CN110348192A - The authentication method of biological characteristic - Google Patents
The authentication method of biological characteristic Download PDFInfo
- Publication number
- CN110348192A CN110348192A CN201810890260.9A CN201810890260A CN110348192A CN 110348192 A CN110348192 A CN 110348192A CN 201810890260 A CN201810890260 A CN 201810890260A CN 110348192 A CN110348192 A CN 110348192A
- Authority
- CN
- China
- Prior art keywords
- image
- template
- merging
- information
- multiple biometric
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 230000001815 facial effect Effects 0.000 claims description 17
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 239000003086 colorant Substances 0.000 claims 1
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012850 discrimination method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000005549 size reduction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- 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
- G06V40/12—Fingerprints or palmprints
-
- 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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Security & Cryptography (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Ophthalmology & Optometry (AREA)
- Computer Hardware Design (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention provides a kind of authentication method of biological characteristic, includes: multiple biometric images of a user is obtained, wherein multiple biometric image includes one first biometric image and one second biometric image of different types of biological characteristic；One, which is generated, according to multiple biometric image merges image；And judge whether the merging image is similar to a template image.
Description
Technical field
A kind of authentication method of the present invention about biological characteristic, espespecially a kind of side that identification is carried out using biological characteristic
Method.
Background technique
There are many mobile devices to use the identity of facial image, fingerprint image or iris image identification user at present.
But every kind of identification mode has its insufficient place, therefore further improvement still in need.
Summary of the invention
Therefore, one of present invention is designed to provide a kind of authentication method of biological characteristic to solve above-mentioned prior art institute
Problems faced.
One embodiment of the invention provides a kind of authentication method of biological characteristic, includes: obtaining the multiple of a user
Biometric image, wherein multiple biometric image includes one first biometric image and one second biology of different types of biological characteristic
Image；One, which is generated, according to multiple biometric image merges image；And judge whether the merging image is similar to a template image.
According to the present invention, the identification of two kinds of biological characteristics can be completed at the same time, with high security, high-efficient advantage,
And additional hardware resource is not needed.
Detailed description of the invention
Fig. 1 shows the trusted area of Android operating system.
Fig. 2 is the flow chart of personal identification method of the present invention.
Fig. 3 is the schematic diagram that the sub-step for the step of merging image is generated in Fig. 2.
Fig. 4 is the schematic diagram according to the merging image of one embodiment of the invention.
Fig. 5 is to judge to merge an image embodiment whether similar with template image in Fig. 2.
Wherein, appended drawing reference:
100 operating systems
The 110 general worlds
112,116 hardware abstraction layer
114,118 driver
120 safer worlds
122 trusted applications
124 packet elements
126SPI/API interface
130 fingerprint sensing devices
140 camera models
200 methods
202~206,2042,2044,2046,2062,2064 steps
Specific embodiment
Below in conjunction with the drawings and specific embodiments, the present invention will be described in detail, but not as a limitation of the invention.
Present invention could apply to a variety of different operating systems, such as Android (Android) operating system or Microsoft
Form (windows) operating system.For convenience of description, illustrate running of the invention by taking Android operating system as an example below.Fig. 1 is aobvious
Show the trusted area (Trust Zone) 100 of an Android operating system.Trusted area 100 includes the general world (normal
Word) 110, safer world (secure world) 120.The general world 110 includes hardware abstraction layer (Hardware
Abstraction Layer, HAL) 112,116 and driver (Kernel Driver) 114,118.Safer world 120 includes
Trusted application (Trusted Application, TA) 122, packet element (wrapper) 124 and SPI/API interface
(Serial Peripheral Interface (SPI)/Application Programming Interface (API) 126,
Wherein trusted application 122 is used to execute various operations (identification is such as executed in the environment for having security fence), storage
The movement such as storage sample version (Template) and acquisition image.Fingerprint sensing device 130 is used to capture the fingerprint image of user.Fingerprint sense
It surveys device 130 and connects SPI/API interface 126, and packet element 124 obtains fingerprint image from SPI/API interface 126, and by fingerprint
Image is transmitted to trusted application 122.Driver 114 is responsible for receiving the interrupt request that fingerprint sensing device 130 transmits
(Interrupt Request, IRQ) signal, and inform that hardware abstraction layer 112 has the event of touch fingerprint sensor 130.
