CN109670417A - Fingerprint identification method and device - Google Patents
Fingerprint identification method and device Download PDFInfo
- Publication number
- CN109670417A CN109670417A CN201811467334.4A CN201811467334A CN109670417A CN 109670417 A CN109670417 A CN 109670417A CN 201811467334 A CN201811467334 A CN 201811467334A CN 109670417 A CN109670417 A CN 109670417A
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- fingerprint image
- image
- finger type
- finger
- 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 44
- 238000005457 optimization Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 10
- 239000000284 extract Substances 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 230000006870 function Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 230000000994 depressogenic effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 239000012528 membrane Substances 0.000 description 2
- 229910044991 metal oxide Inorganic materials 0.000 description 2
- 150000004706 metal oxides Chemical class 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000003825 pressing Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000037303 wrinkles Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 229910052738 indium Inorganic materials 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 239000011505 plaster Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- 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
- G06V40/13—Sensors therefor
-
- 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
- G06V40/1365—Matching; Classification
- G06V40/1376—Matching features related to ridge properties or fingerprint texture
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Collating Specific Patterns (AREA)
- Image Input (AREA)
Abstract
This application discloses fingerprint identification methods and device.This method comprises: using the first fingerprint image of initial acquisition gain of parameter;Finger type is judged according to first fingerprint image;If it is determined that the first finger type obtains the second fingerprint image then using the acquisition parameter reset, and fingerprint recognition is carried out using second fingerprint image.This method obtains fingerprint image using corresponding acquisition parameter according to finger type, so as to improve the success rate of fingerprint recognition when finger perspires or is wet.
Description
Technical field
The present invention relates to biological identification technologies, more particularly, to fingerprint identification method and device.
Background technique
With the fast development of information technology, terminal (such as mobile phone, tablet computer) uses more more and more universal, fingerprint knowledge
Other technology is used widely in unlock, wake-up, mobile payment of terminal etc. as the standard configuration technology in terminal.Referring to
While line identification technology prevalence, in order to realize that quick release, the unlocked time of terminal are also that terminal manufacturer is directly concerned about
Problem.
However, can be wet or easy perspiration because of finger often and cause to unlock successfully in actual fingerprint application scenarios
Rate is very low, this is because, since fingerprint is lines, crestal line is protrusion when finger presses fingerprint recognition mould group, can connect
Contact the surface of fingerprint recognition mould group, and the valley line between crestal line be it is sunk, surface is not reached, in this way, knowing in fingerprint
It will not contact because of contact and two different states in other mould group and identify fingerprint.But when finger moistens, fingerprint
Surface will form water membrane, and moisture film may fill and lead up the depressed section between crestal line.In this way when finger presses fingerprint recognition mould
It is not in the contact fingerprint state alternate with not contacting when group, the first finger type identification success rate is greatly reduced.
The unlock success rate of the first finger type is low in terminal applies, has seriously affected user experience, therefore, it is desirable into
One step improves the success rate of fingerprint recognition when finger perspires or is wet.
Summary of the invention
In view of the above problems, the purpose of the present invention is to provide fingerprint identification method and devices, wherein according to finger type
Fingerprint image is obtained using corresponding acquisition parameter, to improve the success rate of fingerprint recognition.
According to the first aspect of the invention, a kind of fingerprint identification method is provided, comprising: using initial acquisition gain of parameter the
One fingerprint image;Finger type is judged according to first fingerprint image;If it is determined that the first finger type, then using again
The acquisition parameter of setting obtains the second fingerprint image, and carries out fingerprint recognition using second fingerprint image.
Preferably, the initial acquisition parameter is the acquisition parameter for the optimization of second finger type, the reset
Acquisition parameter be for the first finger type optimization acquisition parameter.
Preferably, the initial acquisition parameter and the acquisition parameter of the reset include acquisition density different from each other
And/or frequency acquisition.
Preferably, between the step of obtaining first fingerprint image and the step of judging finger type, further includes: adopt
Fingerprint recognition is carried out with first fingerprint image, if fingerprint recognition fails, further executes the step for judging finger type
Suddenly.
Preferably, fingerprint recognition is carried out using first fingerprint image and is identified using second fingerprint image
Respectively include: extract the fisrt feature information of corresponding fingerprint image;And by the fisrt feature information and default template ratio
Compared with to obtain recognition result.
