CN109910818A - A kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification - Google Patents
A kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification Download PDFInfo
- Publication number
- CN109910818A CN109910818A CN201910116114.5A CN201910116114A CN109910818A CN 109910818 A CN109910818 A CN 109910818A CN 201910116114 A CN201910116114 A CN 201910116114A CN 109910818 A CN109910818 A CN 109910818A
- Authority
- CN
- China
- Prior art keywords
- feature
- identification
- vocal print
- vehicle
- matching
- 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
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The present invention provides a kind of VATS Vehicle Anti-Theft Systems based on human body multiple features fusion identification, comprising: image collecting device for acquiring the facial image of user, and is transferred in system on chip;Voice collection device for acquiring the voice signal of user, and is transferred in system on chip;System on chip, for receiving the facial image and voice signal and being handled, identification matching is carried out simultaneously to the changes in faces feature and vocal print feature of user, then feature and vocal print feature fusion is moved using mouth lip to carry out authentication, whether is legitimate user to determine.The present invention realizes changes in faces feature and vocal print feature first while carrying out identification matching, then mouth lip is recycled to move feature and vocal print feature fusion to carry out authentication, it whether can be out legitimate user with quick discrimination, it not only increases and sets rate, the accuracy of car owner's identification can also be increased, to reduce the generation that vehicle is stolen event.
Description
Technical field
The present invention relates to a kind of VATS Vehicle Anti-Theft Systems based on human body multiple features fusion identification, belong to vehicle anti-theft skill
Art field.
Background technique
As the rapid development of sociaty and economy, people's lives level also increasingly increases, most people is gathered around
There is private car, automobile has become the essential vehicles in human lives.
However, increasing with automobile, car clout case is also frequently to occur.Although vehicle is stolen in order to prevent now,
There is also a series of anti-theft measures, but not can be well solved vehicle theft problem.There are still illegal point in life
The phenomenon that son carries out deception certification using photo and false recording.
Therefore, the accuracy of VATS Vehicle Anti-Theft System car owner identification how is improved, thus reduce the generation that vehicle is stolen event,
It is that those skilled in the art are dedicated to the problem solved.
Summary of the invention
The technical problem to be solved by the present invention is the accuracy of VATS Vehicle Anti-Theft System car owner identification how is improved, to subtract
Small vehicle is stolen the generation of event.
In order to solve the above-mentioned technical problem, the technical solution of the present invention is to provide one kind to be based on human body multiple features fusion identity
The VATS Vehicle Anti-Theft System of identification, it is characterised in that: including
Image collecting device acquires the facial image of user after receiving enabling signal, and is transferred to SOC piece
In upper system;
Voice collection device acquires the voice signal of user after receiving enabling signal, and is transferred to SOC piece
In upper system;
System on chip becomes the face of user for receiving the facial image and voice signal and being handled
Change feature and vocal print feature carries out identification matching simultaneously, then moves feature and vocal print feature fusion using mouth lip to carry out body
Whether part verifying, be legitimate user to determine.
Preferably, the system on chip includes:
Microprocessor is filled for receiving identification signal code after signal to described image acquisition device and sound collection
Transmission enabling signal is set, and starts identification IP kernel, the opening and closing of control vehicle are interacted with communication module;
DRAM dynamic random access memory, for accessing collected face picture and voice signal, and with the body
Part identification IP kernel is connected;
EEPROM Electrically Erasable Programmable Read-Only Memory, for storing the eyes and nose of the legitimate user of preparatory typing
Textural characteristics, mouth lip move feature and vocal print feature is respective and mouth lip moves and the feature templates of vocal print Fusion Features, and
It is connected with the microprocessor and identification IP kernel;
The identification IP kernel changes identification module, voiceprint identification module, characteristic matching module group by facial characteristics
At, and be connected with microprocessor;
The facial characteristics changes identification module, for by the facial image of collected user carry out noise reduction process and
Feature extraction calculates the textural characteristics of each area dividing using the concurrency of hardware simultaneously and mouth lip moves feature;
The voiceprint identification module, for collected voice signal to be carried out noise reduction process and vocal print feature extraction;
The characteristic matching module, it is preparatory with EEPROM for merging the dynamic feature of mouth lip and vocal print feature
The correlated characteristic template of storage is matched;
The communication module, the communication for vehicle and mobile terminal.
