CN106667506A - Method and device for detecting lies on basis of electrodermal response and pupil change - Google Patents
Method and device for detecting lies on basis of electrodermal response and pupil change Download PDFInfo
- Publication number
- CN106667506A CN106667506A CN201611190789.7A CN201611190789A CN106667506A CN 106667506 A CN106667506 A CN 106667506A CN 201611190789 A CN201611190789 A CN 201611190789A CN 106667506 A CN106667506 A CN 106667506A
- Authority
- CN
- China
- Prior art keywords
- normal
- value
- eigenvalues
- pupil
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 210000001747 pupil Anatomy 0.000 title claims abstract description 87
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000004044 response Effects 0.000 title claims abstract description 22
- 238000012360 testing method Methods 0.000 claims abstract description 53
- 231100000430 skin reaction Toxicity 0.000 claims description 51
- 238000012549 training Methods 0.000 claims description 37
- 238000012795 verification Methods 0.000 claims description 30
- 230000037311 normal skin Effects 0.000 claims description 28
- 230000005611 electricity Effects 0.000 claims description 24
- 210000005252 bulbus oculi Anatomy 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 3
- 238000010181 skin prick test Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 9
- 230000006870 function Effects 0.000 description 34
- 238000005516 engineering process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000008921 facial expression Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000001815 facial effect Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 210000001508 eye Anatomy 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
Classifications
-
- 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
- A61B5/164—Lie detection
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/11—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for measuring interpupillary distance or diameter of pupils
- A61B3/112—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for measuring interpupillary distance or diameter of pupils for measuring diameter of pupils
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/053—Measuring electrical impedance or conductance of a portion of the body
- A61B5/0531—Measuring skin impedance
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/197—Matching; Classification
Abstract
The invention discloses a method and a device for detecting lies on the basis of electrodermal response and pupil change. The method includes determining normal feature values of test persons according to normal electrodermal response data of the test persons in non-lying states, normal pupil data and preset weight values; determining test feature values of the test persons according to test electrodermal response data of the test persons in test states, test pupil data and the weight values; comparing the normal feature values to the test feature values to obtain lie detection results of the test persons. The method and the device have the advantages that the method and the device are high in feasibility, and high lie detection accuracy can be guaranteed.
Description
Technical field
The present embodiments relate to lie-detection technology field, more particularly to a kind of survey changed based on galvanic skin response and pupil
Lie method and device.
Background technology
Through the development in more than halfth century, lie-detection technology investigation in the western countries headed by the U.S., safety and
It is widely used in Ji Yao departments.A lie detector is a kind of effective means that public security department is screened for crime fact, is passed through
The mode of asked questions stimulates testee, so as to obtain the corresponding physiological change of testee, by analyzing these
Physiological change is judging whether testee lies.Currently known a lie detector is mainly by detection galvanic skin response, chest
The physiological parameters such as breathing, pupil, blood pressure are being detected a lie.For example, when tested person lies, its psychology can feel nervous, so as to
Pupil is caused independently not amplify, and the salt branch of its skin surface uprises, and skin conductivity rate strengthens, therefore, according to these pupils
The physiological parameters such as hole situation of change, galvanic skin response just can determine whether out whether it lies.
It is the face's video by receiving user based on the scheme of detecting a lie of facial expression in prior art, detection face regards
The least one frame of facial image feature of frequency, being then based on facial image feature carries out human facial expression recognition, so as to using training
The probability of lying of the facial expression that good artificial neural networks are recognized.
Although the face recognition technology utilized in such scheme is awfully hot topic, and adaptable field is also very more,
But it is not high for the feasibility that judges to lie because everyone countenance difference can be than larger, somebody when lying
Can be poker-faced when lying, can smile when somebody lies, individual variation causes the classification of data set extremely difficult.If
One people can smile when lying, then by the photo of a smile, it is impossible to distinguish this people and only smile still
Lie.Thus, discriminate whether that the accuracy rate lied is relatively low, feasibility is not high using the scheme of human facial expression recognition.
The content of the invention
The present invention provides a kind of lie detecting method changed based on galvanic skin response and pupil and device, to realize work(of detecting a lie
Energy.
