A kind of fingerprint identification method and device
Technical field
The present invention relates to technical field of biometric identification, more particularly, to a kind of fingerprint identification method and device.
Background technology
Fingerprint recognition, as one of most widely used in biological identification technology, cheapest technology of identification of price, has good
Good development prospect.
The principle of fingerprint recognition is:Finger is detected to touch, be pressed into opening under Fingerprint Identification Unit or Fingerprint Identification Unit
Close, launch light, penetrate fingerprint recognition cover-plate glass, reflexes Skin superficial makings simultaneously receives, form fingerprint image;Process and divide
Analysis fingerprint image, the storehouse finding clearly characteristic point and original fingerprint feature point is mated, if consistent, is proved to be successful, such as
Really inconsistent then authentication failed.
The front fingerprint recognition part of the optical finger print module in current Fingerprint Identification Unit or Fingerprint Identification Unit, due to printing opacity
Need to be all glass material, and the lipophile of glass material, significantly fingerprint can be left clearly after user's checking.When again
Secondary triggering touch event of not destroying in the case of original fingerprint is it is possible to again identify that fingerprint, and unlocks.Like this, when with
After terminal is lost at family, if the last fingerprint of Fingerprint Identification Unit residual is clear enough, will be unlocked, very dangerous.
Content of the invention
The embodiment of the present invention provides a kind of fingerprint identification method and device, can effectively prevent non-user using residual
Fingerprint is unlocked, and improves the safety of unlocked by fingerprint.
In a first aspect, embodiments providing a kind of method of fingerprint recognition, including:
Obtain fingerprint image to be verified;
If it is determined that described fingerprint image to be verified is mated with history fingerprint image, then refusal carries out fingerprint recognition.
Second aspect, embodiments provides a kind of device of fingerprint recognition, including:
Fingerprint image acquisition module, for obtaining fingerprint image to be verified;
Fingerprint identification module, for if it is determined that described fingerprint image to be verified is mated with history fingerprint image, then refusing
Carry out fingerprint recognition.
The method and apparatus of fingerprint recognition provided in an embodiment of the present invention, by by fingerprint image to be verified and history fingerprint
Image carries out the matching analysis, effectively prevent non-user and is identified using latent prints, has reached raising fingerprint recognition
The beneficial effect of safety.
Brief description
Figure 1A is the flow chart of one of the embodiment of the present invention one fingerprint identification method;
Figure 1B is fingerprint image to be verified and the history fingerprint image of one of the embodiment of the present invention one fingerprint identification method
As method flow diagram when mismatching;
Fig. 2A is a kind of fingerprint schematic diagram of one of the embodiment of the present invention two fingerprint identification method;
Fig. 2 B is the first fingerprint image schematic diagram to be verified of one of the embodiment of the present invention two fingerprint identification method;
Fig. 2 C is the first history fingerprint image schematic diagram of one of the embodiment of the present invention two fingerprint identification method;
Fig. 2 D is the first fingerprint image to be verified and first of one of the embodiment of the present invention two fingerprint identification method
Plant the fingerprint image schematic diagram after the superposition of history fingerprint image;
Fig. 2 E is the second fingerprint image to be verified schematic diagram of one of the embodiment of the present invention two fingerprint identification method;
Fig. 2 F is the second history fingerprint image schematic diagram of one of the embodiment of the present invention two fingerprint identification method;
Fig. 2 G is the second fingerprint image to be verified and second of one of the embodiment of the present invention two fingerprint identification method
Plant the fingerprint image schematic diagram after the superposition of history fingerprint image;
Fig. 3 is the flow chart of one of the embodiment of the present invention three fingerprint identification method;
Fig. 4 is the flow chart of one of the embodiment of the present invention four fingerprint identification method;
Fig. 5 is the flow chart of one of the embodiment of the present invention five fingerprint identification method;
Fig. 6 is the flow chart of one of the embodiment of the present invention six fingerprint identification method;
Fig. 7 A is the flow chart of one of the embodiment of the present invention seven fingerprint identification method;
Fig. 7 B is the detail flowchart of one of the embodiment of the present invention seven fingerprint identification method;
Fig. 8 is the structural representation of one of the embodiment of the present invention eight fingerprint identification device;
Fig. 9 is the structural representation of one of the embodiment of the present invention nine fingerprint identification device.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.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
Illustrate only part related to the present invention rather than entire infrastructure in description, accompanying drawing, and be not restricted to the step of the present embodiment
Rapid order.