Preferably, the fisrt feature information includes line shape, line number, core point, bifurcation, curvature.
Preferably, when carrying out fingerprint recognition using first fingerprint image, optimize using for second finger type
Recognizer, using second fingerprint image carry out fingerprint recognition when, using for the first finger type optimization knowledge
Other algorithm.
Preferably, the step of judgement finger type includes: to analyze first fingerprint image using disaggregated model, with
Obtain the finger type.
Preferably, the step of analyzing first fingerprint image using disaggregated model includes: to extract first fingerprint image
The second feature information of picture;And by the second feature Information application in the disaggregated model to obtain classification results.
Preferably, the second feature information includes that crestal line and valley line area ratio, crestal line width, crestal line intersect points.
Preferably, further includes: collect the fingerprint image of second finger type and the first finger type as training sample;With
And using the training sample training disaggregated model.
Preferably, the disaggregated model includes any of the following model: SVM, BP neural network, cluster algorithm.
Preferably, first finger type is wet finger type, and the second finger type is dry finger type.
According to the second aspect of the invention, a kind of fingerprint identification device is provided, comprising: acquisition unit, for acquiring first
Fingerprint image and the second fingerprint image;Configuration unit is set when acquiring first fingerprint image and second fingerprint image
Set the acquisition parameter of the acquisition unit;Taxon, for judging finger type according to first fingerprint image;And know
Other unit, for carrying out fingerprint recognition using first fingerprint image or second fingerprint image.
Preferably, the recognition unit extracts the fisrt feature information of corresponding fingerprint image, and special by described first
Reference ceases compared with default template, to obtain recognition result.
Preferably, the fisrt feature information includes line shape, line number, core point, bifurcation, curvature.
Preferably, the taxon extracts the second feature information of first fingerprint image, and by described second
Characteristic information is applied in the disaggregated model to obtain classification results.
Preferably, the second feature information includes that crestal line and valley line area ratio, crestal line width, crestal line intersect points.
Preferably, the recognition unit is when carrying out fingerprint recognition failure using first fingerprint image, described in starting
Taxon judges finger type.
Preferably, the taxon starts the configuration unit reset and adopts when being judged as the first finger type
Collect parameter, and the starting acquisition unit acquires second fingerprint image.
Preferably, first finger type is wet finger type, and the second finger type is dry finger type.
Fingerprint identification method according to an embodiment of the present invention judges finger type according to the first fingerprint image, is being judged as
When the first finger type, the second fingerprint image is obtained using the acquisition parameter of reset.This method is used according to finger type
Corresponding acquisition parameter reacquires fingerprint image, so as to improve the success of fingerprint recognition when finger perspires or is wet
Rate.
In a preferred embodiment, joined when acquiring the first fingerprint image using the acquisition for the optimization of second finger type
Number, when acquiring the second fingerprint image using the acquisition parameter for the optimization of the first finger type, to make full use of acquisition single
The hardware performance of member obtains the fingerprint image of optimization.In a further preferred embodiment, using the first fingerprint image and the
The recognizer accordingly optimized is respectively adopted in two fingerprint images when being identified.Thus, for finger classification optimization acquisition ginseng
On the basis of the fingerprint image that number obtains, further using the recognizer accordingly optimized, so as to further increase finger
The success rate of fingerprint recognition when perspiring or is wet.
In a preferred embodiment, in fingerprint recognition, fisrt feature information and default template ratio are extracted from fingerprint image
Compared with, finger classification when, analyzed in disaggregated model from fingerprint image second feature Information application.Fisrt feature information and
Second feature information is the different characteristic informations obtained from fingerprint image, thus may be respectively used for improving fingerprint recognition and hand
Refer to the accuracy of classification.
Detailed description of the invention
By referring to the drawings to the description of the embodiment of the present invention, above-mentioned and other purposes of the invention, feature and
Advantage will be apparent from, in the accompanying drawings:
Fig. 1 shows the flow chart of fingerprint identification method according to a first embodiment of the present invention;
Fig. 2 shows the flow charts of fingerprint identification method according to a second embodiment of the present invention;
Fig. 3 shows svm classifier model used in fingerprint identification method according to an embodiment of the present invention;
Fig. 4 shows the schematic block diagram of the fingerprint identification device of third embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Whole description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Fig. 1 shows the flow chart of fingerprint identification method according to a first embodiment of the present invention, in the following methods, first-hand
Finger type is wet finger type, and second finger type is dry finger type.