To reach this purpose, the embodiment of the present invention is employed the following technical solutions：
A kind of lie detecting method changed based on galvanic skin response and pupil, including：
Normal skin electricity response data, normal pupil data and default power according to tester under the state of not lying
Weight values, determine the normal eigenvalues of the tester；
Test galvanic skin response data, test pupil data and the power according to the tester under test mode
Weight values, determine the test feature value of the tester；
Relatively the normal eigenvalues and the test feature value, obtain the result of detecting a lie of the tester.
Further, in said method, the determination of the weighted value includes：
Normal skin electricity response data xi and normal pupil data according to N number of training sample user under the state of not lying
Yi, builds the decision function comprising weight parameter：
F (X, Y)=y*+α·(x*-y*)
Wherein α is weight parameter,
Normal skin electricity response data and normal pupil data according to M verification sample of users under the state of not lying,
Decision function is trained, the weighted value is obtained.
Further, the normal skin electricity reaction in said method, according to M verification sample of users under the state of not lying
Data and normal pupil data, are trained to decision function, obtain the weighted value, including：
For each value of default weight parameter, the value is respectively obtained according to the decision function corresponding normal
Eigenvalue is interval；
For each value, according to normal skin electricity response data of the M verification sample of users under the state of not lying with
Normal pupil data, and the discriminant function, respectively obtain the normal characteristics of the corresponding M verification sample of users of the value
Value；
For each value, the normal eigenvalues that the corresponding M verification sample of users of the value is determined respectively fall in correspondence
The interval probability of normal eigenvalues；
The corresponding value of maximum of probability is defined as into the weighted value.
Further, in said method, for each value of default weight parameter, distinguish according to the decision function
Obtain the corresponding normal eigenvalues of the value interval, including：
It is normal under the state of not lying according to N number of training sample user for each value of default weight parameter
Galvanic skin response data and normal pupil data, and the discriminant function respectively obtains the normal spy of N number of training sample user
Value indicative；
Maximum normal eigenvalues and minimum normal eigenvalues are selected from the normal eigenvalues of N number of training sample user
Build corresponding normal eigenvalues interval.
Further, in said method, the galvanic skin response data for galvanic skin response current value, the pupil number
According to the ratio that eyeball is occupied for pupil.
Correspondingly, invention additionally discloses a kind of device of detecting a lie changed based on galvanic skin response and pupil, including：
Normal eigenvalues determining module, for according to tester under the state of not lying normal skin electricity response data,
Normal pupil data and default weighted value, determine the normal eigenvalues of the tester；
Test feature value determining module, for the test galvanic skin response number according to the tester under test mode
According to, the test pupil data and weighted value, the test feature value of the tester is determined；
Result of detecting a lie determining module, for the relatively normal eigenvalues and the test feature value, obtains the test
The result of detecting a lie of person.
Further, in said apparatus, including weighted value determining module, the weighted value determining module includes：
Decision function construction unit, it is anti-for the normal skin electricity according to N number of training sample user under the state of not lying
Data xi and normal pupil data yi are answered, the decision function comprising weight parameter is built：
F (X, Y)=y*+α·(x*-y*)
Wherein α is weight parameter,
Decision function training unit, it is anti-for the normal skin electricity according to M verification sample of users under the state of not lying
Data and normal pupil data are answered, decision function is trained, obtain the weighted value.
Further, in said apparatus, the decision function training unit includes：
Characteristic interval subelement, for for each value of default weight parameter, distinguishing according to the decision function
Obtain the corresponding normal eigenvalues of the value interval；
Sample characteristics subelement, for for each value, sample of users is verified under the state of not lying just according to M
Normal galvanic skin response data and normal pupil data, and the discriminant function, respectively obtain the corresponding M verification sample of the value
The normal eigenvalues of this user；
Determine the probability subelement, for for each value, the corresponding M verification sample of users of the value being determined respectively
Normal eigenvalues fall in the interval probability of corresponding normal eigenvalues；
Weighted value subelement, for the corresponding value of maximum of probability to be defined as into the weighted value.
Further, in said apparatus, the characteristic interval subelement specifically for：
It is normal under the state of not lying according to N number of training sample user for each value of default weight parameter
Galvanic skin response data and normal pupil data, and the discriminant function respectively obtains the normal spy of N number of training sample user
Value indicative；
Maximum normal eigenvalues and minimum normal eigenvalues are selected from the normal eigenvalues of N number of training sample user
Build corresponding normal eigenvalues interval.