Embodiment one
A kind of flow chart of fingerprint identification method that Figure 1A provides for the embodiment of the present invention one, the present embodiment is applicable to know
The situation of other latent prints, the method can be executed by fingerprint identification device provided in an embodiment of the present invention, and this device can be adopted
Realized with the mode of software and/or hardware, software form can be the authentication application in the internet virtual world, and this device can be integrated
In any equipment needing fingerprint recognition, such as mobile terminal and fixed terminal, mobile terminal includes smart mobile phone, flat board electricity
Brain, PDA, intelligent watch etc., fixed terminal includes desktop computer, ATM, door lock, Time Attendance Device etc., as shown in figure 1, the method tool
Body comprises the steps:
S110, acquisition fingerprint image to be verified.
Specifically, fingerprint refers to the convex recessed uneven streakline producing of the positive surface skin of the finger tips of people.Streakline is regular
The different stricture of vagina type of arrangement form.The details of the starting point of streakline, terminal, binding site and bifurcation, referred to as fingerprint.By referring to
Stricture of vagina identification obtains fingerprint image to be verified, is differentiated by the details of relatively different fingerprints.Due to everyone
Fingerprint different, even if between ten fingers of same people, fingerprint is also clearly distinguished from, and therefore fingerprint can be used for identity authentication.Due to
The orientation carrying out finger scan every time is not exclusively the same, and impetus difference also brings along different degrees of deformation, so can deposit again
Obscuring fingerprint in a large number, so the key of fingerprint identification technology is intended to correctly extract feature and realizes correct coupling.
If it is desired to identification fingerprint, first have to first set up fingerprint base, allow the original fingerprint data of fingerprint become initial identification code,
Fingerprint scanning systems only store " finger print data ", and processor can be encrypted to these data.Presently used fingerprint image
Collection, is substantially based on three kinds of technical foundation:Optical technology, semiconductor silicon technology, ultrasonic technology.Wherein, optical technology is adopted
Collection fingerprint is that history is the most remote, most popular technology, is recorded by light catoptric imaging and verifies fingerprint.
If it is determined that described fingerprint image to be verified is mated with history fingerprint image, then refusal carries out fingerprint recognition to S120.
Wherein, history fingerprint image, i.e. latent prints image, is the fingerprint image that the user of storage successfully identifies for the last time
Picture;Fingerprint image to be verified is mated with history fingerprint image, can be by calculating fingerprint image to be verified and history fingerprint image
As the registration of superposition is compared the pixel it is also possible to after calculating fingerprint image to be verified is superimposed with history fingerprint image
Value is compared, and first can also process fingerprint image to be verified with history Fingerprint Image Binarization, then to binary conversion treatment
Image afterwards carries out image overlay analysis etc..Specifically, fingerprint identification technology has in good anti-theft feature and personal information
The secret protection held.But, fingerprint is but to accompany us not modifiable throughout one's life, if the fingerprint that we pass through record is illegal
Collect it is meant that we encrypt mobile phone, bank account, fingerprint gate lock etc. using fingerprint identification function there is great peace
Full hidden danger.For example, after user's lost mobile phone, if the Fingerprint Identification Unit last fingerprint of residual is clear enough, will be by
Unblock.Therefore, it is necessary to first determine whether described fingerprint image to be verified is history fingerprint after obtaining fingerprint image to be verified
Image, if mated with history fingerprint image, then can determine as latent prints image, and refusal carries out fingerprint recognition;If
Mismatch with history fingerprint image, then can determine as not being latent prints image, fingerprint recognition can be carried out.
The fingerprint identification method that the present embodiment provides, by being mated fingerprint image to be verified with history fingerprint image
Analysis, be effectively prevent non-user and is identified using latent prints, has reached and has improved the beneficial of fingerprint recognition safety
Effect.
On the basis of above-described embodiment, history fingerprint image can extract from history fingerprint image data storehouse, and history refers to
The foundation of print image data base is the fingerprint image obtaining use after each fingerprint recognition success, using described fingerprint image as
History fingerprint image.
