the driving state of the driver belongs.
- the state detection device 10 for example, based on the feature amount related to the driver represented by biological information such as the opening degree of the driver's eyelid, the frequency of the heartbeat, and the like.
- the class to which the current driving state of the driver belongs is detected.
- This class is one of the driving states of the driver divided into a plurality of classes using the driver's activity as an index. Therefore, based on the detected class, it is possible to detect the estimation result as to how much the driver's activity level is reduced regardless of the individual difference or the individual difference. Thereby, it becomes possible to provide driving assistance more appropriate for the driver.
- the biological information of the driver is not limited to the above-described opening degree of the eyelid and the frequency of the heartbeat.
- a plurality of pieces of biological information such as blood pressure and body temperature are further acquired, and the driving state is also taken into account.
- the classifiers 42 1 to 42 6 are described as binary classifiers that have already been learned by Adaboost.
- the present invention is not limited to this, and the discriminators 42 1 to 42 6 may be binary discriminators such as SVM (Support Vector Machine).
- the ECOC Error Correcting Output Coding
- class 1 is a normal driving state (a state where normal driving is possible)
- class 2 is a non-normal driving state (sleeping, casual driving, fatigue driving, rush driving, stress
- Class 3 can also be in a non-normal driving state and in a conversation with another person.
- the number p of classes is not limited to 3 and is arbitrary.
- the driving state of the driver may be divided into four or more classes using the driver's activity as an index.
- the code table W is as shown in FIG.
- the driver's driving state may be completely sleepless, slightly sleepy, sleepy, quite sleepy,
- the activity level of the driver may be classified as an index into five classes corresponding to sleepy states.
- the size W SIZE (number of rows) of the code table W can be obtained by the following equation.
- the number of columns is equal to the class number G.
- the code table W is fixed.
- the encoder may be removed, and each discriminator may be configured to perform a binary determination that is assigned in advance.
- the identification device 30 is configured as follows.
- the encoder 41 is removed, and the input information x is directly input to the discriminators 42 1 to 42 6 .
- the discriminator 421 1 determines whether the “driver state” represented by the input information x is classified into the first group consisting of class 1 or the second group consisting of class 2 and class 3. identified, the identifier 42 2, "driver's state", or being classified into the first group of class 2, identifies which from class 1 and class 3 is classified into the second group including ,. . .
- Discriminator 42 6 "driver's state", or being classified into the first group of class 2, identifies which are classified from the class 3 to the second group composed, configured and so May be.
- the inventors conducted an experiment using the state detection device of the present invention and a driving simulator in order to verify the effectiveness of the present invention.
- the subjects are 3 men and women (21-23 years old), and the duration of one experiment is 20 minutes.
- 10 experiments were conducted per subject.
- a monotonous highway including a random S-curve was set for the driving simulator.
- the state detection device of the present invention was able to detect the driving state of the driver with a correct answer rate of 93.56%. This accuracy rate is about 1.5 to 3 times higher than other methods.
- the identification device 30 is realized by a general computer.
- the state detection device 10 includes a biological information acquisition device 20 and an identification device 30 realized by a computer.
- the identification device 30 connects a central processing unit (CPU) 30a, a main storage unit 30b, an auxiliary storage unit 30c, a display unit 30d, an input unit 30e, an interface unit 30f, and the units 30a to 30f.
- System bus 30g a central processing unit (CPU) 30a, a main storage unit 30b, an auxiliary storage unit 30c, a display unit 30d, an input unit 30e, an interface unit 30f, and the units 30a to 30f.
- CPU central processing unit
- CPU30a performs the process mentioned later to the input information x acquired by the biometric information acquisition apparatus 20 according to the program memorize
- the main storage unit 30b includes a RAM (Random Access Memory) and the like, and is used as a work area of the CPU 30a.
- RAM Random Access Memory
- the auxiliary storage unit 30c includes a non-volatile memory such as a ROM (Read Only Memory), a magnetic disk, and a semiconductor memory.
- the auxiliary storage unit 30c stores programs executed by the CPU 30a, various parameters, and the like. Moreover, the information including the input information x output from the biometric information acquisition apparatus 20 and the processing result by the CPU 30a is sequentially stored.
- the display unit 30d includes a CRT (Cathode Ray Tube) or LCD (Liquid Crystal Display), and displays the processing result of the CPU 30a.