Preferably, the SOC, which is inserted in the slot of vehicle-mounted ECU, is connected thereto, and electric wave signal is issued out of key, by automobile
Antenna receives the vehicle-mounted ECU identification signal code through control engine after the electric wave signal, then SOC system is driven to start work
Make, the microprocessor opens image collecting device and voice collection device, and it is special to start the face in identification IP kernel
Sign variation identification module, voiceprint identification module, characteristic matching module, verification result feeds back to control by microprocessor after verifying
The vehicle-mounted ECU of engine is controlled the opening and closing of vehicle by vehicle-mounted ECU or alarmed.
Preferably, the characteristic matching module is realized match while facial image and vocal print feature first, in matching
Identification is just exited if there is any one characteristic matching is unsuccessful, the verifying of information is carried out using mobile terminal.
Preferably, the authorization of vehicle is carried out by the mobile terminal, and carries out checking for vehicle-state in time.
It is highly preferred that the mobile terminal is mobile phone, IPAD or computer.
Preferably, the system on chip method for identifying ID specifically includes the following steps:
Step A1: obtain the face texture feature of the legitimate user of preparatory typing, mouth lip move feature and vocal print feature with
And mouth lip moves the correlate template of feature and vocal print Fusion Features, and stores in memory;
Step A2: after the facial image and voice signal that collect user, to the facial image harmony of the acquisition
Sound signal is pre-processed, and face texture feature and vocal print feature are obtained;
Step A3: at the same to information in collected face texture feature and vocal print feature and the memory carry out
Identification matching;
Step A4: if face texture feature and vocal print feature have a kind of matching unsuccessful, feeding back to microprocessor for result,
Microprocessor transmits information to mobile terminal by communication module, determined by car owner according to facial image and acoustic information be
No unlocking vehicle;
Step A5: if successful match, lip dynamic, vocal print feature being merged, then matched again, if
With unsuccessful, result is fed back into microprocessor, microprocessor transmits information to mobile terminal by communication module, by car owner
Decide whether unlocking vehicle according to facial image and acoustic information;If successful match, microprocessor feeds back to result vehicle-mounted
ECU unlocking vehicle.
It is highly preferred that carrying out identifying matched detailed process for facial image are as follows: protect as far as possible first in the step A3
It is motionless to hold face, says password password;Acquired image is divided into multiple homalographic region units, chooses two differences respectively
Position operator calculate the gray level co-occurrence matrixes of each area dividing, then the characteristic values of gray level co-occurrence matrixes is calculated as retouching
State the feature of image texture;Calculate the textural characteristics of each area dividing simultaneously using the concurrency of hardware;According to texture spy
Levy the position for determining eyes and nose, the textural characteristics of nose are to obtain the texture of the first width effective image, the line of eyes
Reason is obtained from multiple image;The inverted triangle geometry having by eyes, nose and mouth determines the position of mouth, and
It is partitioned into mouth region of interest ROI;The ROI image for comparing several frames in front and back analyzes the variation of mouth region, extracts mouth lip
Dynamic feature;Feature templates, which are moved, with the eyes of pre-stored user and the textural characteristics of nose and mouth lip carries out multiple features fusion
Matching;In matching, a threshold value T is preset, only when the feature and pre-stored legal use that identified person extracts
When the feature templates information matches at family are more than the threshold value T, illustrate the success of this characteristic matching.