For example, the i Phone that user A uses, first collects and records the index fingerprint of user A in mobile phone by pressing HOME key
Data, as long as so mobile phone can identify the finger print data of user it is possible to unlock.Unblock falls in HOME key user A every time
Fingerprint is different due to pressing power, scope, leads to the fingerprint of HOME key residual all different.But, as long as fingerprint clear, that
As long as user B is put a lighttight black glass in HOME key and pressed, the touch time of fingerprint recognition will be made to touch
Send out, checking fingerprint is correct, successfully unlocks.Specifically for example, user A using mobile phone, stays the fingerprint of HOME key for the last time
For fingerprint a, and gather image as history fingerprint image, record is in history fingerprint image data storehouse.User B is referred to using residual
The fingerprint that stricture of vagina pressing obtains is fingerprint b, now in order to prevent fingerprint b to be identified successfully, before identification fingerprint B, first obtains record
Fingerprint a image in history fingerprint image data storehouse, fingerprint a image is mated with fingerprint b image, if mated into
Work(, then can determine as latent prints image, refusal carries out fingerprint recognition.
If it is determined that described fingerprint image to be verified is mated with history fingerprint image, then illustrate that fingerprint image to be verified is residual
Stay fingerprint image, refusal carries out fingerprint recognition.However, if it is determined that described fingerprint image to be verified and history fingerprint image are not
Join, then illustrate that fingerprint image to be verified is not latent prints image, then fingerprint recognition is carried out to fingerprint image to be verified, and
Obtain the fingerprint image being used as history fingerprint image after each fingerprint recognition success, the flow chart of fingerprint identification method is such as
Shown in Figure 1B.
For example, the history fingerprint image of above-mentioned user A is fingerprint a, if user A wants to open mobile phone again, pins
HOME key, because defevent experiment pressing is it is impossible to as fingerprint a, so have new fingerprint occur, by fingerprint recognition
Obtaining fingerprint image to be verified is fingerprint e.Fingerprint e and history fingerprint image fingerprint a mismatches, then fingerprint image to be verified is described
Fingerprint e is not latent prints image, obtains the fingerprint image fingerprint e being used as new history fingerprint image after fingerprint recognition success
Picture.
Embodiment two
A kind of flow chart of fingerprint identification method that Fig. 2A provides for the embodiment of the present invention two, the present embodiment is in above-mentioned enforcement
It is optimized on the basis of example, there is provided determine the processing method that described fingerprint image to be verified is mated with history fingerprint image,
Specifically, described fingerprint image to be verified is overlapped with described history fingerprint image, obtains being superimposed fingerprint image;Determine institute
State fingerprint image to be verified and the registration of superposition fingerprint image, if described registration exceedes default registration it is determined that described
Fingerprint image to be verified is mated with history fingerprint image.
Accordingly, the method for the present embodiment includes:
S210, acquisition fingerprint image to be verified.
S220, described fingerprint image to be verified is overlapped with described history fingerprint image, obtains being superimposed fingerprint image.
In order to preferably determine described fingerprint image to be verified and history fingerprint image matching degree, need to described to be tested
Card fingerprint image and described history fingerprint image are overlapped processing.Wherein it is possible to it is complete using described fingerprint image to be verified
Portion's image is overlapped with all images of described history fingerprint image, obtains being superimposed fingerprint image;Can also be described to be tested
After all images of card fingerprint image are overlapped with all images of described history fingerprint image, intercept parts of images as folded
Adding fingerprint image, intercepting can be partly centre position or optional position;Described fingerprint image to be verified can also first be intercepted
Parts of images and described history fingerprint image parts of images, then be overlapped, obtain being superimposed fingerprint, image interception part can
Think described fingerprint image to be verified and the centre position of described history fingerprint image or any position of user's setting
Put, but fingerprint image to be verified described in premise is identical with the position that described history fingerprint image intercepts.
S230, determine described fingerprint image to be verified and the registration of described superposition fingerprint image, if described registration surpasses
Cross default registration it is determined that described fingerprint image to be verified is mated with history fingerprint image, refusal carries out fingerprint recognition.
Wherein, can intuitively observe whether described fingerprint image to be verified is latent prints using registration, preset weight
Right can be the quiescent value of system default, and the scope as 80% or personal settings is any in 0%~100%
Value.Specifically, after obtaining being superimposed fingerprint image, fingerprint recognition system obtains described fingerprint image to be verified through image procossing
Picture and the registration of superposition fingerprint image.If described registration exceedes default registration it is determined that described fingerprint image to be verified
As mating with history fingerprint image, refusal carries out fingerprint recognition;If described registration is not less than default registration it is determined that institute
State fingerprint image to be verified to mismatch with history fingerprint image, carry out fingerprint recognition.