- CRT Cathode Ray Tube
- LCD Liquid Crystal Display
- the input unit 30e includes a pointing device such as a keyboard and a mouse.
- the operator's instruction is input via the input unit 30e and notified to the CPU 30a via the system bus 30g.
- the interface unit 30f includes a serial interface or a LAN (Local Area Network) interface.
- the biological information acquisition device 20 is connected to the system bus 30g via the interface unit 30f.
- This process is executed after the CPU 30a instructs the biometric information acquisition apparatus 20 to acquire biometric information and the input information x is input from the biometric information acquisition apparatus 20.
- the CPU 30a has a code [1, -1, -1], code 2 [-1, 1, -1] each including three elements in the first to sixth lines of the code table W. , Code 3 [-1, -1, 1], code 4 [1, -1, 0], code 5 [1, 0, -1], code 6 [0, 1, -1] and input information x Associate with.
- the CPU 30a classifies classes 1 to 3 into classes belonging to the first group and classes belonging to the second group based on the codes 1 to 6. Then, based on the input information x, the CPU 30a identifies whether the driving state of the driver belongs to the first group or the second group, and the identification values h 1 (x) to h according to the identification result 6 (x) is calculated.
- the CPU 30a uses the identification values h 1 (x) to h 6 (x) and the elements in the column direction of the code table W, and the loss values L corresponding to the classes 1 to 3, respectively. to calculate the 1 ⁇ L 3.
- the identification values h 1 (x) to h 6 (x) are ⁇ 2, ⁇ 7, 0.5, ⁇ 1, ⁇ 9, and 12, respectively, the corresponding loss value L 1 is calculated as the above formula (4).
- the CPU 30a detects a class corresponding to the loss value having the smallest value among the calculated loss values L 1 to L 3 .
- the CPU 30a assigns the class 2 corresponding to the minimum loss value L 2 to which the driving state of the driver belongs. Detect as a class.
- step S105 the CPU 30a outputs information on the detected class to an external device or the like. Thereafter, the CPU 30a repeatedly executes the processing from step S101 to step S105.
- the state detection device 10 for example, based on the feature amount related to the driver represented by biological information such as the opening degree of the driver's eyelid, the frequency of the heartbeat, and the like.
- the class to which the current driving state of the driver belongs is detected.
- This class is one of the driving states of the driver divided into a plurality of classes using the driver's activity as an index. Therefore, based on the detected class, it is possible to detect the fact of how much the driver's activity level is reduced, regardless of the individual difference or the individual difference. Thereby, it becomes possible to provide driving assistance more appropriate for the driver.
- biometric information is used as information about the driver, but the present invention is not limited to this.
- Information related to the driver includes, for example, information related to the vehicle such as the steering angle, the degree of wobbling of the vehicle, the timing of the brake, the speed and acceleration of the vehicle, the ambient environment such as the temperature and humidity in the passenger compartment, audio, Information regarding operation of equipment other than driving, such as operation of an air conditioner, may be acquired.
- the state detection apparatus of the present invention may detect the class to which the driving state belongs in consideration of these information.
- the state detection device 10 detects the state of the driver who drives the automobile
- the present invention is not limited to this.
- the state detection device of the present invention is also suitable for detecting the state of a driver or operator who drives a train, an airplane, or the like.
- the driver's activity may be determined based on an index that varies depending on, for example, the sleepiness, fatigue level, etc. of the monitored person.
- the driver's activity may be defined based on, for example, the accuracy of the driver's (monitored person's) operation, attention, or the like.
- the loss value may be calculated using, for example, a sigmoid function.
- the function of the identification device 30 can be realized by dedicated hardware or by a normal computer system.
- the programs stored in the auxiliary storage unit 30c of the identification device 30 are a flexible disk, a CD-ROM (Compact Disk Read-Only Memory), a DVD (Digital Versatile Disk), an MO (Magneto- An apparatus that performs the above-described processing may be configured by storing and distributing in a computer-readable recording medium such as an optical disk) and installing the program in the computer.
- a computer-readable recording medium such as an optical disk
- the program may be stored in a disk device or the like of a predetermined server device on a communication network such as the Internet. For example, it may be superimposed on a carrier wave and downloaded to a computer.
- the program may be activated and executed while being transferred via a communication network.