It is highly preferred that carrying out identifying matched detailed process for voice signal in the step A3 are as follows: obtain each short
When analysis window frequency spectrum, Mel frequency spectrum is then obtained by Mel filter group, cepstral analysis is carried out on Mel frequency spectrum, obtain
MFCC carries out the feature extraction of vocal print as characteristic parameter, if owner name is " LL ", then user is carrying out authentication
When should be verified with the speech form of " hello, I requests the vehicle using LL ", discriminate whether as legitimate user.
It is highly preferred that the step A5 is specifically included: the MFCC feature of the behavioral characteristics of lip, vocal print being carried out feature and is melted
It is combined into:In formula,For the weight of the behavioral characteristics of lip,For vocal print
The weight of MFCC feature, diIndicate the similitude of ith feature, i is positive integer;α=(a1,a2,……ak) it is the lip dynamic spy extracted
Levy vector, α '=(a '1,a′2,……a′k) it is pre-stored lip behavioral characteristics vector, dimension k;β=(b1,b2,……
bj) it is the vocal print feature vector extracted, β '=(b '1,b′2,……b′j) it is pre-stored vocal print feature vector, dimension j;For the attenuation function of weight, wherein a is setting coefficient, TiFor weights omegaiDecision threshold
Value,For the inverse of noise coefficient, η ' is fused feature;It is and preparatory after fusion feature extracts
The fusion feature template of storage is matched, and equally in matching, presets a threshold value T, only mentioning as identified person
When the matching of the feature of taking-up and the feature templates information of pre-stored legitimate user is more than the threshold value T, illustrate this feature
Successful match.
Method provided by the invention overcomes the deficiencies in the prior art, designs vehicle anti-theft system using system on chip
System, first realization changes in faces feature and vocal print feature simultaneously carry out identification matching, then recycle mouth lip move feature and
Vocal print feature merges to carry out authentication.The method is a variety of using the parallel identification method of image and vocal print and fusion human body
Whether feature carries out authentication and applies on VATS Vehicle Anti-Theft System, can be out legitimate user with quick discrimination, not only increase
Recognition rate, additionally it is possible to increase the accuracy of car owner's identification, to reduce the generation that vehicle is stolen event.
Detailed description of the invention
Fig. 1 is the structure of vehicle anti-theft system figure provided in this embodiment based on human body multiple features fusion identification;
Fig. 2 is personal identification method flow chart;
Fig. 3 is cell phone application service content schematic diagram.
Specific embodiment
Present invention will be further explained below with reference to specific examples.
As shown in Figure 1, being a kind of vehicle anti-theft system based on human body multiple features fusion identification provided in this embodiment
System, comprising:
Image collecting device after receiving signal, acquires the facial image of user, is transferred in microprocessor, pass through
I2C bus is connect with microprocessor;Camera is used in the present embodiment;
Voice collection device after receiving signal, acquires the voice signal of user, is transferred in microprocessor, pass through
I2C bus is connect with microprocessor;Microphone is used in the present embodiment;
Vehicle-mounted ECU, for the vehicle-mounted ECU for controlling engine, the slot that system on chip is inserted in vehicle-mounted ECU is connected thereto.
SOC includes microprocessor, DRAM, EEPRAM, identification IP kernel and communication module.
Microprocessor receives identification signal code after signal, opens image collecting device and voice collection device, and
Start facial characteristics variation identification matching module, Application on Voiceprint Recognition matching module, characteristic matching module, can be carried out with communication module
Interaction, controls the opening and closing of vehicle.
DRAM (dynamic random access memory), for accessing collected picture and voice signal, and and identification
IP kernel is connected;
EEPROM (Electrically Erasable Programmable Read-Only Memory), for storing the eyes and nose of the legitimate user of preparatory typing
Textural characteristics, mouth lip moves feature and vocal print feature is respective and mouth lip is dynamic and the feature templates of vocal print Fusion Features,
And it is connected with microprocessor and identification IP kernel;
Identification IP kernel is made of facial characteristics variation identification module, voiceprint identification module, characteristic matching module, leads to
AXI interface is crossed to be connected with microprocessor;
Facial characteristics changes identification module, for proposing the facial image progress noise reduction process and feature of collected user
It takes, calculates the textural characteristics of each area dividing simultaneously using the concurrency of hardware;
Voiceprint identification module, for collected voice signal to be carried out noise reduction process and feature extraction;
The vocal print feature that lip moves feature and extracts is merged, is stored in advance with memory by characteristic matching module
Correlated characteristic template matched;
Communication module, the communication for vehicle and mobile terminal.