For example, the default registration of above-mentioned user A is set as the quiescent value 80% of system default.Obtain above-mentioned user A's
History fingerprint image fingerprint a as shown in Figure 2 B, fingerprint image fingerprint b to be verified as shown in Figure 2 C, by all images of fingerprint a with
After all images of fingerprint b are overlapped, draw the fingerprint image of superposition as shown in Figure 2 D, intercept mid portion image as folded
Adding fingerprint image c.Obtaining fingerprint image fingerprint b to be verified with the registration being superimposed fingerprint image c through image procossing is 90%,
Due to exceeding default registration 80%, determine that fingerprint image fingerprint b to be verified is matched with history fingerprint image fingerprint a, therefore
Refusal carries out fingerprint recognition to fingerprint b.That is, the big absolutely probability of fingerprint b is stayed the residual on HOME key for fingerprint a and is referred to
Stricture of vagina.
Or, obtain the history fingerprint image fingerprint a of above-mentioned user A as shown in Figure 2 E, fingerprint image fingerprint b to be verified is such as
Shown in Fig. 2 F, after all images of all images of fingerprint a and fingerprint b are overlapped, the fingerprint image c drawing superposition is as schemed
It is seen then that fingerprint image fingerprint b to be verified is relatively low with the fingerprint image c registration ratio being superimposed shown in 2G, obtained by image procossing
Fingerprint image fingerprint b to be verified and the registration being superimposed fingerprint image c are 20%, due to not less than default registration 80%, really
Fixed fingerprint image fingerprint b to be verified and history fingerprint image fingerprint a mismatches, and therefore carries out fingerprint recognition to fingerprint b.Namely
Say, the fingerprint that the big absolutely probability of fingerprint b stays for active user, not latent prints.
The fingerprint identification method that the present embodiment provides, by the coincidence based on fingerprint image to be verified and superposition fingerprint image
Degree carries out the matching analysis, effectively prevent non-user and is identified using latent prints, improves the accurate of fingerprint recognition
Property and safety.
Embodiment three
A kind of flow chart of fingerprint identification method that Fig. 3 provides for the embodiment of the present invention three, the present embodiment is in above-mentioned enforcement
It is optimized on the basis of example, there is provided determine the processing method that described fingerprint image to be verified is mated with history fingerprint image,
Specifically, described fingerprint image to be verified is overlapped with described history fingerprint image, obtains being superimposed fingerprint image;Determine institute
State fingerprint image to be verified and the registration of described history fingerprint image, if described registration exceed default registration it is determined that
Described fingerprint image to be verified is mated with history fingerprint image.
Accordingly, the method for the present embodiment includes:
S310, acquisition fingerprint image to be verified.
S320, described fingerprint image to be verified is overlapped with described history fingerprint image, obtains being superimposed fingerprint image.
S330, determine described fingerprint image to be verified and the registration of described history fingerprint image, if described registration surpasses
Cross default registration it is determined that described fingerprint image to be verified is mated with described history fingerprint image, refusal carries out fingerprint recognition.
Equally, the superposition fingerprint that registration can also directly be obtained by fingerprint image to be verified and the superposition of history fingerprint image
Image determines, default registration can be the quiescent value of system default, such as 80% or personal settings scope 0%
Arbitrary value in~100%.Specifically, after obtaining being superimposed fingerprint image, fingerprint recognition system obtains through image procossing
The registration of superposition fingerprint image.If described registration exceed default registration it is determined that described fingerprint image to be verified with
History fingerprint image mates, and refusal carries out fingerprint recognition;If described registration is not less than default registration it is determined that described treat
Checking fingerprint image and history fingerprint image mismatch, and carry out fingerprint recognition.
Further for example, the default registration of above-mentioned user A is set as the quiescent value 80% of system default.Obtain above-mentioned user A
History fingerprint image fingerprint a as shown in Figure 2 B, fingerprint image fingerprint b to be verified as shown in Figure 2 C, by all images of fingerprint a
After being overlapped with all images of fingerprint b, draw the fingerprint image of superposition as shown in Figure 2 D, intercept mid portion image conduct
Superposition fingerprint image c.The registration obtaining being superimposed fingerprint image c through image procossing is 90%, due to exceeding default registration
80%, equally can determine that fingerprint image fingerprint b to be verified is matched with history fingerprint image fingerprint a, therefore refuse to fingerprint b
Carry out fingerprint recognition.