- the program may be executed entirely or partially on the server device, and the above-described processing may be executed while transmitting / receiving information regarding the processing via the communication network.
- the state detection device, state detection method, and program of the present invention are suitable for detecting the state of the monitored person.
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Theoretical Computer Science (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Cardiology (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Child & Adolescent Psychology (AREA)
- Social Psychology (AREA)
- Mathematical Physics (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Developmental Disabilities (AREA)
- Educational Technology (AREA)
- Hospice & Palliative Care (AREA)
- Psychology (AREA)
- Fuzzy Systems (AREA)
- Multimedia (AREA)
Abstract
Description
以下、本発明の第1の実施形態に係る状態検出装置10を、図1~図4を参照して説明する。本実施形態に係る状態検出装置10は、自動車を運転する運転者の生体情報を取得する。状態検出装置10は、取得した生体情報に基づいて運転者個々の運転状態を、その度合いとともに検出する。
次に、本発明の第2の実施形態を、図6及び図7を参照して説明する。
なお、第1の実施形態に係る状態検出装置10と同一又は同等の構成については、同一の符号を用い、その説明を省略する。
20 生体情報取得装置
30 識別装置
30a CPU
30b 主記憶部
30c 補助記憶部
30d 表示部
30e 入力部
30f インターフェイス部
30g システムバス
40 識別ユニット
41 符号器
421~426 識別器
43 復号器
1 クラス検出ユニット
Claims (11)
- 被監視者に関する情報を特徴量として検出する特徴量検出手段と、
前記特徴量を入力値とし、当該特徴量をもとに、前記被監視者の活性度合を指標として定められた複数のクラスが分類される第1グループと第2グループのうち、前記被監視者の状態が属する前記グループを識別し、当該識別した結果を含む情報を出力値として出力する識別手段と、
前記識別手段が出力した出力値から損失値を算出する損失値算出手段と、
前記損失値に基づいて前記被監視者の状態が属する前記クラスを検出するクラス検出手段と、
を備える状態検出装置。 - 複数の前記クラスは、複数通りのパターンで前記第1グループと前記第2グループとに分類される請求項1に記載の状態検出装置。
- 前記損失値算出手段は、前記損失値を、指数関数で定義する請求項1または2に記載の状態検出装置。
- 前記クラス数をp、前記識別手段の数をGとしたときに、
前記各識別手段は、前記被監視者の活性度合が、W∈{1,0,-1}p・Gで表される符号表Wの対応する行の値が1の要素に対応するクラスから構成される第1グループと、値が-1の要素に対応するクラスから構成される第2グループのいずれに属すかを表す符号とその信頼度を表す絶対値とを出力する、請求項1に記載の状態検出装置。 - 前記損失値算出手段は、
前記各識別手段から出力された前記符号及び前記信頼度と前記符号表Wとに基づいて、
指数関数で定義される損失を、前記識別手段別及び前記クラス別に求め、
さらに、同一クラスに対応する損失の和を求める、
請求項4に記載の状態検出装置。 - 前記クラス検出手段は、前記損失の和が最も小さいクラスを、前記被監視者の状態が属する前記クラスとして検出する、
請求項5に記載の状態検出装置。 - 前記被監視者に関する情報は、前記被監視者の生体情報である請求項1乃至3のいずれか一項に記載の状態検出装置。
- 前記被監視者の生体情報は、前記被監視者の視線情報と、前記被監視者の心拍情報の少なくとも一方を含む請求項7に記載の状態検出装置。
- 前記被監視者に関する情報を特徴量として検出し、前記識別手段に供給する特徴量検出手段をさらに備える、
請求項1に記載の状態検出装置。 - 被監視者の状態を示す指標である特徴量を入力値とし、当該特徴量をもとに、前記被監視者の活性度合を指標として定められた複数のクラスが分類される第1グループと第2グループのうち、前記被監視者の状態が属する前記グループを識別し、当該識別した結果を含む情報を出力値として出力する識別工程と、
前記識別工程で出力した出力値から損失値を算出する損失値算出工程と、
前記損失値に基づいて前記被監視者の状態が属する前記クラスを検出するクラス検出工程と、
を含む状態検出方法。 - コンピュータに、
被監視者の状態を示す指標である特徴量を入力値とし、当該特徴量をもとに、前記被監視者の活性度合を指標として定められた複数のクラスが分類される第1グループと第2グループのうち、前記被監視者の状態が属する前記グループを識別し、当該識別した結果を含む情報を出力値として出力する識別ステップと、
前記識別ステップで出力した前記出力値から損失値を算出する損失値算出ステップと、
前記損失値に基づいて前記被監視者の状態が属する前記クラスを検出するクラス検出ステップと、
を実行させるためのプログラム。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201080031025.8A CN102473354B (zh) | 2009-07-09 | 2010-03-23 | 状态检测装置、状态检测方法 |