As shown in Fig. 2, be a kind of method for identifying ID provided in this embodiment, specifically includes the following steps:
Step A1: obtaining the eyes of preparatory typing legitimate user and the textural characteristics of nose etc., mouth lip move feature harmony
Line feature and mouth lip move the correlate template of feature and vocal print Fusion Features, store it in memory;
Step A2: after collecting facial image and voice signal, the image and vocal print of acquisition are pre-processed;
Step A3: while identification matching is carried out to facial variation characteristic and vocal print feature;
Step A4: if changes in faces feature and vocal print feature have a kind of matching unsuccessful, feeding back to microprocessor for result,
Microprocessor transmits information to mobile terminal by communication module, determined by car owner according to facial image and acoustic information be
No unlocking vehicle;
Step A5: if successful match, lip dynamic, vocal print feature being merged, then matched again, if
With unsuccessful, result is fed back into microprocessor, microprocessor transmits information to mobile terminal by communication module, by car owner
Decide whether unlocking vehicle according to facial image and acoustic information;If successful match, microprocessor feeds back to result vehicle-mounted
ECU unlocking vehicle;
For facial image, step A2 further comprises: collected facial image is first carried out median filtering then into
Column hisgram equalization, filters out the noise in image, improves signal-to-noise ratio;For voice signal, step A2 further comprises: if vehicle
Primary name is small strong, then the form of sound acquired is " hello, I requests using small strong vehicle ", pretreatment mainly includes going
It makes an uproar, preemphasis, framing and adding window.This method improves the accuracy of identification, avoid some criminals using photo and
The false defect verified and cause vehicle stolen of recording.
For facial image, step A3 further comprises: keeping face motionless as far as possible first, says password password.It will adopt
The image collected is divided into multiple homalographic region units, chooses two different positions operator respectively and calculates each area dividing
Gray level co-occurrence matrixes, then feature of the characteristic value of gray level co-occurrence matrixes as description image texture is calculated.Utilize hardware
Concurrency can calculate the textural characteristics of each area dividing simultaneously.The position of eyes and nose is determined according to textural characteristics,
To obtain the texture of the first width effective image, the texture of eyes obtains the textural characteristics of nose from multiple image.Pass through
The inverted triangle geometry that eyes, nose and mouth have, determines the position of mouth, and is partitioned into mouth area-of-interest
(ROI).The ROI image for comparing five frame of front and back analyzes the variation of mouth region, extracts mouth lip and moves feature.Be stored in advance
The eyes of user and the textural characteristics of nose and mouth lip move feature templates and carry out multiple features fusion matching.In matching, in advance
First set a threshold value T, the feature templates information of the feature and pre-stored legitimate user that only extract as identified person
When matching is more than threshold value T, illustrate the success of this characteristic matching.
For voice signal, step 3 further comprises the frequency spectrum for obtaining each short-time analysis window, is then filtered by Mel
Device group obtains Mel frequency spectrum, and cepstral analysis is carried out on Mel frequency spectrum, obtains the spy that MFCC carries out vocal print as characteristic parameter
Sign is extracted, if owner name be it is small strong, then user when carrying out authentication should with " hello, I request using it is small by force
The speech form of vehicle " is verified, and is discriminated whether as legitimate user.