The fingerprint identification method that the present embodiment provides, by the coincidence based on fingerprint image to be verified and history fingerprint image
Degree carries out the matching analysis, effectively prevent non-user and is identified using latent prints, improves the accurate of fingerprint recognition
Property and safety.
Example IV
A kind of flow chart of fingerprint identification method that Fig. 4 A provides for the embodiment of the present invention four, the present embodiment is in above-mentioned enforcement
It is optimized on the basis of example, there is provided described fingerprint image to be verified is overlapped with described history fingerprint image, obtains
The processing method of superposition fingerprint image, specifically, described fingerprint image to be verified and described history fingerprint image is carried out two-value
Change is processed;The binary image of the binary image of described fingerprint image to be verified and described history fingerprint image is folded
Plus, using the binary image obtaining as superposition fingerprint image.
Accordingly, the method for the present embodiment includes:
S410, acquisition fingerprint image to be verified.
S420, described fingerprint image to be verified and described history fingerprint image are carried out binary conversion treatment.
Wherein, image binaryzation is processed is exactly that the gray value of the pixel on image is set to 0 or 255, will entirely scheme
As presenting obvious black and white effect.Specifically, the fingerprint image to be verified of acquisition and the history fingerprint image extracting are ash
Degree image, gray level image is by the Lycoperdon polymorphum Vitt picture of the different colours depth between black pixel, white pixel point and monochrome pixels point
Vegetarian refreshments is formed, the tonal gradation corresponding to pixel in gray level image generally 0~255, the corresponding gray scale of black pixel
It is worth for 0, the corresponding gray value of white pixel point is 255, and it is exactly will to have the ash of 256 tonal gradations that image carries out binary conversion treatment
Degree image is converted into the image only with 0 and 255 gray values, that is, containing only black pixel and white pixel point.At the binaryzation of image
By choosing suitable threshold value, this threshold value can be the quiescent value of fingerprint recognition system acquiescence to reason, as 125 it is also possible to according to
The dynamic value of personal set, when certain point gray value is more than or equal to the pixel of threshold value, this gray value is classified as 255, no
Then pixel is classified as 0.
S430, the binary image of the binary image of described fingerprint image to be verified and described history fingerprint image is entered
Row superposition, using the binary image obtaining as described superposition fingerprint image.
Specifically, fingerprint image to be verified complete for binary conversion treatment and history fingerprint image are carried out after image procossing
Superposition, obtain is binary image.The binaryzation of image is conducive to the process further of image, makes image become simple, and
And data volume reduces.Or, first history fingerprint image is overlapped with fingerprint image to be verified, chooses centre position or appoint
The image of meaning position, as superposition fingerprint image, carries out binary conversion treatment to superposition fingerprint image, obtains binaryzation superposition fingerprint
Image, finally on the basis of based on binaryzation superposition fingerprint image, determines that fingerprint image to be verified with history fingerprint image is
No coupling.
S440, determine described fingerprint image to be verified and superposition fingerprint image registration, if described registration exceed pre-
If registration is it is determined that described fingerprint image to be verified is mated with history fingerprint image, refusal carries out fingerprint recognition.
For example, above-mentioned user A is existing carries out history fingerprint image fingerprint a and fingerprint image fingerprint b to be verified at binaryzation
Reason, threshold value is set as default value 125, when certain point gray value is more than or equal to 125 pixel, this gray value is classified as
255, otherwise pixel is classified as 0, obtain the binary image aa of history fingerprint image and the two-value of fingerprint image to be verified
Change image bb, obtain after being overlapped being superimposed binary image cc, referring specifically to Fig. 2 B-2G.Intercept fingerprint two-value to be verified again
The mid portion image of the mid portion image and superposition fingerprint binary image cc of changing image aa is compared.If exceeding default
Registration, equally can determine that fingerprint image fingerprint b to be verified is matched with history fingerprint image fingerprint a, therefore refuses to finger
Stricture of vagina b carries out fingerprint recognition.Wherein, default registration be similarly can be system default quiescent value, such as 80% or individual
Arbitrary value in 0%~100% for the scope of people's setting.