In step s101, using the first fingerprint image of initial acquisition gain of parameter.For example, pressing fingerprint recognition in user
The first fingerprint image is obtained during mould group.
Initial acquisition parameter is, for example, the acquisition parameter defaulted, or the acquisition parameter for the optimization of second finger type.
Acquisition parameter includes acquisition density and frequency acquisition.For example, the fingerprint clear of second finger type, can use lesser acquisition
Density and frequency acquisition reduce the whole waiting time of unlocked by fingerprint to improve the speed of fingerprint collecting.
First fingerprint image can be general image or topography.For example, it may be determined that fingerprint recognition mould group is adopted
Collect range, and above-mentioned acquisition range is divided into multiple regions, obtains in multiple region corresponding to the maximum region of signal-to-noise ratio
Fingerprint image.
First fingerprint image can also be the fingerprint image selected in the multiple fingerprint images continuously obtained.Example
Such as, the first fingerprint image can be clearest in the image of a finger or multiple fingerprint images of a finger of user
One.During acquiring fingerprint image, some fingerprint images may be caused fuzzy because finger is mobile.Therefore,
Multiple fingerprint image can be grabbed, using last width fingerprint image, or arbitrarily select a width clearly fingerprint image as the
One fingerprint image, the specific present invention is without limitation.
In step s 102, finger type is judged according to the first fingerprint image.For example, extracting feature from the first fingerprint image
Then characteristic information is applied to by disaggregated model trained in advance by information.Thus, believed according to the feature of the first fingerprint image
First fingerprint image is divided into the first finger types of image and second finger types of image by breath.
The characteristic information that can be used for finger classification in first fingerprint image includes that crestal line and valley line area ratio, crestal line are wide
Degree, crestal line intersect at least one of points.For example, the crestal line and valley line area ratio substantially 1:1 of second finger type, the
The fingerprint surface of one finger type will form water membrane, and moisture film may fill and lead up the depressed section between crestal line, as a result crestal line
It may change with valley line area ratio as greater than 2:1.
Disaggregated model includes SVM, BP neural network, clustering etc..
SVM refers to support vector machines, is a kind of common method of discrimination.SVM method is by a Nonlinear Mapping
Sample space is mapped in a higher-dimension or even infinite dimensional feature space (space Hilbert), so that in original sample by p
The problem of Nonlinear separability, is converted into the problem of linear separability in feature space in this space.SVM method application kernel function
Expansion theorem, there is no need to know the explicit expression of Nonlinear Mapping;Due to being established linearly in high-dimensional feature space
Learning machine, so not only hardly increasing the complexity of calculating, and avoid to a certain extent compared with linear model
" dimension disaster ".
BP neural network is a kind of multilayer feedforward neural network according to the training of error backpropagation algorithm, and algorithm is known as
BP algorithm, its basic thought are gradient descent methods, using gradient search technology, to make the real output value and expectation of network
The error mean square difference of output valve is minimum.Trained neural network can voluntarily be handled the input information of similar sample
The smallest information by non-linear conversion of output error.
Clustering refers to that the set by physics or abstract object is grouped into the analysis for the multiple classes being made of similar object
Process.The target of clustering is exactly that data are collected on the basis of similar to classify.Clustering technique is measured between different data sources
Similitude, and data source is categorized into different clusters.
In step s 103, different subsequent steps is selected according to fingerprint pattern.If it is determined that the first fingerprint image is the
Two finger types of image, then follow the steps S104.If it is determined that the first fingerprint image is the first finger types of image, then step is executed
Rapid S105.
In step S104, for second finger type (e.g. dry finger type), carried out using the first fingerprint image
Fingerprint recognition.For example, characteristic information is extracted from the first fingerprint image, then by the first fingerprint image characteristics information and default template
It compares, obtains recognition result, fingerprint recognition terminates.
Can be used in first fingerprint image identify fisrt feature information include line shape, line number, core point, bifurcation,
At least one of curvature.