Step A5 further comprises that the MFCC feature of the behavioral characteristics of lip, vocal print is carried out Fusion Features are as follows:In formula,For the weight of the behavioral characteristics of lip,It is special for the MFCC of vocal print
The weight of sign, diIndicate the similitude of ith featureα=
(a1,a2,……ak) it is the lip behavioral characteristics vector extracted, α '=(a '1,a′2,……a′k) it is that pre-stored lip is dynamic
State feature vector, dimension k;β=(b1,b2,……bj) it is the vocal print feature vector extracted, β '=(b '1,b′2,……b′j)
For pre-stored vocal print feature vector, dimension j;For the attenuation function of weight, its energy
Enough to guarantee to make its weight become very little when a certain characteristic matching effect is impacted larger, wherein a is set as 10, TiFor weights omegai's
Decision threshold,For the inverse of noise coefficient, in formula it can be seen that weight shared by δ more loud noise more
Small, η ' is fused feature.After fusion feature extracts, is matched with pre-stored fusion feature template, equally existed
When matching, a threshold value T, the only spy when the feature of identified person extracted and pre-stored legitimate user are preset
When the matching of sign Template Information is more than threshold value T, illustrate the success of this characteristic matching.
Identification matching when, preset a threshold value T, only when the feature of identified person extracted with it is pre-stored
When the matching of the feature templates information of legitimate user is more than threshold value T, illustrate the success of this characteristic matching.If matching is unsuccessful, then
This result is fed back into microprocessor, and microprocessor transmits information to mobile terminal by communication module according to result, and inquiry is
No authorization.If successful match, then result is fed back to the vehicle-mounted ECU of control engine by it after result feedback to microprocessor,
Unlocking vehicle.
Preferably, the mobile terminal is mobile phone or IPAD equipment or computer, using the APP on mobile terminal via
Communication module carries out data transmission, and realizes that APP changes Cha Gongneng, user's comprehensive verification result letter to the additions and deletions of user related data
Breath and authorization conditions, which control its vehicle, (to be issued enabled instruction or refuses vehicle launch and lock car door car window to carry out
Alarm).
As shown in figure 3, being the cell phone application service content of present system.In terms of carrying out any one of identification not
When can match, it can decide whether to authorize user when car owner receives result information in cell phone application, and also can
The current state of vehicle is checked by APP.The step of being operated using APP is as follows:
B1: car owner downloads installation APP software in mobile phone application software;
B2: it is logged in using account and password;
B3: after logining successfully, strictly according to the facts first fills in vehicle and personal information, then by filling in the ID of Vehicular system
Code is bound with vehicle.
After B4:B3 is completed, the state of vehicle at this time can be checked.
Car owner checks the use state of vehicle using cell phone application, and makes decision in the unsuccessful situation of identities match
Vehicle is opened, is closed or door and window locks and carries out alarm operation, very simple is convenient.
The above, only presently preferred embodiments of the present invention, not to the present invention in any form with substantial limitation,
It should be pointed out that under the premise of not departing from the method for the present invention, can also be made for those skilled in the art
Several improvement and supplement, these are improved and supplement also should be regarded as protection scope of the present invention.All those skilled in the art,
Without departing from the spirit and scope of the present invention, when made using disclosed above technology contents it is a little more
Dynamic, modification and the equivalent variations developed, are equivalent embodiment of the invention;Meanwhile all substantial technologicals pair according to the present invention
The variation, modification and evolution of any equivalent variations made by above-described embodiment, still fall within the range of technical solution of the present invention
It is interior.
Claims (10)
1. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification, it is characterised in that: including
Image collecting device acquires the facial image of user after receiving enabling signal, and is transferred to SOC on piece system
In system;
Voice collection device acquires the voice signal of user after receiving enabling signal, and is transferred to SOC on piece system
In system;
System on chip, it is special to the changes in faces of user for receiving the facial image and voice signal and being handled
Vocal print feature of seeking peace while identification matching is carried out, then moves feature and vocal print feature fusion using mouth lip and tested to carry out identity
Whether card, be legitimate user to determine.
2. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as described in claim 1, feature exist
In: the system on chip includes:
Microprocessor is sent out for receiving identification signal code after signal to described image acquisition device and voice collection device
Enabling signal is sent, and starts identification IP kernel, the opening and closing of control vehicle are interacted with communication module;
DRAM dynamic random access memory is known for accessing collected face picture and voice signal, and with the identity
Other IP kernel is connected;
EEPROM Electrically Erasable Programmable Read-Only Memory, for storing the eyes of the legitimate user of preparatory typing and the texture of nose
Feature, mouth lip move feature and vocal print feature is respective and mouth lip is dynamic and the feature templates of vocal print Fusion Features, and with institute
Microprocessor is stated to be connected with identification IP kernel;
The identification IP kernel is made of facial characteristics variation identification module, voiceprint identification module, characteristic matching module, and
It is connected with microprocessor;
The facial characteristics changes identification module, for the facial image of collected user to be carried out noise reduction process and feature
It extracts, calculates the textural characteristics of each area dividing simultaneously using the concurrency of hardware and mouth lip moves feature;
The voiceprint identification module, for collected voice signal to be carried out noise reduction process and vocal print feature extraction;
The characteristic matching module is stored in advance for merging the dynamic feature of mouth lip and vocal print feature with EEPROM
Correlated characteristic template matched;
The communication module, the communication for vehicle and mobile terminal.
3. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 2, feature exist
In: be connected thereto in the slot that the SOC is inserted in vehicle-mounted ECU, electric wave signal issued out of key, received as car antenna described in
Through controlling the vehicle-mounted ECU identification signal code of engine after electric wave signal, SOC system starts, micro- place are then driven
It manages device and opens image collecting device and voice collection device, and start the variation identification mould of the facial characteristics in identification IP kernel
Block, voiceprint identification module, characteristic matching module, verification result feeds back to the vehicle-mounted of control engine by microprocessor after verifying
ECU is controlled the opening and closing of vehicle by vehicle-mounted ECU or alarmed.
4. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 2, feature exist
In: the characteristic matching module matches while realizing facial image and vocal print feature first, in matching if there is any one
Kind characteristic matching is unsuccessful just exits identification, and mobile terminal is utilized to carry out the verifying of information.
5. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 2, feature exist
In: the authorization of vehicle is carried out by the mobile terminal, and carries out checking for vehicle-state in time.
6. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 5, feature exist
In: the mobile terminal is mobile phone, IPAD or computer.
7. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 1 or 2, feature
Be: the method for identifying ID of the system on chip specifically includes the following steps:
Step A1: the face texture feature of the legitimate user of preparatory typing is obtained, mouth lip moves feature and vocal print feature and mouth
Portion's lip moves the correlate template of feature and vocal print Fusion Features, and stores in memory;
Step A2: after the facial image and voice signal that collect user, to the facial image harmony message of the acquisition
It number is pre-processed, obtains face texture feature and vocal print feature;
Step A3: at the same to collected face texture feature and vocal print feature identified with the information in the memory
Matching;
Step A4: if face texture feature and vocal print feature have a kind of matching unsuccessful, result is fed back into microprocessor, micro- place
It manages device and mobile terminal is transmitted information to by communication module, decided whether out by car owner according to facial image and acoustic information
Open vehicle;
Step A5: if all successful match, lip dynamic, vocal print feature is merged, then matched again, if matching is not
Success, result is fed back into microprocessor, microprocessor transmits information to mobile terminal by communication module, by car owner according to
Facial image and acoustic information decide whether unlocking vehicle;If successful match, result is fed back to vehicle-mounted ECU and opened by microprocessor
Open vehicle.
8. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 7, feature exist
In: it in the step A3, carries out identifying matched detailed process for facial image are as follows: keep face motionless as far as possible first, say
Password password out;Acquired image is divided into multiple homalographic region units, chooses two different positions operator meter respectively
The gray level co-occurrence matrixes of each area dividing are calculated, then the characteristic value of gray level co-occurrence matrixes is calculated as description image texture
Feature;Calculate the textural characteristics of each area dividing simultaneously using the concurrency of hardware;According to textural characteristics determine eyes and
The position of nose, the textural characteristics of nose are to obtain the texture of the first width effective image, and the textures of eyes is from multiple image
Middle acquisition;The inverted triangle geometry having by eyes, nose and mouth, determines the position of mouth, and is partitioned into mouth sense
Interest region ROI;The ROI image for comparing several frames in front and back analyzes the variation of mouth region, extracts mouth lip and moves feature;With it is pre-
The eyes of the user first stored and the textural characteristics of nose and mouth lip move feature templates and carry out multiple features fusion matching;It is matching
When, preset a threshold value T, the feature templates of the feature and pre-stored legitimate user that only extract as identified person
When information matches are more than the threshold value T, illustrate the success of this characteristic matching.
9. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 7, feature exist
In: it in the step A3, carries out identifying matched detailed process for voice signal are as follows: obtain the frequency of each short-time analysis window
Spectrum, then obtains Mel frequency spectrum by Mel filter group, and cepstral analysis is carried out on Mel frequency spectrum, obtains MFCC as feature
Parameter carries out the feature extraction of vocal print, if owner name is " LL ", then user should be with " you when carrying out authentication
Good, I requests the vehicle using LL " speech form verify, discriminate whether as legitimate user.
10. a kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification as claimed in claim 7, feature exist
In: the step A5 is specifically included: the MFCC feature of the behavioral characteristics of lip, vocal print is carried out Fusion Features are as follows:In formula,For the weight of the behavioral characteristics of lip,It is special for the MFCC of vocal print
The weight of sign, diIndicate the similitude of ith feature, i is positive integer;For the lip dynamic of extraction
Feature vector, α '=(a '1,a′2,……a′k) it is pre-stored lip behavioral characteristics vector, dimension k;β=(b1,
b2,……bj) it is the vocal print feature vector extracted, β '=(b '1,b′2,……b′j) it is pre-stored vocal print feature vector, dimension
Number is j;For the attenuation function of weight, wherein a is setting coefficient, TiFor weights omegaiSentence
Determine threshold value,For the inverse of noise coefficient, η ' is fused feature;After fusion feature extracts, with
Pre-stored fusion feature template is matched, and equally in matching, is preset a threshold value T, is only worked as identified person
The feature extracted and the matching of feature templates information of pre-stored legitimate user when being more than the threshold value T, illustrate this
Characteristic matching success.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910116114.5A CN109910818B (en) | 2019-02-15 | 2019-02-15 | Vehicle anti-theft system based on human body multi-feature fusion identity recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910116114.5A CN109910818B (en) | 2019-02-15 | 2019-02-15 | Vehicle anti-theft system based on human body multi-feature fusion identity recognition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109910818A true CN109910818A (en) | 2019-06-21 |
CN109910818B CN109910818B (en) | 2021-10-08 |
Family
ID=66961599
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910116114.5A Active CN109910818B (en) | 2019-02-15 | 2019-02-15 | Vehicle anti-theft system based on human body multi-feature fusion identity recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109910818B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110363148A (en) * | 2019-07-16 | 2019-10-22 | 中用科技有限公司 | A kind of method of face vocal print feature fusion verifying |
CN110364163A (en) * | 2019-07-05 | 2019-10-22 | 西安交通大学 | The identity identifying method that a kind of voice and lip reading blend |