Further, in said apparatus, the galvanic skin response data for galvanic skin response current value, the pupil number
According to the ratio that eyeball is occupied for pupil.
The technical scheme that the embodiment of the present invention is provided, by gathering galvanic skin response data and pupil data, and is carried out
Corresponding data processing, realizes the higher lie detection function of accuracy rate, can have for the delinquent one kind that provides of public security department's strike
Effect approach, it is also possible to as an interesting application in life.
Description of the drawings
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, can be with basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is a kind of flow process of lie detecting method changed based on galvanic skin response and pupil that the embodiment of the present invention one is provided
Schematic diagram；
Fig. 2 is the idiographic flow schematic diagram of detect a lie stage-training part and part of detecting that the embodiment of the present invention one is provided；
Fig. 3 is the overall flow schematic diagram of detect a lie data acquisition and process that the embodiment of the present invention one is provided；
Fig. 4 is a kind of structure of lie detecting method changed based on galvanic skin response and pupil that the embodiment of the present invention two is provided
Schematic diagram.
Specific embodiment
With reference to the accompanying drawings and examples the present invention is described in further detail.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that, in order to just
Part related to the present invention rather than entire infrastructure are illustrate only in description, accompanying drawing.
Embodiment one
Accompanying drawing 1 is referred to, is a kind of based on detecting a lie that galvanic skin response and pupil change of the offer of the embodiment of the present invention one
The schematic flow sheet of method, the method is applied to and judges the scene whether testee lies by physiological reaction, the method by
Detect a lie device to perform based on what galvanic skin response and pupil changed, the device can be realized by software and/or hardware, be integrated in
The inside of operation terminal unit.The method specifically includes following steps：
S110, the normal skin electricity response data according to tester under the state of not lying, normal pupil data and pre-
If weighted value, determine the normal eigenvalues of the tester.
Specifically, put question to some normal and consentient problem to be trained tester, such issues that answer can only
For "Yes" or "No", these problems primarily to obtain galvanic skin response data of the tester in the case where not lying and
Pupil data, and the value of galvanic skin response data several times is made after normalized as the normal model of galvanic skin response data
Enclose, the data value of pupil size several times is made as the normal range of pupil size data after normalized, finally to two
The each in addition corresponding weight of kind of data, determines the scope of the normal eigenvalues of whole test.
S120, the test galvanic skin response data according to the tester under test mode, test pupil data and
The weighted value, determines the test feature value of the tester.
Specifically, by puing question to tester event to be confirmed related tender subject, tester is obtained in test mode
Under galvanic skin response data and pupil data, and respectively the value of galvanic skin response data and pupil data made into normalized
Afterwards each in addition corresponding weight, obtains test value.
Normal eigenvalues described in S130, comparison and the test feature value, obtain the result of detecting a lie of the tester.
Specifically, with reference to Fig. 2, on the basis of S110 and S120, it is known that the technical scheme that the present invention is provided can divide
To train and testing two parts.
Training part：The part mainly for obtain tester do not lie in the case of galvanic skin response data and pupil
Hole data, so as to obtain normal eigenvalues scope.
Part of detecting：Galvanic skin response data and pupil number of the part mainly for acquisition tester under test mode
According to so as to obtain test value.Further, by the way that test value is compared with the scope of normal eigenvalues, if test value
Beyond normal eigenvalues scope, then it is determined as lying；Otherwise, do not lie.
It should be noted that the galvanic skin response data are the current value of galvanic skin response, galvanic skin response can lead to
Cross tester to wear the wearable devices such as bracelet to gather galvanic skin response data；The pupil data occupies eyeball for pupil
Ratio, pupil situation of change can gather pupil data using the first-class mobile terminal device of cell-phone camera, and by two kinds
Data in addition weight finally carrying out judging whether to lie.
Explanation is needed further exist for, with reference to Fig. 3, tester is anti-by wearing the wearable devices such as bracelet collection skin pricktest
Data are answered, and data storage is got off to be sent to mobile phone terminal；Again by the picture data of mobile phone camera collecting test person, then
Eyes are oriented using algorithm of target detection, the ratio that pupil occupies eyeball is obtained, and this data are big as pupil
Little data are preserved, and eventually through mobile phone terminal data processing is carried out, and obtain result of detecting a lie.