EP10796952.9A EP2453427A4 (en) | 2009-07-09 | 2010-03-23 | STATUS DETECTION DEVICE, STATUS RECOGNITION PROCESS AND PROGRAM THEREFOR |
US13/383,085 US8786448B2 (en) | 2009-07-09 | 2010-03-23 | State detecting device, state detecting method, and non-transitory computer-readable medium |
KR1020127003098A KR101311552B1 (ko) | 2009-07-09 | 2010-03-23 | 상태검출장치, 상태검출방법 및 기록매체 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2009-163143 | 2009-07-09 | ||
JP2009163143A JP5444898B2 (ja) | 2009-07-09 | 2009-07-09 | 状態検出装置、状態検出方法及びプログラム |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011004641A1 true WO2011004641A1 (ja) | 2011-01-13 |
Family
ID=43429066
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/055010 WO2011004641A1 (ja) | 2009-07-09 | 2010-03-23 | 状態検出装置、状態検出方法及びプログラム |
Country Status (6)
Country | Link |
---|---|
US (1) | US8786448B2 (ja) |
EP (1) | EP2453427A4 (ja) |
JP (1) | JP5444898B2 (ja) |
KR (1) | KR101311552B1 (ja) |
CN (1) | CN102473354B (ja) |
WO (1) | WO2011004641A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103403756A (zh) * | 2011-03-03 | 2013-11-20 | 爱信精机株式会社 | 状态推断装置、状态推断方法及程序 |
CN105361863A (zh) * | 2015-12-17 | 2016-03-02 | 无锡桑尼安科技有限公司 | 飞机机长生理参数监控*** |
CN105564659A (zh) * | 2016-01-02 | 2016-05-11 | 无锡桑尼安科技有限公司 | 用于飞机上的弹簧式紧急报警装置 |
Families Citing this family (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2925736B1 (fr) * | 2007-12-21 | 2010-09-24 | Eads Secure Networks | Procede de production d'une preuve de presence ou de fonctionnement d'une entite dans une zone identifiee pendant une duree superieure a un seuil donne, et systeme de surveillance |
JP2012166579A (ja) * | 2011-02-09 | 2012-09-06 | Aisin Seiki Co Ltd | 状態判定装置、状態判定方法及びプログラム |
JP5720462B2 (ja) * | 2011-07-20 | 2015-05-20 | 日産自動車株式会社 | 運転者疲労度推定装置 |
CN103077603A (zh) * | 2012-06-06 | 2013-05-01 | 王晓原 | 基于动态人车环境协同推演的自由流状态汽车驾驶倾向性辨识*** |
KR20140092634A (ko) * | 2013-01-16 | 2014-07-24 | 삼성전자주식회사 | 전자장치와 그 제어방법 |
WO2014156055A1 (ja) * | 2013-03-28 | 2014-10-02 | パナソニック株式会社 | 提示情報学習方法、サーバおよび端末装置 |
CN103426004B (zh) * | 2013-07-04 | 2016-12-28 | 西安理工大学 | 基于纠错输出编码的车型识别方法 |
DE102014201650A1 (de) * | 2013-12-19 | 2015-06-25 | Robert Bosch Gmbh | Verfahren zum Ermitteln des Belastungszustands des Fahrers |
CN103770733B (zh) * | 2014-01-15 | 2017-01-11 | 中国人民解放军国防科学技术大学 | 一种驾驶员安全驾驶状态检测方法及装置 |
US20150294547A1 (en) * | 2014-04-15 | 2015-10-15 | Denso Corporation | Driver condition detection apparatus, driving assistance apparatus, driver condition detection method, and driving assistance method |
KR101659027B1 (ko) | 2014-05-15 | 2016-09-23 | 엘지전자 주식회사 | 이동 단말기 및 차량 제어 장치 |
CN105313898B (zh) | 2014-07-23 | 2018-03-20 | 现代摩比斯株式会社 | 驾驶员状态感应装置及其方法 |
KR102401607B1 (ko) * | 2014-12-04 | 2022-05-25 | 현대모비스 주식회사 | 운전자의 운행 집중도 레벨 분석방법 |
KR101646401B1 (ko) * | 2014-12-08 | 2016-08-12 | 현대자동차주식회사 | 운전자의 졸음상태 판단 시스템 및 그 판단 방법 |