CN111862990A (en) * | 2020-07-21 | 2020-10-30 | 苏州思必驰信息科技有限公司 | Speaker identity verification method and system |
CN112491844A (en) * | 2020-11-18 | 2021-03-12 | 西北大学 | Voiceprint and face recognition verification system and method based on trusted execution environment |
CN113271587A (en) * | 2021-06-11 | 2021-08-17 | 焦作大学 | Internet of things credible authentication system for vehicle |
CN113347608A (en) * | 2021-06-11 | 2021-09-03 | 焦作大学 | Internet of things credibility authentication method for vehicle |
CN114241459A (en) * | 2022-02-24 | 2022-03-25 | 深圳壹账通科技服务有限公司 | Driver identity verification method and device, computer equipment and storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104376250A (en) * | 2014-12-03 | 2015-02-25 | 优化科技(苏州)有限公司 | Real person living body identity verification method based on sound-type image feature |
CN109145791A (en) * | 2018-08-09 | 2019-01-04 | 深圳大学 | One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system |
-
2019
- 2019-02-15 CN CN201910116114.5A patent/CN109910818B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104376250A (en) * | 2014-12-03 | 2015-02-25 | 优化科技(苏州)有限公司 | Real person living body identity verification method based on sound-type image feature |
CN109145791A (en) * | 2018-08-09 | 2019-01-04 | 深圳大学 | One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110364163A (en) * | 2019-07-05 | 2019-10-22 | 西安交通大学 | The identity identifying method that a kind of voice and lip reading blend |
CN110363148A (en) * | 2019-07-16 | 2019-10-22 | 中用科技有限公司 | A kind of method of face vocal print feature fusion verifying |
CN111862990A (en) * | 2020-07-21 | 2020-10-30 | 苏州思必驰信息科技有限公司 | Speaker identity verification method and system |
CN112491844A (en) * | 2020-11-18 | 2021-03-12 | 西北大学 | Voiceprint and face recognition verification system and method based on trusted execution environment |
CN113271587A (en) * | 2021-06-11 | 2021-08-17 | 焦作大学 | Internet of things credible authentication system for vehicle |
CN113347608A (en) * | 2021-06-11 | 2021-09-03 | 焦作大学 | Internet of things credibility authentication method for vehicle |
CN113347608B (en) * | 2021-06-11 | 2023-05-12 | 焦作大学 | Internet of things trusted authentication method for vehicle |
CN113271587B (en) * | 2021-06-11 | 2023-12-26 | 北京白龙马云行科技有限公司 | Internet of things trusted authentication system for vehicle |
CN114241459A (en) * | 2022-02-24 | 2022-03-25 | 深圳壹账通科技服务有限公司 | Driver identity verification method and device, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109910818B (en) | 2021-10-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109910818A (en) | A kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification | |
CN110217196B (en) | Vehicle door control system and method | |
CN107122641B (en) | Intelligent equipment owner identification method and intelligent equipment owner identification device based on use habit | |
CN106599660A (en) | Terminal safety verification method and terminal safety verification device | |
CN107472195A (en) | A kind of car door automatic opening system and method based on facial image track identification | |
CN107909040B (en) | Car renting verification method and device | |
CN110392082B (en) | Vehicle control method and system | |
CN106627486A (en) | Identity identification key system for rented automobile and control method | |
DE102012217728A1 (en) | METHOD AND DEVICES FOR USER AUTHENTICATION AND SAFETY | |
CN1972186A (en) | A mobile identity authentication system and its authentication method | |
CN110827453A (en) | Fingerprint and voiceprint double authentication method and authentication system | |
CN105721460A (en) | Voiceprint password vehicle control system and method | |
Pawar et al. | IoT based embedded system for vehicle security and driver surveillance | |
CN112819997A (en) | Method, apparatus and computer-readable storage medium for creating vehicle key | |
CN110077361B (en) | Vehicle control method and device | |
CN113643489B (en) | Public transit pos machine based on face identification | |
CN108583508B (en) | Keyless opening and closing method and device for vehicle door | |
CN106599843A (en) | Fingerprint authentication method and device | |
CN109285260A (en) | A kind of gate control method and device | |
CN108674367A (en) | A kind of automobile driver's vehicle control method and the system for realizing this method | |
CN110217194B (en) | Shared automobile control method and device and electronic equipment | |
CN107491759B (en) | Mixed mode registration method and device | |
CN215006766U (en) | Entrance guard and ignition system based on multiple biological recognition technology | |
CN108508862A (en) | A kind of authentication system and vehicle for vehicle | |
CN115862200A (en) | Shared automobile management method and device |
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 |