CN109863505A - Method, processor and the electronic equipment of fingerprint recognition - Google Patents
Method, processor and the electronic equipment of fingerprint recognition Download PDFInfo
- Publication number
- CN109863505A CN109863505A CN201980000111.3A CN201980000111A CN109863505A CN 109863505 A CN109863505 A CN 109863505A CN 201980000111 A CN201980000111 A CN 201980000111A CN 109863505 A CN109863505 A CN 109863505A
- Authority
- CN
- China
- Prior art keywords
- fingerprint image
- fingerprint
- feature data
- fisrt feature
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Collating Specific Patterns (AREA)
- Image Input (AREA)
Abstract
The embodiment of the present application provides a kind of method of fingerprint recognition, can be improved the efficiency of fingerprint recognition.This method comprises: primary processor receives the fisrt feature data for the fingerprint image that the coprocessor that coprocessor is sent handles the fingerprint image of finger;The primary processor carries out fingerprint recognition according to the fisrt feature data of the fingerprint image.
Description
Technical field
This application involves information technology fields, and a kind of method more particularly, to fingerprint recognition, processor and electricity
Sub- equipment.
Background technique
Fingerprint identification technology is the fingerprint image that finger is acquired by fingerprint sensor, and passes through central processing unit
(Central Processing Unit, CPU) carries out fingerprint recognition according to the fingerprint image.Due to fingerprint image acquisition process
In the interference that is subject to it is more, and the data volume carried in fingerprint image is huge, so that algorithm when CPU handles fingerprint image
Complexity increases, the collected fingerprint image especially under the cunning scenes such as low temperature, the more dry and strong light direct beam of finger, and CPU pairs
The treatment effeciency of fingerprint image is decreased obviously, it is difficult to meet the needs of fingerprint recognition.
Summary of the invention
The embodiment of the present application provides method, processor and the electronic equipment of a kind of fingerprint recognition, can be improved fingerprint recognition
Efficiency.
In a first aspect, providing a kind of method of fingerprint recognition, comprising: primary processor receives the described of coprocessor transmission
The fisrt feature data for the fingerprint image that coprocessor handles the fingerprint image of finger;The primary processor
According to the fisrt feature data of the fingerprint image, fingerprint recognition is carried out.
In one possible implementation, the method also includes: at coprocessor is to the fingerprint image
During reason, primary processor is concurrently handled the fingerprint image, obtains the second feature number of the fingerprint image
According to, and fingerprint recognition is carried out according to the second feature data of the fingerprint image;
Wherein, the primary processor carries out fingerprint recognition according to the fisrt feature data of the fingerprint image, comprising: if
Fingerprint recognition failure is carried out according to the second feature data of the fingerprint image, the primary processor is according to the fingerprint image
Fisrt feature data carry out fingerprint recognition.
In one possible implementation, the method also includes: if according to the fisrt feature number of the fingerprint image
According to fingerprint recognition failure is carried out, the primary processor receives another fingerprint image for the finger that the coprocessor is sent
Fisrt feature data;The primary processor carries out fingerprint recognition according to the fisrt feature data of another fingerprint image.
In one possible implementation, the method also includes: the coprocessor to the fingerprint image into
During row processing, the primary processor handles the fingerprint image, obtains the second feature of the fingerprint image
Data;
Wherein, the primary processor carries out fingerprint recognition, comprising: institute according to the fisrt feature data of the fingerprint image
Primary processor is stated according to the fisrt feature data of the fingerprint image and the second feature data of the fingerprint image, carries out fingerprint
Identification.
In one possible implementation, the second feature data of the fingerprint image include the fingerprint of coarse extraction
The characteristic of image, the fisrt feature data of the fingerprint image include the characteristic of the fingerprint image carefully extracted.
In one possible implementation, it is connected between the primary processor and the coprocessor by api interface.
In one possible implementation, the primary processor is CPU, and the coprocessor is DSP.