CN104464194A (zh) * | 2014-12-15 | 2015-03-25 | 宇龙计算机通信科技(深圳)有限公司 | 报警提示方法、报警提示***、终端和可穿戴设备 |
CN105788364B (zh) * | 2014-12-25 | 2019-05-14 | ***通信集团公司 | 一种预警信息的发布方法及装置 |
JP6399311B2 (ja) * | 2015-05-07 | 2018-10-03 | スズキ株式会社 | 居眠り検知装置 |
DE102015224739A1 (de) * | 2015-12-09 | 2017-06-14 | Robert Bosch Gmbh | Verfahren und Vorrichtung zum Auslösen von Fußgängerschutzmitteln und/oder Fußgängerwarnmitteln |
DE102015225109A1 (de) | 2015-12-14 | 2017-06-14 | Robert Bosch Gmbh | Verfahren und Vorrichtung zum Klassieren von Augenöffnungsdaten zumindest eines Auges eines Insassen eines Fahrzeugs und Verfahren und Vorrichtung zum Erfassen einer Schläfrigkeit und/oder eines Sekundenschlafes eines Insassen eines Fahrzeugs |
US10357195B2 (en) * | 2017-08-01 | 2019-07-23 | Panasonic Intellectual Property Management Co., Ltd. | Pupillometry and sensor fusion for monitoring and predicting a vehicle operator's condition |
CN107730835B (zh) * | 2017-11-14 | 2019-07-02 | 吉林大学 | 一种基于应激反应能力的汽车驾驶员疲劳识别方法 |
CN109620269B (zh) * | 2019-01-28 | 2021-10-22 | 锦图计算技术(深圳)有限公司 | 疲劳检测方法、装置、设备及可读存储介质 |
KR102171236B1 (ko) * | 2019-03-14 | 2020-10-28 | 서울시립대학교 산학협력단 | 심장 박동의 타입을 분류하는 방법 및 이를 이용한 장치 |
JP7358958B2 (ja) * | 2019-12-11 | 2023-10-11 | トヨタ自動車株式会社 | 運転意識推定装置 |
US20220261657A1 (en) * | 2021-02-17 | 2022-08-18 | International Business Machines Corporation | Monte-carlo adversarial autoencoder for multi-source domain adaptation |
WO2023188130A1 (ja) * | 2022-03-30 | 2023-10-05 | 株式会社Subaru | 診断装置 |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005158077A (ja) * | 2003-11-26 | 2005-06-16 | Daimler Chrysler Ag | 車両の運転者による不注意を識別する方法及びコンピュータプログラム |
JP2007265377A (ja) * | 2006-03-01 | 2007-10-11 | Toyota Central Res & Dev Lab Inc | 運転者状態判定装置及び運転支援装置 |
JP2008217274A (ja) * | 2007-03-01 | 2008-09-18 | Equos Research Co Ltd | 運転者状態判定装置、及び運転支援装置 |
JP2008243031A (ja) | 2007-03-28 | 2008-10-09 | Toyota Central R&D Labs Inc | 漫然運転判定装置 |
JP2008250859A (ja) | 2007-03-30 | 2008-10-16 | Toyota Central R&D Labs Inc | 覚醒度判定装置 |
JP2008282022A (ja) * | 2000-10-14 | 2008-11-20 | Temic Automotive Of North America Inc | 車両運転手の能力を向上させるための方法及び装置 |
JP2009090028A (ja) * | 2007-10-11 | 2009-04-30 | Denso Corp | 眠気度判定装置 |
JP2009163143A (ja) | 2008-01-09 | 2009-07-23 | Dainippon Printing Co Ltd | 高色濃度カラーフィルタ、およびその修正方法 |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5416861A (en) * | 1994-04-29 | 1995-05-16 | University Of Cincinnati | Optical synchronous clock distribution network and high-speed signal distribution network |
US6580973B2 (en) | 2000-10-14 | 2003-06-17 | Robert H. Leivian | Method of response synthesis in a driver assistance system |
US6909947B2 (en) | 2000-10-14 | 2005-06-21 | Motorola, Inc. | System and method for driver performance improvement |
US6925425B2 (en) | 2000-10-14 | 2005-08-02 | Motorola, Inc. | Method and apparatus for vehicle operator performance assessment and improvement |