Preferably, the determination of the weighted value includes：
Normal skin electricity response data x according to N number of training sample user under the state of not lyingiWith normal pupil data
yi, build the decision function comprising weight parameter：
F (X, Y)=y*+α·(x*-y*)
Wherein α is weight parameter,
Normal skin electricity response data and normal pupil data according to M verification sample of users under the state of not lying,
Decision function is trained, the weighted value is obtained.
It should be noted that galvanic skin response data xiMeansigma methodss beIts standard deviation isValue after normalization isCan obtain in the same manner, pupil data yiMeansigma methodss
ForIts standard deviation isValue after normalization is
The overall eigenvalue of order is F (X, Y), and the corresponding weight of galvanic skin response data is α, and α ∈ (0,1), pupil data
Corresponding weight is β, and β=1- α, then decision function can be obtained in the following manner：
Finally draw the decision function comprising weight parameter α：F (X, Y)=y*+α·(x*-y*)。
Preferably, it is described according to the electric response data of M normal skin of the verification sample of users under the state of not lying and just
Often pupil data, is trained to decision function, obtains the weighted value, including：
Part I, for each value of default weight parameter, according to the decision function value is respectively obtained
Corresponding normal eigenvalues are interval；
Part II, it is anti-according to normal skin electricity of the M verification sample of users under the state of not lying for each value
Data and normal pupil data, and the discriminant function are answered, the corresponding M verification sample of users of the value is being respectively obtained just
Normal eigenvalue；
Part III, for each value, determines respectively the normal eigenvalues of the corresponding M verification sample of users of the value
Fall in the interval probability of corresponding normal eigenvalues；The corresponding value of maximum of probability is defined as into the weighted value.
Preferably, in Part I, for each value of default weight parameter, obtain respectively according to the decision function
It is interval to the corresponding normal eigenvalues of the value, including：
It is normal under the state of not lying according to N number of training sample user for each value of default weight parameter
Galvanic skin response data and normal pupil data, and the discriminant function respectively obtains the normal spy of N number of training sample user
Value indicative；
Maximum normal eigenvalues and minimum normal eigenvalues are selected from the normal eigenvalues of N number of training sample user
Build corresponding normal eigenvalues interval.
It should be noted that F (X, Y) correspondence when can obtain the α of optimum and not lie by the method for machine learning
Scope, specifically, obtain F (X, Y) by way of continuous iteration α, and verify in F (X, Y) area using verification sample
Between probability, it is optimal value to take during maximum probability corresponding α.
In order to the optimal value and normal eigenvalues F (X, Y) that represent weight value α in the embodiment of the present invention that become apparent from it is right
The interval scheme implementation process answered, is described in detail below with an instantiation.The normal training sample that hypothesis is collected
Number is 10, and therein 60% is used for training, and is denoted as N, i.e. N=0.6*10=6, during training, if taking α=0.1, and decision function
F (X, Y)=y*+α·(x*-y*) several to there is 6 normal eigenvalues, maximum normal characteristics are selected in this 6 normal eigenvalues
It is interval that value and minimum normal eigenvalues build corresponding normal eigenvalues；Intersection is carried out by remaining 40% as verification sample to test
Card, is denoted as M, i.e. M=0.4*10=4, during verification, first this 4 verification samples is carried out with training identical to process, if being such as
α=0.1 is taken, 4 verification normal eigenvalues are obtained, and counts in the interval probability of corresponding normal eigenvalues.If taking α in the same manner
=0.2, another 4 verifications normal eigenvalues are obtained, and count in the interval probability of corresponding normal eigenvalues, until statistics
To the corresponding probability of all values of α, therefrom choose the corresponding α values of maximum of probability and be defined as the weighted value, and α ∈ (0,1), can
The concrete value of α is determined according to practical situation.
The technical scheme that the embodiment of the present invention is provided, by gathering galvanic skin response data and pupil data, and is carried out
Corresponding data processing, realizes the higher lie detection function of accuracy rate, can have for the delinquent one kind that provides of public security department's strike
Effect approach, it is also possible to as an interesting application in life.