Or, history fingerprint image fingerprint a is first overlapped by above-mentioned user A with fingerprint image fingerprint b to be verified, chooses
The image of centre position or optional position, as superposition fingerprint image c, carries out binary conversion treatment to superposition fingerprint image c, obtains
To superposition fingerprint image cc, finally on the basis of based on superposition fingerprint image cc, determine that fingerprint image to be verified and history refer to
Whether print image mates.
The present embodiment provide fingerprint identification method, by by fingerprint image to be verified and history fingerprint image through two-value
Change is processed, and carries out the matching analysis to the registration of binary image, reduces the complexity of image procossing, improve fingerprint recognition
Rate.
Embodiment five
A kind of flow chart of fingerprint identification method that Fig. 5 provides for the embodiment of the present invention five, the present embodiment is in above-mentioned enforcement
It is optimized on the basis of example, there is provided determine the process side of the registration of described fingerprint image to be verified and superposition fingerprint image
Method, specifically, relatively described fingerprint image to be verified and the pixel value being superimposed on fingerprint image correspondence position;According to comparative result
Determine described fingerprint image to be verified and the registration of superposition fingerprint image.
Accordingly, the method for the present embodiment includes:
S510, acquisition fingerprint image to be verified.
S520, described fingerprint image to be verified is overlapped with described history fingerprint image, obtains being superimposed fingerprint image.
Pixel value on fingerprint image to be verified described in S530, comparison and described superposition fingerprint image correspondence position.
Specifically, the fingerprint image to be verified of acquisition and the history fingerprint image extracting are gray level image, gray level image
It is made up of the gray pixels point of the different colours depth between black pixel, white pixel point and monochrome pixels point, Lycoperdon polymorphum Vitt
Pixel value between 0~255, black pixel value is 0, and white pixel value is 255, counts fingerprint image to be verified and folded respectively
Pixel value on adding fingerprint image correspondence position, obtains registration 1 through equation below;
Or, by obtain fingerprint image to be verified and superposition each position of fingerprint image on pixel value one by one than
Right, if difference little then it is assumed that consistent.Count the percentage ratio that consistent pixel accounts for total pixel, using percentage ratio as registration 2, such as
Lower formula:
S540, the registration determining described fingerprint image to be verified according to comparative result and being superimposed fingerprint image, if described
Registration exceedes default registration it is determined that described fingerprint image to be verified is mated with history fingerprint image, and refusal carries out fingerprint
Identification.
Equally, default registration can be the quiescent value of system default, and the scope as 80% or personal settings exists
Arbitrary value in 0%~100%.If described registration is than or equal to default registration it is determined that described fingerprint to be verified
Image is mated with history fingerprint image, and refusal carries out fingerprint recognition;If described registration not less than default registration it is determined that
Described fingerprint image to be verified is mismatched with history fingerprint image, carries out fingerprint recognition.
For example, above-mentioned user A use system default default registration quiescent value, by history fingerprint image fingerprint a with treat
Checking fingerprint image fingerprint b is overlapped obtaining being superimposed fingerprint image c, counts fingerprint image fingerprint a to be verified and superposition respectively
Pixel value on fingerprint image c correspondence position, is calculated 80% through registration 2, because registration is equal to default registration
80% it is determined that described fingerprint image to be verified is mated with history fingerprint image, and refusal carries out fingerprint recognition.
The fingerprint identification method that the present embodiment provides, by relatively described fingerprint image to be verified and superposition fingerprint image pair
Answer the pixel value on position, determine that registration carries out the matching analysis according to comparative result, improve the degree of accuracy of fingerprint recognition.
Embodiment six
A kind of flow chart of fingerprint identification method that Fig. 6 provides for the embodiment of the present invention six, the present embodiment is in above-mentioned enforcement
On the basis of example, also include:Refusal also includes after carrying out fingerprint recognition:Collection intravenous data carries out hand vein recognition.
Accordingly, the method for the present embodiment includes:
S610, acquisition fingerprint image to be verified.
If it is determined that described fingerprint image to be verified is mated with history fingerprint image, then refusal carries out fingerprint recognition to S620.
S630, collection intravenous data carry out hand vein recognition.