US20020151297A1 (en) | 2000-10-14 | 2002-10-17 | Donald Remboski | Context aware wireless communication device and method |
DE602004031659D1 (de) * | 2003-10-23 | 2011-04-14 | Delta Tooling Co Ltd | Htung zum nachweis von müdigkeit und computerprogramm |
JP4466571B2 (ja) | 2005-05-12 | 2010-05-26 | 株式会社デンソー | ドライバ状態検出装置、車載警報装置、運転支援システム |
JP4702100B2 (ja) * | 2006-02-27 | 2011-06-15 | トヨタ自動車株式会社 | 居眠り判定装置および居眠り運転警告装置 |
CN100453043C (zh) | 2007-01-23 | 2009-01-21 | 武汉理工大学 | 汽车驾驶员疲劳实时监测*** |
CN100462047C (zh) * | 2007-03-21 | 2009-02-18 | 汤一平 | 基于全方位计算机视觉的安全驾驶辅助装置 |
WO2009003170A1 (en) * | 2007-06-27 | 2008-12-31 | Radow Scott B | Stationary exercise equipment |
US8744738B2 (en) * | 2007-09-28 | 2014-06-03 | The Boeing Company | Aircraft traffic separation system |
JP4948379B2 (ja) * | 2007-12-18 | 2012-06-06 | キヤノン株式会社 | パターン識別器生成方法、情報処理装置、プログラム及び記憶媒体 |
CN101281646A (zh) * | 2008-05-09 | 2008-10-08 | 山东大学 | 基于视觉的驾驶员疲劳实时检测方法 |
CN101375796B (zh) * | 2008-09-18 | 2010-06-02 | 浙江工业大学 | 疲劳驾驶实时检测*** |
JP5270415B2 (ja) | 2009-03-19 | 2013-08-21 | トヨタ自動車株式会社 | 眠気判定装置及びプログラム |
-
2009
- 2009-07-09 JP JP2009163143A patent/JP5444898B2/ja active Active
-
2010
- 2010-03-23 KR KR1020127003098A patent/KR101311552B1/ko active IP Right Grant
- 2010-03-23 US US13/383,085 patent/US8786448B2/en active Active
- 2010-03-23 WO PCT/JP2010/055010 patent/WO2011004641A1/ja active Application Filing
- 2010-03-23 EP EP10796952.9A patent/EP2453427A4/en not_active Withdrawn
- 2010-03-23 CN CN201080031025.8A patent/CN102473354B/zh not_active Expired - Fee Related
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008282022A (ja) * | 2000-10-14 | 2008-11-20 | Temic Automotive Of North America Inc | 車両運転手の能力を向上させるための方法及び装置 |
JP2005158077A (ja) * | 2003-11-26 | 2005-06-16 | Daimler Chrysler Ag | 車両の運転者による不注意を識別する方法及びコンピュータプログラム |
JP2007265377A (ja) * | 2006-03-01 | 2007-10-11 | Toyota Central Res & Dev Lab Inc | 運転者状態判定装置及び運転支援装置 |
JP2008217274A (ja) * | 2007-03-01 | 2008-09-18 | Equos Research Co Ltd | 運転者状態判定装置、及び運転支援装置 |
JP2008243031A (ja) | 2007-03-28 | 2008-10-09 | Toyota Central R&D Labs Inc | 漫然運転判定装置 |
JP2008250859A (ja) | 2007-03-30 | 2008-10-16 | Toyota Central R&D Labs Inc | 覚醒度判定装置 |
JP2009090028A (ja) * | 2007-10-11 | 2009-04-30 | Denso Corp | 眠気度判定装置 |
JP2009163143A (ja) | 2008-01-09 | 2009-07-23 | Dainippon Printing Co Ltd | 高色濃度カラーフィルタ、およびその修正方法 |
Non-Patent Citations (2)
Title |
---|
INFORMATION PROCESSING SOCIETY OF JAPAN JOURNAL, vol. 50, 2009, pages 171 - 180 |
See also references of EP2453427A4 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103403756A (zh) * | 2011-03-03 | 2013-11-20 | 爱信精机株式会社 | 状态推断装置、状态推断方法及程序 |
EP2682917A4 (en) * | 2011-03-03 | 2014-08-20 | Aisin Seiki | STATUS ESTIMATING APPARATUS, STATUS ESTIMATION METHOD AND PROGRAM THEREFOR |