Embodiment two
Accompanying drawing 4 is referred to, is a kind of based on detecting a lie that galvanic skin response and pupil change of the offer of the embodiment of the present invention two
The structural representation of device, the device is specifically comprising such as lower module：
Normal eigenvalues determining module 21, for the normal skin electricity stoichiometric number according to tester under the state of not lying
According to, normal pupil data and default weighted value, the normal eigenvalues of the tester are determined；
Test feature value determining module 22, for the test galvanic skin response number according to the tester under test mode
According to, the test pupil data and weighted value, the test feature value of the tester is determined；
Result of detecting a lie determining module 23, for the relatively normal eigenvalues and the test feature value, obtains the survey
The result of detecting a lie of examination person.
Preferably, also including weighted value determining module, the weighted value determining module includes：
Decision function construction unit, it is anti-for the normal skin electricity according to N number of training sample user under the state of not lying
Answer data xiWith normal pupil data yi, build the decision function comprising weight parameter：
F (X, Y)=y*+α·(x*-y*)
Wherein α is weight parameter,
Decision function training unit, it is anti-for the normal skin electricity according to M verification sample of users under the state of not lying
Data and normal pupil data are answered, decision function is trained, obtain the weighted value.
Preferably, the decision function training unit includes：Characteristic interval subelement, for for default weight parameter
Each value, respectively obtain the corresponding normal eigenvalues of the value according to the decision function interval；
Sample characteristics subelement, for for each value, sample of users is verified under the state of not lying just according to M
Normal galvanic skin response data and normal pupil data, and the discriminant function, respectively obtain the corresponding M verification sample of the value
The normal eigenvalues of this user；
Determine the probability subelement, for for each value, the corresponding M verification sample of users of the value being determined respectively
Normal eigenvalues fall in the interval probability of corresponding normal eigenvalues；
Weighted value subelement, for the corresponding value of maximum of probability to be defined as into the weighted value.
Preferably, the characteristic interval subelement specifically for：For each value of default weight parameter, according to N
Normal skin electricity response data and normal pupil data of the individual training sample user under the state of not lying, and the judgement letter
Number respectively obtains the normal eigenvalues of N number of training sample user；Select from the normal eigenvalues of N number of training sample user
It is interval that maximum normal eigenvalues and minimum normal eigenvalues build corresponding normal eigenvalues.
Needs say that the galvanic skin response data are the current value of galvanic skin response, and the pupil data is pupil
Occupy the ratio of eyeball.The galvanic skin response data can be gathered by Wearable devices such as bracelets, the pupil number
According to can be by mobile phone camera gather.
The technical scheme that the embodiment of the present invention is provided, by gathering galvanic skin response data and pupil data, and is carried out
Corresponding data processing, with higher feasibility, it is ensured that the accuracy rate detected a lie is higher.
The said goods can perform the method that any embodiment of the present invention is provided, and possess the corresponding functional module of execution method
And beneficial effect.
Note, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious changes,
Readjust and substitute without departing from protection scope of the present invention.Therefore, although the present invention is carried out by above example
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
More other Equivalent embodiments can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
1. a kind of lie detecting method changed based on galvanic skin response and pupil, it is characterised in that include：
Normal skin electricity response data, normal pupil data and default weight according to tester under the state of not lying
Value, determines the normal eigenvalues of the tester；
Test galvanic skin response data, test pupil data and the weight according to the tester under test mode
Value, determines the test feature value of the tester；
Relatively the normal eigenvalues and the test feature value, obtain the result of detecting a lie of the tester.
2. method according to claim 1, it is characterised in that the determination of the weighted value includes：
Normal skin electricity response data xi and normal pupil data yi according to N number of training sample user under the state of not lying,
Build the decision function comprising weight parameter：
F (X, Y)=y*+α·(x*-y*)
Wherein α is weight parameter,
Normal skin electricity response data and normal pupil data according to M verification sample of users under the state of not lying, to sentencing
Certainly function is trained, and obtains the weighted value.
3. method according to claim 2, it is characterised in that verify sample of users under the state of not lying according to M
Normal skin electricity response data and normal pupil data, are trained to decision function, obtain the weighted value, including：
For each value of default weight parameter, according to the decision function the corresponding normal characteristics of the value are respectively obtained
Value is interval；
For each value, normal skin electricity response data according to M verification sample of users under the state of not lying and normally
Pupil data, and the discriminant function, respectively obtain the normal eigenvalues of the corresponding M verification sample of users of the value；
For each value, determine respectively the corresponding M verification sample of users of the value normal eigenvalues fall it is corresponding just
The interval probability of normal eigenvalue；
The corresponding value of maximum of probability is defined as into the weighted value.