The function of hardware abstraction layer 112 first is that after receiving the interrupt request that kernel driver 114 transmits, inform and believable apply journey
Sequence 122 is received and is recognized the operation such as fingerprint image, and in addition a function is returned receiving trusted application 122
The identification of fingerprint of report is as a result, pass on to the system of more top.It is enterprising in the electronic device of installation Android operating system at present
The framework of row identification of fingerprint is as shown in Figure 1.Relevant details is the personage institute that is familiar with Android operating system or identification field
Known, details are not described herein.
The fingerprint sensing device 130 of Fig. 1 and camera model 140 by installation Android operating system electronic device (such as wisdom
Type mobile phone) it is provided.Camera model 140 is used to shoot the image of face of user.Camera model 140 is connected to driver 118,
And driver 118 is connected to hardware abstraction layer 116.Image of face captured by camera model 140 can be transmitted via driver 118
To hardware abstraction layer 116.Hardware abstraction layer 116 is connected to hardware abstraction layer 112, and 112 received face, institute of hardware abstraction layer
Image can send trusted application 122 to via hardware abstraction layer 112.In other embodiments, hardware is taken out as layer 116
It may connect to trusted application 122, to transmit image of face to trusted application 122.
In the prior art, two kinds of biological identifications (such as identification of fingerprint and human face recognition) is carried out, there are two types of possible sides
Formula may be implemented.A kind of possible mode is by single trusted application (such as trusted application of the invention
122), for example, trusted application has first carried out a kind of biological identification and then has executed another biological identification, this
The shortcomings that kind mode is that efficiency is bad.Alternatively possible mode is exactly additionally to increase a trusted application again.By two
A trusted application carries out above two biological identification simultaneously.The shortcomings that this mode is to need more hardware resource.
Fig. 2 is referred to, Fig. 2 is the flow chart of personal identification method 200 of the present invention.At least the one of personal identification method 200
Partially (such as a part of or whole) is executed by trusted application 122.Firstly, in step 202, by one or more
A different images acquisition device obtains multiple biometric images of user, wherein multiple biometric image includes at least different type
Biological characteristic one first biometric image and one second biometric image.The first or second biometric image can be face figure
Picture, fingerprint image or iris image.For convenience of explanation, below in an example, the first biometric image is facial image,
Second biometric image is fingerprint image, but the present invention is not limited thereto.
Trusted application 122 executes step 204 after obtaining multiple biometric image.In step 204, believable
Application program 122 generates a merging image according to multiple biometric image.One embodiment of step 204 is as shown in figure 3, wherein
Including for 3 sub-steps 2042,2044 and 2046.Step 2042 includes the size of adjustment biometric image, such as will be obtained
Multiple biometric images are adjusted to identical size.For example, the size of fingerprint image is 160*160, the size of facial image
For 80*80, adjusting the mode of size for the size reduction of fingerprint image is 80*80, or by the size of facial image with interpolation
Mode be converted into 160*160.In another embodiment, it is preset that the size of facial image and fingerprint image is all adjusted to one
Size, so that facial image is identical as the size of fingerprint image.Next, will do it step 2044 to adjust the side of biometric image
To.The direction of biometric image is determined according to the content of the biometric image.For example, the direction of facial image can be basis
The position of two and nose determines.The direction of fingerprint image can be the phase of the position according to characteristic points multiple in fingerprint image
Relationship is determined.In one embodiment, to be identified to adjust according to the direction of facial image registered in operating system
Facial image direction, and according to the direction of fingerprint image registered in operating system, to adjust fingerprint to be identified
The direction of image.In one embodiment of step 2044, the direction of facial image and fingerprint image is all adjusted to such as Fig. 4
Y-direction, however, the present invention is not limited thereto.