US9064397B2 (en) | 2011-03-03 | 2015-06-23 | Aisin Seiki Kabushiki Kaisha | State estimation device, state estimation method, and program |
CN105361863A (zh) * | 2015-12-17 | 2016-03-02 | 无锡桑尼安科技有限公司 | 飞机机长生理参数监控*** |
CN105564659A (zh) * | 2016-01-02 | 2016-05-11 | 无锡桑尼安科技有限公司 | 用于飞机上的弹簧式紧急报警装置 |
Also Published As
Publication number | Publication date |
---|---|
US8786448B2 (en) | 2014-07-22 |
CN102473354B (zh) | 2014-07-16 |
JP2011018240A (ja) | 2011-01-27 |
KR20120046230A (ko) | 2012-05-09 |
KR101311552B1 (ko) | 2013-09-26 |
EP2453427A1 (en) | 2012-05-16 |
US20120105234A1 (en) | 2012-05-03 |
CN102473354A (zh) | 2012-05-23 |
EP2453427A4 (en) | 2015-01-21 |
JP5444898B2 (ja) | 2014-03-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2011004641A1 (ja) | 状態検出装置、状態検出方法及びプログラム | |
Jiang et al. | Learning visual attention to identify people with autism spectrum disorder | |
US11823386B2 (en) | Diagnosis assistance apparatus, and information processing method | |
US10332638B2 (en) | Methods and systems for pre-symptomatic detection of exposure to an agent | |
US20190370982A1 (en) | Movement learning device, skill discriminating device, and skill discriminating system | |
CN111898642A (zh) | 关键点检测方法、装置、电子设备及存储介质 | |
EP3872818A2 (en) | Method for providing diagnostic system using semi-supervised learning, and diagnostic system using same | |
CN112216402A (zh) | 基于人工智能的疫情预测方法、装置、计算机设备及介质 | |
JP2016066114A (ja) | 特徴点検出装置、方法及びプログラム | |
Tran et al. | Phase segmentation methods for an automatic surgical workflow analysis | |
KR101852774B1 (ko) | 처방전 인식 오류 보정 방법 및 처방전 인식 오류 보정 시스템 | |
JP5917932B2 (ja) | 状態推定装置、状態推定方法及びプログラム | |
JP2012166579A (ja) | 状態判定装置、状態判定方法及びプログラム | |
WO2012118206A1 (ja) | 状態推定装置、状態推定方法及びプログラム | |
JP2019080809A (ja) | 状態推定装置 | |
CN107995181B (zh) | 一种基于步态的身份验证方法、装置、设备及存储介质 | |
CN114266777A (zh) | 分割模型的训练方法、分割方法、装置、电子设备及介质 | |
Nilashi et al. | Accuracy Analysis of Type-2 Fuzzy System in Predicting Parkinson’s Disease Using Biomedical Voice Measures | |
WO2021024297A1 (ja) | 敵対的事例検知システム、方法およびプログラム | |
de Barros et al. | Computer aided diagnosis of neurodevelopmental disorders and genetic syndromes based on facial images–A systematic literature review | |
Xie et al. | An SVM parameter learning algorithm scalable on large data size for driver fatigue detection | |
JP2021185465A (ja) | 音響認識システム、音響認識方法 | |
EP3940626A1 (en) | Information processing method and information processing system | |
CN112580724A (zh) | 基于扭矩轮的匀速转动识别的润滑剂的润滑性能检测方法 | |
CN111797784B (zh) | 驾驶行为监测方法、装置、电子设备及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201080031025.8 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10796952 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13383085 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010796952 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 20127003098 Country of ref document: KR Kind code of ref document: A |