4. method according to claim 3, it is characterised in that for each value of default weight parameter, according to institute
State decision function and respectively obtain the corresponding normal eigenvalues interval of the value, including：
For each value of default weight parameter, according to normal skin of N number of training sample user under the state of not lying
Electric response data and normal pupil data, and the discriminant function respectively obtains the normal eigenvalues of N number of training sample user；
Maximum normal eigenvalues and minimum normal eigenvalues are selected to build from the normal eigenvalues of N number of training sample user
Corresponding normal eigenvalues are interval.
5. the method according to any one of claim 1-4, it is characterised in that the galvanic skin response data are that skin pricktest is anti-
The current value answered, the pupil data occupies the ratio of eyeball for pupil.
6. a kind of device of detecting a lie changed based on galvanic skin response and pupil, it is characterised in that include：
Normal eigenvalues determining module, for the normal skin electricity response data according to tester under the state of not lying, normally
Pupil data and default weighted value, determine the normal eigenvalues of the tester；
Test feature value determining module, for the test galvanic skin response data according to the tester under test mode, surveys
Examination pupil data and the weighted value, determine the test feature value of the tester；
Result of detecting a lie determining module, for the relatively normal eigenvalues and the test feature value, obtains the tester's
Detect a lie result.
7. device according to claim 6, it is characterised in that including weighted value determining module, the weighted value determines mould
Block includes：
Decision function construction unit, for the normal skin electricity stoichiometric number according to N number of training sample user under the state of not lying
According to xi and normal pupil data yi, the decision function comprising weight parameter is built：
F (X, Y)=y*+α·(x*-y*)
Wherein α is weight parameter,
Decision function training unit, for the normal skin electricity stoichiometric number according to M verification sample of users under the state of not lying
According to normal pupil data, decision function is trained, obtain the weighted value.
8. device according to claim 7, it is characterised in that the decision function training unit includes：
Characteristic interval subelement, for for each value of default weight parameter, respectively obtaining according to the decision function
The corresponding normal eigenvalues of the value are interval；
Sample characteristics subelement, for for each value, according to normal skin of the M verification sample of users under the state of not lying
Electric skin response data and normal pupil data, and the discriminant function, respectively obtain the corresponding M verification sample of the value and use
The normal eigenvalues at family；
Determine the probability subelement, for for each value, the normal of the corresponding M verification sample of users of the value being determined respectively
Eigenvalue falls in the interval probability of corresponding normal eigenvalues；
Weighted value subelement, for the corresponding value of maximum of probability to be defined as into the weighted value.
9. device according to claim 8, it is characterised in that the characteristic interval subelement specifically for：
For each value of default weight parameter, according to normal skin of N number of training sample user under the state of not lying
Electric response data and normal pupil data, and the discriminant function respectively obtains the normal eigenvalues of N number of training sample user；
Maximum normal eigenvalues and minimum normal eigenvalues are selected to build from the normal eigenvalues of N number of training sample user
Corresponding normal eigenvalues are interval.