Followed by step 2046 to merge multiple biometric image.For example, by facial image and fingerprint image
Result after merging is the equal of situation shown in Fig. 4.Facial image shown in Fig. 4 includes the image letter of tri- figure layers of RGB
Breath, and fingerprint image is the equal of the image information of a figure layer.Therefore, facial image and fingerprint image are merged,
It is equivalent to and obtains the image comprising four figure layers.
Step 206 is carried out after image after being merged.In step 206, trusted application 122 judges the conjunction
And whether image is similar to a template image.The template image is by facial image registered in operating system and fingerprint image
It is constituted.For example, user can first register its face image and fingerprint image in operating system, by step
202, the facial image of the registration and fingerprint image can be merged into a template image by 204 method.
In one embodiment, if judging that the merging image is similar to a template image in step 206, believable application
Program 122 provides an output valve " 1 " and indicates to recognize successfully, if judging the merging image and a template image not in step 206
Similar, then trusted application 122, which provides an output valve " 0 ", indicates identification failure.
Fig. 5 illustrates one embodiment of step 206.Step 206 includes sub-step 2062 and 2064.Step 2062 includes:
Extraction captures the feature of the merging image, to generate a characteristic information F.In step 2062, an image identification can be used and drill
Algorithm process merging image, to extract the feature of the merging image.The image identification algorithm can be (but being not limited to):
Convolutional neural networks (Convolutional Neural Networks, CNN) algorithm, local binary pattern (Local
Binary Patterns, LBP) algorithm, histograms of oriented gradients (Histogram of Oriented Gradient, HOG)
In algorithm or scale invariant feature conversion (Scale Invariant Feature Transform, SIFT) algorithm at least
One extraction.Above-mentioned a variety of algorithms can select a progress, and can also arrange in pairs or groups utilization.
After obtaining the characteristic information F to be identified for merging image, step 2064 is then carried out.Step 2064 judgement
The similitude of this feature information F and a template information T.Template information T is according to template image characteristic information obtained.By
The algorithm that step 2062 uses can extract the feature of the template image of registration and generate template information T.Example edition information T storage
There are in the archives economy of trusted application 122, when carrying out step 2064, trusted application 122 takes out storage
The template information T characteristic information F that merges image with to be identified carry out operation.In one embodiment, this feature information F and template
Information T respectively includes N number of coefficient, and wherein N is positive integer.Step 2064 with this feature information F and template information T carry out it is European away from
From (Euclidean Distance) operation to generate one first numerical value.First numerical value indicates that this feature information F and the template believe
The difference between T is ceased, the difference between the merging image and template image is also represented.First numerical value is smaller, indicates the merging figure
As more similar to template image.First numerical value is bigger, indicates that the merging image and template image are more dissimilar.The believable application
Program 122 judges that this feature information F and the example edition information T-phase are seemingly or dissimilar according to first numerical value, to generate an output valve.
In one embodiment, trusted application 122 compares first numerical value and a threshold value.When first numerical value is less than or equal to
When (that is, being not more than) threshold value, trusted application 122 generates an output valve " 1 ".Output valve " 1 " indicates the merging
Image is similar to template image, and the result of identification is to pass through (pass).It is credible when first numerical value is greater than the threshold value
Application program 122 is relied to generate an output valve " 0 ".Output valve " 0 " represents the merging image and template image is dissimilar, and identity is known
Other result is failure (fail).In other examples, trusted application 122 can be generated according to first numerical value
One score (score), and the score is made comparisons with a preset fraction, to generate the output valve.Above-mentioned Euclidean distance operation
It is to judge the similitude between this feature information F and template information T.In various embodiments, other similarity measurements
(Similarity Measurement) method of measuring, such as manhatton distance (Manhattan Distance), Chebyshev away from
From (Chebyshev Distance), Minkowski Distance (Minkowski Distance), mahalanobis distance
(Mahalanobis Distance), Hamming distance (Hamming distance) and correlation distance (Correlation
Distance it) is applicable to realize step 2064.
From above description, it will be seen that, step 206 is similar to the principle of current human face recognition.Therefore, for human face recognition
For the usual skill in field, when will appreciate that above-mentioned content and accordingly implement the present invention.