Fingerprint image is by wrinkle ridge and line Gu Zucheng, and wrinkle ridge is also known as lines, crestal line, the lug boss of corresponding finger skin
Point, the deeper thick lines of gray scale are rendered as in fingerprint image;Line paddy, that is, valley line then corresponds to the part that finger skin is fallen in, folder
Between two crestal lines, the gray scale of opposite crestal line is brighter.Line shape is the configuration of fingerprint, for example, be divided into left dustpan type, right dustpan type,
Bucket type, arch and Gothic arch.Line number refers to the quantity of fingerprint lines in Mode Areas, also the lines number between some algorithm two o'clocks
As the lines number between fingerprint characteristic, such as two nodes.Core point is located at the progressive center of fingerprint lines, it is typically used as
Reference point when reading fingerprint and comparing fingerprint.Bifurcation is the fine feature in lines, such as a lines is divided into two herein
Item or a plurality of lines.Curvature is the fine feature in lines, indicates the speed that ridge orientation changes.The characteristic information of fingerprint image
Including but not limited to features described above.
Default template is several fingerprint templates being stored in template database in advance, and the fingerprint template quantity prestored is extremely
Less it is one, and includes at least at least one fingerprint template of same finger in the fingerprint template prestored.For example, can pre-deposit
The fingerprint template of the fingerprint template of ten fingers of same people, same finger can only be stored in one, can also be stored in multiple.It is stored in certain
The recognition success rate that fingerprint matching identification is carried out using the finger can be improved in multiple fingerprint templates of one finger.
Fingerprinting step based on the first fingerprint image includes by the characteristic information and template data of the first fingerprint image
The template fingerprint information stored in library is compared.Using existing recognizer, according to the fisrt feature of the first fingerprint image
Information judges whether to match with the similarity size of default template.Similarity refers to can be with the ginseng of similarity degree between representative image
Number, including at least one of Feature Points Matching number and histogram distribution similarity, but be not limited to Feature Points Matching number and
Histogram distribution similarity.For example, characteristic matching number threshold value and histogram distribution similarity threshold can be set, if be not achieved
Respective threshold then judges to mismatch.If there is multiple template finger print information in template database, by the first finger print information successively with
These template fingerprint information are compared.
In step s105, it for the first finger type (the first finger type is, for example, wet finger type), resets
Acquisition parameter.The acquisition parameter of reset is, for example, to be directed to the acquisition parameter of the first finger type optimization.
As described above, acquisition parameter includes acquisition density and frequency acquisition.For example, the fingerprint fuzzy of the first finger type,
Higher acquisition density and frequency acquisition can be used, to improve the resolution ratio and clarity of fingerprint image.Although fingerprint collecting
Speed reduction cause unlocked time to extend, but the success rate of fingerprint recognition can be improved.
Preferably, further adjustment is according to the type adjustment sensor parameters for identifying mould group, to improve the clear of fingerprint image
Clear master.For example, optical finger print identification mould group is the refraction and principle of reflection using light, light is from bottom directive prism, and through rib
Mirror goes out, the light angle reflected on the rough strain line of finger surface fingerprint of injection and the bright and dark light reflected back
It will be different.CMOS (Complementary Metal Oxide Semiconductor, complementary metal oxide semiconductor)
Or the optical device of CCD (Charge Coupled Device, charge-coupled device) will be collected into different bright-dark degrees
Pictorial information just completes the acquisition of fingerprint.Capacitance type fingerprint identifies that mould group is to be integrated with thousands of semiconductor devices at one piece
" plate " on, finger plaster on it with which constitute the another sides of capacitor, it is at salient point and recessed since finger flat is rough and uneven in surface
The actual range size that plate is contacted at point is just different, and the capacitance values of formation are also just different, and equipment is around this principle
Collected different numerical value is summarized, the acquisition of fingerprint is also just completed.
The sensitivity that optical device is improved in optical finger print identification mould group changes electricity in capacitance type fingerprint identification mould group
Hold threshold value, the clarity and contrast of fingerprint image can be improved, obtain more characteristic informations to improve recognition success rate.
In step s 106, the second fingerprint image is obtained using the acquisition parameter of reset.For example, referring in user's pressing
The second fingerprint image is obtained during line identification mould group.