Specifically, it is possible to make in the case that fingerprint image to be verified is very short with the time that history fingerprint image obtains
Become fingerprint image to be verified and history fingerprint image, then then further collection user's intravenous data carries out hand vein recognition.Its
In, hand vein recognition is one kind of bio-identification.A kind of mode of vein recognition system is to obtain personal vein by hand vein recognition instrument
Scattergram, extracts eigenvalue according to special alignment algorithm from vein scattergram, and another way is obtained by infrared C CD photographic head
Take the image of finger, palm, hand back vein, the digital picture of vein is stored in computer systems, realize eigenvalue storage.
When vein compares, take vein figure in real time, special to digital image zooming-out with advanced filtering, image binaryzation, refinement means
Levy, coupling is compared with being stored in main frame medium-sized vein eigenvalue using complicated matching algorithm, thus identity authentication is carried out to individual,
Confirm identity.Hand vein recognition is divided into:Refer to hand vein recognition and vena metacarpea identification.Refer to hand vein recognition, because its capacity is big, recognition speed
Hurry up, the features such as degree of accuracy is high.
Fingerprint recognition system carries out hand vein recognition to fingerprint, can gather mobile terminal front it is also possible to collection is mobile eventually
End back, during fingerprint collecting, gathers the intravenous data information of authenticatee simultaneously, preferentially finger print information is sentenced
Certainly, in the state of finger print information is judged as latent prints, continue to judge intravenous data, if last intravenous data to be verified and
The intravenous data of user's storage is consistent, then identify successfully;If the intravenous data that intravenous data to be verified is stored with user is not
Unanimously, then recognition failures, mobile terminal stores this intravenous data information, and locks mobile terminal simultaneously.
For example, if it is fingerprint b that above-mentioned user B utilizes the fingerprint that latent prints pressing obtains, fingerprint image to be verified refers to
Stricture of vagina b is mated with history fingerprint image fingerprint a image, if the match is successful, then fingerprint image fingerprint b to be verified is carried out
Hand vein recognition.Mobile phone gathers the intravenous data information at user B back, the intravenous data information phase with the user A of storage in mobile phone
Contrast.Unmatch due to, therefore recognition failures, record the intravenous data information of this user B, and lock mobile phone.
The fingerprint identification method that the present embodiment provides, carries out hand vein recognition by gathering intravenous data, further to fingerprint
Mate unsuccessful user and carry out a deep step analysis, while reducing mobile terminal identification error, contribute to recording disabled user
Intravenous data information, carry out authentication, improve the safety that user uses.
Embodiment seven
A kind of flow chart of fingerprint identification method that Fig. 7 A provides for the embodiment of the present invention seven, the present embodiment is in above-mentioned enforcement
On the basis of example, also include:Determine the time difference of described fingerprint image to be verified and the acquisition of described history fingerprint image;If institute
When stating time difference and exceeding preset duration, then triggering execution matching operation.
Accordingly, the method for the present embodiment includes:
S710, determine described fingerprint image to be verified and described history fingerprint image acquisition time difference.
Specifically, user within the extremely short time it is possible to make fingerprint image to be verified and history fingerprint image complete
Sample;And after the time exceeding certain time length, fingerprint image to be verified and history fingerprint image duplicate probability very little.Cause
This, it is necessary to obtain the time difference that fingerprint image to be verified is obtained with history fingerprint image, is analyzed to time difference.
If the described time difference of S720 exceedes preset duration, triggering execution matching operation.
Wherein, preset duration can be the quiescent value of system default, the such as arbitrary value of 10min or personal settings,
As 1min, 2min, 3min etc..When the time difference that fingerprint image to be verified is obtained with history fingerprint image exceedes preset duration,
Then trigger matching operation, verify whether fingerprint image to be verified is consistent with history fingerprint image.
S730, acquisition fingerprint image to be verified.
If it is determined that described fingerprint image to be verified is mated with history fingerprint image, then refusal carries out fingerprint recognition to S740.
For example, above-mentioned user A adopts preset duration 10min of system default, and fingerprint image fingerprint b to be verified is referred to history
The time difference that print image fingerprint a obtains is 15min, because time difference exceedes preset duration 10min, then obtain fingerprint to be verified
Finger image b, triggering determines the operation whether fingerprint image fingerprint b to be verified is mated, such as Fig. 7 B with history fingerprint image fingerprint a
Shown.