10. the device according to any one of claim 6-9, it is characterised in that the galvanic skin response data are skin pricktest
The current value of reaction, the pupil data occupies the ratio of eyeball for pupil.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611190789.7A CN106667506B (en) | 2016-12-21 | 2016-12-21 | A kind of lie detecting method and device changed based on electrodermal response and pupil |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611190789.7A CN106667506B (en) | 2016-12-21 | 2016-12-21 | A kind of lie detecting method and device changed based on electrodermal response and pupil |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106667506A true CN106667506A (en) | 2017-05-17 |
| CN106667506B CN106667506B (en) | 2019-11-01 |
Family
ID=58869882
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611190789.7A Active CN106667506B (en) | 2016-12-21 | 2016-12-21 | A kind of lie detecting method and device changed based on electrodermal response and pupil |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106667506B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107453778A (en) * | 2017-06-22 | 2017-12-08 | 广州番禺职业技术学院 | A kind of intelligent watch and the method for authentication |
| CN110393504A (en) * | 2018-04-24 | 2019-11-01 | 高金铎 | A kind of method and device of pupil detection |
| CN110638472A (en) * | 2019-09-27 | 2020-01-03 | 新华网股份有限公司 | Emotion recognition method and device, electronic equipment and computer readable storage medium |
| CN111222374A (en) * | 2018-11-26 | 2020-06-02 | 广州慧睿思通信息科技有限公司 | Lie detection data processing method and device, computer equipment and storage medium |
| CN111714142A (en) * | 2020-06-12 | 2020-09-29 | 京东数字科技控股有限公司 | Psychological state evaluation method, device, equipment and computer readable storage medium |
| CN113729708A (en) * | 2021-09-10 | 2021-12-03 | 上海理工大学 | Lie evaluation method based on eye movement technology |
| US11607160B2 (en) | 2019-06-05 | 2023-03-21 | Carlos Andres Cuestas Rodriguez | System and method for multi modal deception test scored by machine learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6042548A (en) * | 1997-11-14 | 2000-03-28 | Hypervigilant Technologies | Virtual neurological monitor and method |
| CN202604845U (en) * | 2011-12-12 | 2012-12-19 | 张占强 | Pupillometric lie detector based on platform TMS320DM642 |
| CN103211605A (en) * | 2013-05-14 | 2013-07-24 | 重庆大学 | Psychological testing system and method |
| CN204336935U (en) * | 2014-12-24 | 2015-05-20 | 重庆大学 | A kind of multi-path physiology psychological tester based on pulse, breathing and skin resistance |
| CN105022929A (en) * | 2015-08-07 | 2015-11-04 | 北京环度智慧智能技术研究所有限公司 | Cognition accuracy analysis method for personality trait value test |
-
2016
- 2016-12-21 CN CN201611190789.7A patent/CN106667506B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6042548A (en) * | 1997-11-14 | 2000-03-28 | Hypervigilant Technologies | Virtual neurological monitor and method |
| CN202604845U (en) * | 2011-12-12 | 2012-12-19 | 张占强 | Pupillometric lie detector based on platform TMS320DM642 |
| CN103211605A (en) * | 2013-05-14 | 2013-07-24 | 重庆大学 | Psychological testing system and method |
| CN204336935U (en) * | 2014-12-24 | 2015-05-20 | 重庆大学 | A kind of multi-path physiology psychological tester based on pulse, breathing and skin resistance |
| CN105022929A (en) * | 2015-08-07 | 2015-11-04 | 北京环度智慧智能技术研究所有限公司 | Cognition accuracy analysis method for personality trait value test |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107453778A (en) * | 2017-06-22 | 2017-12-08 | 广州番禺职业技术学院 | A kind of intelligent watch and the method for authentication |
| CN110393504A (en) * | 2018-04-24 | 2019-11-01 | 高金铎 | A kind of method and device of pupil detection |
| CN110393504B (en) * | 2018-04-24 | 2022-02-15 | 高金铎 | Intelligent virus identification method and device |
| CN111222374A (en) * | 2018-11-26 | 2020-06-02 | 广州慧睿思通信息科技有限公司 | Lie detection data processing method and device, computer equipment and storage medium |
| US11607160B2 (en) | 2019-06-05 | 2023-03-21 | Carlos Andres Cuestas Rodriguez | System and method for multi modal deception test scored by machine learning |
| CN110638472A (en) * | 2019-09-27 | 2020-01-03 | 新华网股份有限公司 | Emotion recognition method and device, electronic equipment and computer readable storage medium |
| CN110638472B (en) * | 2019-09-27 | 2022-07-05 | 新华网股份有限公司 | Emotion recognition method and device, electronic equipment and computer readable storage medium |
| CN111714142A (en) * | 2020-06-12 | 2020-09-29 | 京东数字科技控股有限公司 | Psychological state evaluation method, device, equipment and computer readable storage medium |
| CN111714142B (en) * | 2020-06-12 | 2023-12-08 | 京东科技控股股份有限公司 | Psychological state assessment method, psychological state assessment device, psychological state assessment equipment and computer readable storage medium |