Second fingerprint image can be general image or topography.For example, it may be determined that fingerprint recognition mould group
Acquisition range, and above-mentioned acquisition range is divided into multiple regions, it is right to obtain the maximum region institute of signal-to-noise ratio in multiple region
The fingerprint image answered.
Second fingerprint image can also be the fingerprint image selected in the multiple fingerprint images continuously obtained.Example
Such as, the second fingerprint image can be clearest in the image of a finger or multiple fingerprint images of a finger of user
One.During acquiring fingerprint image, some fingerprint images may be caused fuzzy because finger is mobile.Therefore,
Multiple fingerprint image can be grabbed, using last width fingerprint image, or arbitrarily select a width clearly fingerprint image as the
Two fingerprint images, the specific present invention is without limitation.
In step s 107, fingerprint recognition is carried out using the second fingerprint image.For example, extracting first from the second fingerprint image
Then characteristic information compares fisrt feature information with default template, obtain fingerprint recognition as a result, fingerprint recognition terminates.
Can be used in second fingerprint image identify fisrt feature information and the first fingerprint image for identification first
Characteristic information is essentially identical, including at least one of line shape, line number, core point, bifurcation, curvature.
It is big according to the fisrt feature information of the second fingerprint image and the similarity of default template using existing recognizer
It is small to judge whether to match, to obtain fingerprint recognition result.
Preferably, the second fingerprint can be further increased using the algorithm for recognizing fingerprint for the optimization of the first finger type
The success rate of image recognition.For example, reducing phase in recognizer according to the feature that the fingerprint image of the first finger type obscures
Like degree threshold value, fingerprint recognition success rate can be improved.
In step S108, fingerprint recognition terminates.
The recognition result includes using the success of the first Fingerprint recognition, using the failure of the first Fingerprint recognition, use
Second Fingerprint recognition succeeds, using any one of the second Fingerprint recognition failure.
Fingerprint identification method according to first embodiment judges finger type according to the first fingerprint image, is being judged as
When one finger type, resets acquisition parameter and obtain the second fingerprint image.This method is according to finger type using reset
Acquisition parameter reacquire fingerprint image, so as to improve finger perspire or it is wet when fingerprint recognition success rate.
Fig. 2 shows the flow charts of fingerprint identification method according to a second embodiment of the present invention.Second embodiment of the invention
Fingerprint identification method and first embodiment are the difference is that carry out pre-identification using the first fingerprint image.Phase is sampled in figure
Same appended drawing reference indicates the same steps of the two, only describes the step related to pre-identification below.
In step s101, using the first fingerprint image of initial acquisition gain of parameter.Then, step S111 is executed to first
Fingerprint image carries out pre-identification.
For example, fisrt feature information is extracted from the first fingerprint image, then by fisrt feature information compared with default template
Compared with acquisition fingerprint recognition is as a result, fingerprint recognition terminates.
Can be used in first fingerprint image identify fisrt feature information, including line shape, line number, core point, bifurcation,
At least one of curvature.
It is similar to default template according to the fisrt feature information of the first fingerprint image using existing algorithm for recognizing fingerprint
Degree size judges whether to match, to obtain fingerprint recognition result.
In step S112, different subsequent steps is selected according to the recognition result of the first fingerprint image.If first refers to
The identification success of print image, thens follow the steps S108, terminates fingerprint recognition.If the recognition failures of the first fingerprint image, hold
Row step S102.
In step s 102, finger type is judged according to the first fingerprint image.For example, extracting feature from the first fingerprint image
Then characteristic information is applied to by disaggregated model trained in advance by information.Thus, believed according to the feature of the first fingerprint image
Fingerprint image is divided into the first finger types of image and second finger types of image by breath.
In step s 103, different subsequent steps is selected according to fingerprint pattern.If it is determined that the first fingerprint image is the
Two finger types of image, then follow the steps S108, and fingerprint recognition terminates.If it is determined that the first fingerprint image is the first finger type
Image thens follow the steps S105.
Subsequent step S105-S107 according to the second embodiment is identical as the corresponding steps of first embodiment, herein no longer
It is described in detail.