From above description, it will be seen that, the present invention can be completed at the same time two kinds of biologies using single trusted application and distinguish
Know, has the advantages that hardware resource that is high-efficient, and not needing be additional.
On the other hand, the above embodiments are the equal of recognizing fingerprint and face simultaneously, when both biological characteristics simultaneously
When meeting when registered template image, authentication could be passed through, therefore, the present invention can greatly improve the peace of identification
Quan Xing.
Furthermore the present invention can improve rate of accidentally refusing (False Rejection Rate, FRR) and accidentally by rate (False
Acceptance Rate, FAR) bad problem, wherein " accidentally by " refers to that the biological characteristic for not meeting registration template is but judged to
It is broken into and meets, and " accidentally refusing " refers to meets the biological characteristic of registration template and be judged to be broken into and do not meet.When accidentally lower by rate, safety
Property is better, but accidentally the rate of refusing can be improved thus.Accidentally refuse rate it is lower when, the use experience of user better, but accidentally will be improved by rate.
The accidentally rate of refusing of industry either identification of fingerprint or human face recognition is about 3% at present, is about accidentally 1/50,000 by rate, using this
When the discrimination method of invention, accidentally the rate of refusing it can will force down to 1% that (but this will cause and is accidentally mentioned by rate by threshold value above-mentioned is adjusted
Up to 1/5,000, because accidentally refusing rate and accidentally by trade-off relation is presented between rate).According to the present invention, by fingerprint image and face figure
The combined image of picture is recognized, and always accidentally the rate of refusing can be the respective summation for accidentally refusing rate of the two, that is, 1%+1%=2%.And
Total can be accidentally product of the respective mistake of the two by rate, that is, (1/5,000) * (1/5,000)=1/ (25*10 by rate6).With it is existing
Some identification of fingerprint or human face recognition are compared, the present invention can force down accidentally refuse rate in the case where, still can obtain low-down
Accidentally by rate, safety is taken into account and user experiences.
The foregoing is merely presently preferred embodiments of the present invention, all equivalent changes done according to scope of the present invention patent with
Modification, is all covered by the present invention.
Claims (7)
1. a kind of authentication method of biological characteristic, characterized by comprising:
Multiple biometric images of a user are obtained, wherein multiple biometric image includes the one the of different types of biological characteristic
One biometric image and one second biometric image；
One, which is generated, according to multiple biometric image merges image；And
Judge whether the merging image is similar to a template image.
2. the method as described in claim 1, which is characterized in that first biometric image is the facial image of the user,
And second biometric image is the fingerprint image of the user.
3. method according to claim 2, which is characterized in that the merging image includes that the RGB three primary colors of the facial image are believed
The finger print information of breath and the fingerprint image.
4. the method as described in claim 1, which is characterized in that this generates the step of the merging image according to multiple biometric image
Suddenly include:
At least one of size in multiple biometric image is adjusted, multiple biometric image is of the same size；
Adjust the direction of multiple biometric image；And
Multiple biometric image adjusted is merged.
5. the method as described in claim 1, which is characterized in that this judges whether the merging image and a template image are similar
Step includes:
A characteristic information is generated according to the feature of the merging image；And
Judge the similitude of this feature information Yu a template information, wherein the template information is obtained according to the template image.
6. method as claimed in claim 5, which is characterized in that the feature according to the merging image generates a characteristic information
Step includes handling the merging image using an image identification algorithm, which drills comprising convolutional neural networks
Algorithm, local binary pattern algorithm, histograms of oriented gradients algorithm, at least one in scale invariant feature conversion calculus method
Person.