The fingerprint identification method that the present embodiment provides, by determining the acquisition of fingerprint image to be verified and history fingerprint image
Time difference, judge whether triggering execution matching operation, improve fingerprint recognition integrity.
Embodiment eight
A kind of structural representation of fingerprint identification device that Fig. 8 provides for the embodiment of the present invention eight, the present embodiment is applicable
In the situation of identification latent prints, the method can be executed by fingerprint identification device provided in an embodiment of the present invention, this device
Can be realized by the way of software and/or hardware, software form can be the authentication application in the internet virtual world, and this device can
It is integrated in any equipment needing fingerprint recognition, such as mobile terminal and fixed terminal, mobile terminal includes smart mobile phone, flat
Plate computer, PDA, intelligent watch etc., fixed terminal includes desktop computer, ATM, door lock, Time Attendance Device etc., as shown in figure 8, specifically
Including:Fingerprint image acquisition module 81 and fingerprint identification module 82.
Fingerprint image acquisition module 81 is used for obtaining fingerprint image to be verified;
Fingerprint identification module 82 is used for if it is determined that described fingerprint image to be verified is mated with history fingerprint image, then refuse
Carry out fingerprint recognition.
Fingerprint identification device described in the present embodiment is used for executing the fingerprint identification method described in the various embodiments described above, its skill
Art principle is similar with the technique effect producing, and repeats no more here.
Embodiment nine
Fig. 9 show a kind of structural representation of fingerprint identification device of the embodiment of the present invention six offer, as shown in Figure 9:
On the basis of above-described embodiment, fingerprint identification module 82 specifically includes, fingerprint image superpositing unit 91 and overlap
Degree determining unit 92;
Fingerprint image superpositing unit 91 is used for being overlapped described fingerprint image to be verified with described history fingerprint image,
Obtain being superimposed fingerprint image;
Registration determining unit 92 is used for determining described fingerprint image to be verified and the registration of described superposition fingerprint image,
If described registration exceedes default registration it is determined that described fingerprint image to be verified is mated with described history fingerprint image;Or
Person, determines described fingerprint image to be verified and the registration of described history fingerprint image, if described registration exceedes default coincidence
Degree is it is determined that described fingerprint image to be verified is mated with described history fingerprint image.
On the basis of above-described embodiment, fingerprint image superpositing unit 91 is specifically for by described fingerprint image to be verified
Carry out binary conversion treatment with described history fingerprint image;The binary image of described fingerprint image to be verified and described history are referred to
The binary image of print image is overlapped, using the binary image obtaining as described superposition fingerprint image.
On the basis of above-described embodiment, registration determining unit 92 is specifically for relatively described fingerprint image to be verified
With the pixel value on described superposition fingerprint image correspondence position;Described fingerprint image to be verified and described is determined according to comparative result
The registration of superposition fingerprint image.
On the basis of above-described embodiment, described fingerprint identification module 82 is additionally operable to:If it is determined that described fingerprint to be verified
Image is mismatched with described history fingerprint image, then carry out fingerprint recognition using described fingerprint image to be verified.
On the basis of above-described embodiment, described device also includes:Hand vein recognition module 93;
Described hand vein recognition module 93 is used for, after described fingerprint identification module 82 refusal carries out fingerprint recognition, being used for adopting
Collection intravenous data carries out hand vein recognition.
On the basis of above-described embodiment, described device also includes:DBM 94;
DBM 94 is used for obtaining the fingerprint image of use after each fingerprint recognition success, by described fingerprint image
As history fingerprint image, history of forming fingerprint image data storehouse.
On the basis of above-described embodiment, described device, also include:Time difference determining module 95;
Time difference determining module 95 is used for determining described fingerprint image to be verified and the acquisition of described history fingerprint image
Time difference;If described time difference exceedes preset duration, triggering execution matching operation.
Said apparatus can perform the method that the aforementioned all embodiments of the present invention are provided, and possesses execution said method corresponding
Functional module and beneficial effect.The not ins and outs of detailed description in the present embodiment, can be found in the aforementioned all enforcements of the present invention
The method that example is provided.
Fingerprint identification device described in the present embodiment is used for executing the fingerprint identification method described in the various embodiments described above, its skill
Art principle is similar with the technique effect producing, and repeats no more here.
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 being carried out to the present invention 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
Other Equivalent embodiments more can be included, and the scope of the present invention is determined by scope of the appended claims.