Fingerprint identification method according to the second embodiment, using the first fingerprint image carry out pre-identification, if pre-identification at
Function then terminates fingerprint recognition, if pre-identification fails, further judges finger type according to the first fingerprint image, is being judged as
When the first finger type, the second fingerprint image is obtained using the acquisition parameter of reset.This method can reduce unnecessary
Finger type deterministic process, to improve recognition speed.Further, corresponding acquisition parameter is used again according to finger type
Fingerprint image is obtained, so as to improve the success rate of fingerprint recognition when finger perspires or is wet.
Fig. 3 shows svm classifier model used in fingerprint identification method according to an embodiment of the present invention.
In svm classifier model, only there are two different classification results, the first finger type fingerprint or second finger types
Fingerprint.The crestal line and valley line area ratio of first finger type and second finger type, crestal line width, that crestal line intersects points etc. is special
Sign is different, therefore can judge that it belongs to second finger type by carrying out tagsort to the finger print test set of input
Fingerprint still falls within the first finger type fingerprint.
The training of svm classifier model: selecting suitable kernel function, by second finger type training set and the first finger type
Training set is mapped to high-dimensional feature space.Second finger type feature sample and are found out in sample characteristics space using SVM
The optimal separating hyper plane of one finger type feature sample obtains representing second finger type and the first finger type sample characteristics
Supporting vector collection and its corresponding VC confidence level, formed and judge the discriminant function of each characteristic type.
The use of svm classifier model: after obtaining the first fingerprint image, using svm classifier model to the first fingerprint image
As carrying out classification judgement, the characteristic information of the first fingerprint image to be sorted is mapped to feature space by kernel function effect
In, as the input of discriminant function, classification results are obtained using classification decision function, i.e. fingerprint image is that second finger type refers to
Print image or the first finger type fingerprint image.
Fig. 4 goes out the schematic block diagram of the fingerprint identification device of third embodiment of the invention.
Fingerprint identification device 100 includes acquisition unit 101, configuration unit 102, recognition unit 103 and taxon 105.
Acquisition unit 101 is for acquiring the first fingerprint image and the second fingerprint image.
It is single that the acquisition is arranged when acquiring first fingerprint image and second fingerprint image for configuration unit 102
The acquisition parameter of member 101.
Recognition unit 103 is connected with acquisition unit 101, for using first fingerprint image or second fingerprint
Image carries out fingerprint recognition.Recognition unit 103 extracts the fisrt feature information of corresponding fingerprint image, and special by described first
Reference ceases compared with default template, to obtain recognition result.Fisrt feature information include line shape, line number, core point, bifurcation,
Any one in curvature.
Taxon 105 is connected to obtain the first fingerprint image with acquisition unit 101, is connected with recognition unit 103
To obtain the recognition result of the first fingerprint image.For fingerprint identification method according to first embodiment, taxon 105 is being obtained
After obtaining the first fingerprint image, finger type is further judged according to the first fingerprint image.For finger according to the second embodiment
Line recognition methods, taxon 105 is after recognition unit 103 is to the recognition failures of the first fingerprint image, further according to
One fingerprint image judges finger type.
Taxon 105 extracts the second feature information of first fingerprint image, and by the second feature information
Applied in the disaggregated model to obtain classification results.Second feature information includes that crestal line and valley line area ratio, crestal line are wide
Degree, crestal line intersect any one in points.
Taxon 105 is connected with configuration unit 102, to start configuration unit 102 after judging finger type
Acquisition parameter for the optimization of wet mobile phone is provided.
It is as described above according to the embodiment of the present invention, these embodiments details all there is no detailed descriptionthe, also not
Limiting the invention is only the specific embodiment.Obviously, as described above, can make many modifications and variations.This explanation
These embodiments are chosen and specifically described to book, is principle and practical application in order to better explain the present invention, thus belonging to making
Technical field technical staff can be used using modification of the invention and on the basis of the present invention well.The present invention is only by right
The limitation of claim and its full scope and equivalent.
Claims (21)
1. a kind of fingerprint identification method characterized by comprising
Using the first fingerprint image of initial acquisition gain of parameter;
Finger type is judged according to first fingerprint image;
If it is determined that the first finger type obtains the second fingerprint image then using the acquisition parameter reset, and use
Second fingerprint image carries out fingerprint recognition.
2. fingerprint identification method according to claim 1, which is characterized in that the initial acquisition parameter is for second-hand
Refer to that the acquisition parameter of type optimization, the acquisition parameter of the reset are the acquisition parameters for the optimization of the first finger type.