| CN113729708A (en) * | 2021-09-10 | 2021-12-03 | 上海理工大学 | Lie evaluation method based on eye movement technology |
| CN113729708B (en) * | 2021-09-10 | 2023-06-20 | 上海理工大学 | Lie judgment method based on eye movement technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106667506B (en) | 2019-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106667506A (en) | Method and device for detecting lies on basis of electrodermal response and pupil change | |
| JP4401079B2 (en) | Subject behavior analysis | |
| Bartlett et al. | Automatic decoding of facial movements reveals deceptive pain expressions | |
| CN105559802B (en) | Based on the depressed diagnostic system and data processing method paid attention to and emotion information merges | |
| CN109543526B (en) | True and false facial paralysis recognition system based on depth difference characteristics | |
| CN104274191B (en) | A kind of Psychological Evaluation method and system thereof | |
| CN107536617A (en) | The apparatus and method of bio-identification signal detection driver status based on driver | |
| CN107007257A (en) | The automatic measure grading method and apparatus of the unnatural degree of face | |
| CN109805944B (en) | Children's ability analytic system that shares feelings | |
| Matta et al. | Real-time continuous identification system using ECG signals | |
| CN106128032A (en) | A kind of fatigue state monitoring and method for early warning and system thereof | |
| KR20120131043A (en) | Biometric system using ecg signal and fingerprint identification | |
| CN110598608A (en) | Non-contact and contact cooperative psychological and physiological state intelligent monitoring system | |
| CN106529377A (en) | Age estimating method, age estimating device and age estimating system based on image | |
| CN109567832A (en) | A kind of method and system of the angry driving condition of detection based on Intelligent bracelet | |
| CN111062021B (en) | Method and device for identity authentication based on wearable equipment | |
| CN108717872A (en) | Health analysis method and system based on face, hard recognition and big data | |
| CN106901689A (en) | Cognitive and motor function detecting system and data processing method based on interaction in kind | |
| CN110464369A (en) | A kind of state of mind based on human body physical sign numerical value judges algorithm | |
| CN110175522A (en) | Work attendance method, system and Related product | |
| CN111222374A (en) | Lie detection data processing method and device, computer equipment and storage medium | |
| Rawat et al. | Real-Time Heartbeat Sensing with Face Video using a Webcam and OpenCV | |
| Perdana et al. | Lie Detector with Eye Movement and Eye Blinks Analysis Based on Image Processing using Viola-Jones Algorithm | |
| Alariki et al. | A Review Study of Heartbeat Biometric Authentication. | |
| Fernández et al. | Automatic eye gesture recognition in audiometries for patients with cognitive decline |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 200233 Shanghai, Jinshan District Ting Wei Road, building 1419, room 4, No. 6558 Patentee after: Shanghai Hede Intelligent Equipment Co.,Ltd. Address before: 200233 Shanghai, Jinshan District Ting Wei Road, building 1419, room 4, No. 6558 Patentee before: SHANGHAI YUDE INFORMATION TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200515 Address after: 830002 Room 102, unit 3, building 9, family home, Bayi Middle School, Wuxing Road, Tianshan District, Urumqi City, Xinjiang Uygur Autonomous Region Patentee after: Wang Zhenhuan Address before: 200233 Shanghai, Jinshan District Ting Wei Road, building 1419, room 4, No. 6558 Patentee before: Shanghai Hede Intelligent Equipment Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| CP02 | Change in the address of a patent holder |
Address after: Unit 201, unit 2, building 3, Anping community, Jiyang street, Zhuji City, Shaoxing City, Zhejiang Province Patentee after: Wang Zhenhuan Address before: 830002 Room 102, unit 3, building 9, family home, Bayi Middle School, Wuxing Road, Tianshan District, Urumqi City, Xinjiang Uygur Autonomous Region Patentee before: Wang Zhenhuan |
|
| CP02 | Change in the address of a patent holder | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210113 Address after: 210000 212, 2nd floor, building 9, xingzhihui business garden, No.19 Xinghuo Road, Jiangbei new district, Nanjing City, Jiangsu Province Patentee after: Nanjing Lizhi psychological big data Industry Research Institute Co.,Ltd. Address before: 311800 unit 2, building 3, Anping community, Jiyang street, Zhuji City, Shaoxing City, Zhejiang Province Patentee before: Wang Zhenhuan |
|
| TR01 | Transfer of patent right |