7. method as claimed in claim 5, which is characterized in that the similitude of judgement this feature information and a template information
Step includes carrying out a similarity measurement according to this feature information and the template information to generate one first numerical value, the similarity measurements
Amount be Euclidean distance, manhatton distance, Chebyshev's distance, Minkowski Distance, mahalanobis distance, Hamming distance to it is related away from
One of from.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201862651270P | 2018-04-02 | 2018-04-02 | |
| US62/651,270 | 2018-04-02 | ||
| TW107124570A TW201942780A (en) | 2018-04-02 | 2018-07-17 | Method for identifying biological characteristics |
| TW107124570 | 2018-07-17 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110348192A true CN110348192A (en) | 2019-10-18 |
Family
ID=68173323
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810890260.9A Pending CN110348192A (en) | 2018-04-02 | 2018-08-07 | The authentication method of biological characteristic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110348192A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116741A (en) * | 2013-01-28 | 2013-05-22 | 天津理工大学 | Capture and identification system for blending images of palm veins and palm prints |
| CN103886283A (en) * | 2014-03-03 | 2014-06-25 | 天津科技大学 | Method for fusing multi-biometric image information for mobile user and application thereof |
| US20140321718A1 (en) * | 2013-04-24 | 2014-10-30 | Accenture Global Services Limited | Biometric recognition |
| CN105227860A (en) * | 2014-07-02 | 2016-01-06 | 索尼公司 | Image generating method, device and mobile terminal |
| CN107609365A (en) * | 2016-07-11 | 2018-01-19 | 三星电子株式会社 | For the method and apparatus using multiple biometrics validators checking user |
-
2018
- 2018-08-07 CN CN201810890260.9A patent/CN110348192A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116741A (en) * | 2013-01-28 | 2013-05-22 | 天津理工大学 | Capture and identification system for blending images of palm veins and palm prints |
| US20140321718A1 (en) * | 2013-04-24 | 2014-10-30 | Accenture Global Services Limited | Biometric recognition |
| CN103886283A (en) * | 2014-03-03 | 2014-06-25 | 天津科技大学 | Method for fusing multi-biometric image information for mobile user and application thereof |
| CN105227860A (en) * | 2014-07-02 | 2016-01-06 | 索尼公司 | Image generating method, device and mobile terminal |
| CN107609365A (en) * | 2016-07-11 | 2018-01-19 | 三星电子株式会社 | For the method and apparatus using multiple biometrics validators checking user |
Non-Patent Citations (1)
| Title |
|---|
| 张露 等: "基于分类距离分数的自适应多模态生物特征融合", 《计算机研究与发展》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107438854B (en) | System and method for performing fingerprint-based user authentication using images captured by a mobile device | |
| Jain et al. | Biometric identification | |
| CN102332093B (en) | Identity authentication method and device adopting palmprint and human face fusion recognition | |
| Delac et al. | A survey of biometric recognition methods | |
| Tian et al. | KinWrite: Handwriting-Based Authentication Using Kinect. | |
| Jain et al. | Biometrics: a grand challenge | |
| CN101661557B (en) | Face recognition system and face recognition method based on intelligent card | |
| US20190392129A1 (en) | Identity authentication method | |
| KR20170046448A (en) | Method and device for complex authentication | |
| KR20150143304A (en) | Image recognition device and method for registering feature data in image recognition device | |
| KR20170082074A (en) | Face recognition apparatus and method using physiognomic information | |
| Charity et al. | A bimodal biometrie student attendance system | |
| Bong et al. | Palm print verification system | |
| JP2002236666A (en) | Personal authentication device | |
| Li et al. | Adaptive deep feature fusion for continuous authentication with data augmentation | |
| Deriche | Trends and challenges in mono and multi biometrics | |
| Tahmasebi et al. | Robust intra-class distance-based approach for multimodal biometric game theory-based rank-level fusion of ear, palmprint and signature | |
| WO2014132570A1 (en) | Authentication device, authentication method and program storage medium | |
| Yin et al. | Fusion of face recognition and facial expression detection for authentication: a proposed model | |
| Dewangan | Human authentication using biometric recognition | |
| Kumar | Hand image biometric based personal authentication system | |
| CN110348192A (en) | The authentication method of biological characteristic | |
| Sanches et al. | A single sensor hand biometric multimodal system | |
| Malik et al. | Personal authentication using palmprint with Sobel code, Canny edge and phase congruency feature extraction method | |
| TWI792017B (en) | Biometric identification system and identification method thereof |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191018 |
|
| WD01 | Invention patent application deemed withdrawn after publication |