3. fingerprint identification method according to claim 2, which is characterized in that the initial acquisition parameter and described set again
Fixed acquisition parameter includes acquisition density and/or frequency acquisition different from each other.
4. fingerprint identification method according to claim 1, which is characterized in that in the step of obtaining first fingerprint image
Between the step of judging finger type, further includes:
Fingerprint recognition is carried out using first fingerprint image, if fingerprint recognition fails, further executes and judges finger class
The step of type.
5. according to the method described in claim 4, it is characterized in that, carrying out fingerprint recognition using first fingerprint image and adopting
It is identified with second fingerprint image respectively include:
Extract the fisrt feature information of corresponding fingerprint image;And
By the fisrt feature information compared with default template, to obtain recognition result.
6. fingerprint identification method according to claim 5, which is characterized in that the fisrt feature information includes line shape, line
At least one of number, core point, bifurcation, curvature.
7. fingerprint identification method according to claim 5, which is characterized in that referred to using first fingerprint image
When line identifies, using the recognizer for the optimization of second finger type, fingerprint knowledge is being carried out using second fingerprint image
When other, using the recognizer for the optimization of the first finger type.
8. fingerprint identification method according to claim 1, which is characterized in that the step of judgement finger type includes:
First fingerprint image is analyzed using disaggregated model, to obtain the finger type.
9. fingerprint identification method according to claim 8, which is characterized in that analyze first fingerprint using disaggregated model
The step of image includes:
Extract the second feature information of first fingerprint image;And
By the second feature Information application to obtain classification results in the disaggregated model.
10. fingerprint identification method according to claim 9, which is characterized in that the second feature information include crestal line and
Valley line area ratio, crestal line width, crestal line intersect at least one of points.
11. fingerprint identification method according to claim 8, which is characterized in that further include:
The fingerprint image of second finger type and the first finger type is collected as training sample;And use the training sample
The training disaggregated model.
12. fingerprint identification method according to claim 8, which is characterized in that the disaggregated model includes following any one
Kind model: SVM, BP neural network, cluster algorithm.
13. fingerprint identification method according to claim 1, which is characterized in that first finger type is wet finger class
Type, the second finger type are dry finger type.
14. a kind of fingerprint identification device characterized by comprising
Acquisition unit, for acquiring the first fingerprint image and the second fingerprint image;
The acquisition of the acquisition unit is arranged when acquiring first fingerprint image and second fingerprint image for configuration unit
Parameter;
Taxon, for judging finger type according to first fingerprint image;And
Recognition unit, for carrying out fingerprint recognition using first fingerprint image or second fingerprint image.
15. fingerprint identification device according to claim 14, which is characterized in that the recognition unit extracts corresponding fingerprint
The fisrt feature information of image, and by the fisrt feature information compared with default template, to obtain recognition result.
16. fingerprint identification device according to claim 15, which is characterized in that the fisrt feature information include line shape,
At least one of line number, core point, bifurcation, curvature.
17. fingerprint identification device according to claim 14, which is characterized in that the taxon is extracted described first and referred to
The second feature information of print image, and the second feature Information application is obtained into classification knot in the disaggregated model
Fruit.
18. fingerprint identification device according to claim 17, which is characterized in that the second feature information include crestal line and
Valley line area ratio, crestal line width, crestal line intersect at least one of points.
19. fingerprint identification device according to claim 14, which is characterized in that the recognition unit is using described first
When fingerprint image carries out fingerprint recognition failure, starts the taxon and judge finger type.
20. fingerprint identification device according to claim 19, which is characterized in that the taxon is being judged as first-hand
When referring to type, starts the configuration unit and reset acquisition parameter, and the starting acquisition unit acquisition described second refers to
Print image.
21. fingerprint identification device according to claim 14, which is characterized in that first finger type is wet finger class
Type, the second finger type are dry finger type.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811467334.4A CN109670417A (en) | 2018-12-03 | 2018-12-03 | Fingerprint identification method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811467334.4A CN109670417A (en) | 2018-12-03 | 2018-12-03 | Fingerprint identification method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109670417A true CN109670417A (en) | 2019-04-23 |
Family
ID=66145002
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811467334.4A Pending CN109670417A (en) | 2018-12-03 | 2018-12-03 | Fingerprint identification method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109670417A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112989888A (en) * | 2019-12-17 | 2021-06-18 | 华为技术有限公司 | Fingerprint anti-counterfeiting method and electronic equipment |
US11450141B2 (en) | 2020-09-09 | 2022-09-20 | Egis Technology Inc. | Electronic device with fingerprint sensing function and fingerprint comparison method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106022067A (en) * | 2016-05-30 | 2016-10-12 | 广东欧珀移动通信有限公司 | Unlocking control method and terminal equipment |
CN106096354A (en) * | 2016-05-27 | 2016-11-09 | 广东欧珀移动通信有限公司 | A kind of unlocked by fingerprint method and terminal |
CN107451444A (en) * | 2017-07-17 | 2017-12-08 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
-
2018
- 2018-12-03 CN CN201811467334.4A patent/CN109670417A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106096354A (en) * | 2016-05-27 | 2016-11-09 | 广东欧珀移动通信有限公司 | A kind of unlocked by fingerprint method and terminal |
CN106022067A (en) * | 2016-05-30 | 2016-10-12 | 广东欧珀移动通信有限公司 | Unlocking control method and terminal equipment |
CN107451444A (en) * | 2017-07-17 | 2017-12-08 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112989888A (en) * | 2019-12-17 | 2021-06-18 | 华为技术有限公司 | Fingerprint anti-counterfeiting method and electronic equipment |
WO2021121112A1 (en) * | 2019-12-17 | 2021-06-24 | 华为技术有限公司 | Fingerprint anti-counterfeiting method and electronic device |
US11881011B2 (en) | 2019-12-17 | 2024-01-23 | Huawei Technologies Co., Ltd. | Fingerprint anti-counterfeiting method and electronic device |
CN112989888B (en) * | 2019-12-17 | 2024-06-07 | 华为技术有限公司 | Fingerprint anti-counterfeiting method and electronic equipment |
US11450141B2 (en) | 2020-09-09 | 2022-09-20 | Egis Technology Inc. | Electronic device with fingerprint sensing function and fingerprint comparison method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109063565B (en) | Low-resolution face recognition method and device | |
Tabassi et al. | A novel approach to fingerprint image quality | |
WO2021238455A1 (en) | Data processing method and device, and computer-readable storage medium | |
Peng et al. | A hybrid convolutional neural network for intelligent wear particle classification | |
Hasan et al. | Fingerprint image enhancement and recognition algorithms: a survey | |
CN110781829A (en) | Light-weight deep learning intelligent business hall face recognition method | |
Liu et al. | One-class fingerprint presentation attack detection using auto-encoder network | |
CN101777131A (en) | Method and device for identifying human face through double models | |
CN109284675A (en) | A kind of recognition methods of user, device and equipment | |
CN104346503A (en) | Human face image based emotional health monitoring method and mobile phone | |
CN111954250B (en) | Lightweight Wi-Fi behavior sensing method and system | |
Sudiro et al. | Adaptable fingerprint minutiae extraction algorithm based-on crossing number method for hardware implementation using FPGA device | |
CN113312989A (en) | Finger vein feature extraction network based on aggregation descriptor and attention | |
Verma et al. | Contactless palmprint verification system using 2-D gabor filter and principal component analysis. | |
CN101819629A (en) | Supervising tensor manifold learning-based palmprint identification system and method | |
CN109670417A (en) | Fingerprint identification method and device | |
Khodadoust et al. | Fingerprint indexing for wrinkled fingertips immersed in liquids | |
CN110263726B (en) | Finger vein identification method and device based on deep correlation feature learning | |
Christensen et al. | Empirical evaluation methods in computer vision | |
Evangelin et al. | Biometric authentication of physical characteristics recognition using artificial neural network with PSO algorithm | |
Ibrahim et al. | A hyprid technique for human footprint recognition | |
Kamarajugadda et al. | Stride towards aging problem in face recognition by applying hybrid local feature descriptors | |
Safavipour et al. | A hybrid approach for multimodal biometric recognition based on feature level fusion in reproducing kernel Hilbert space | |
Jaiswal et al. | Brief description of image based 3D face recognition methods | |
Cenys et al. | Genetic algorithm based palm recognition method for biometric authentication systems |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190423 |