Second aspect provides a kind of method of fingerprint recognition, comprising: coprocessor to the fingerprint image of finger at
Reason, obtains the fisrt feature data of the fingerprint image;The coprocessor sends the of the fingerprint image to primary processor
One characteristic, wherein the fisrt feature data of the fingerprint image carry out fingerprint recognition for the primary processor.
In one possible implementation, the coprocessor handles the fingerprint image of finger, obtains described
The fisrt feature data of fingerprint image, comprising: the fingerprint image is handled in the primary processor to obtain the fingerprint
The second feature data of image, and according to the second feature data of the fingerprint image carry out fingerprint recognition during, it is described
Coprocessor is concurrently handled the fingerprint image, obtains the fisrt feature data of the fingerprint image.
Wherein, the fisrt feature data of the fingerprint image are for the primary processor according to the of the fingerprint image
Two characteristics carry out carrying out fingerprint recognition when fingerprint recognition failure.
In one possible implementation, the method also includes: if the primary processor is according to the fingerprint image
Fisrt feature data carry out fingerprint recognition failure, the coprocessor handles another fingerprint image of the finger,
Obtain the fisrt feature data of another frame image;The data processor sends another fingerprint to the primary processor
The fisrt feature data of the fisrt feature data of image, another fingerprint image carry out fingerprint knowledge for the primary processor
Not.
In one possible implementation, the coprocessor handles the fingerprint image of finger, obtains described
The fisrt feature data of fingerprint image, comprising: the fingerprint image is handled in the primary processor to obtain the fingerprint
During the second feature data of image, the coprocessor is concurrently handled the fingerprint image, is obtained described
The fisrt feature data of fingerprint image.
Wherein, the fisrt feature data of the fingerprint image and the second feature data of the fingerprint image are provided commonly for institute
It states primary processor and carries out fingerprint recognition mistake.
In one possible implementation, the second feature data of the fingerprint image include the fingerprint of coarse extraction
The characteristic of image, the fisrt feature data of the fingerprint image include the characteristic of the fingerprint image carefully extracted.
In one possible implementation, pass through application programming interfaces between the primary processor and the coprocessor
API connection.
In one possible implementation, the primary processor is CPU, and the coprocessor is DSP.
The third aspect provides a kind of primary processor for fingerprint recognition, comprising:
Communication unit, the coprocessor for receiving coprocessor transmission handle to the fingerprint image of finger
The fisrt feature data of the fingerprint image arrived;
Fingerprint identification unit carries out fingerprint recognition for the fisrt feature data according to the fingerprint image.
In one possible implementation, the primary processor further includes processing unit, and the processing unit is used for:
Coprocessor in the process of processing, concurrently handles the fingerprint image fingerprint image, obtains described
The second feature data of fingerprint image;The fingerprint identification unit is also used to: according to the second feature data of the fingerprint image
Carry out fingerprint recognition;If fingerprint recognition failure is carried out according to the second feature data of the fingerprint image, according to the fingerprint
The fisrt feature data of image carry out fingerprint recognition.
In one possible implementation, the communication unit is also used to, and receives the described of the coprocessor transmission
The fisrt feature data of another fingerprint image of finger;The fingerprint identification unit is also used to, according to another fingerprint image
Fisrt feature data, carry out fingerprint recognition.
In one possible implementation, the primary processor further includes processing unit, and the processing unit is used for:
The coprocessor in the process of processing, handles the fingerprint image fingerprint image, obtains the finger
The second feature data of print image;Wherein, the fingerprint identification unit is specifically used for: the primary processor is according to the fingerprint image
The fisrt feature data of picture and the second feature data of the fingerprint image carry out fingerprint recognition.
In one possible implementation, the second feature data of the fingerprint image include the fingerprint of coarse extraction
The characteristic of image, the fisrt feature data of the fingerprint image include the characteristic of the fingerprint image carefully extracted.
In one possible implementation, it is connected between the primary processor and the coprocessor by api interface.
In one possible implementation, the primary processor is CPU, and the coprocessor is DSP.
Fourth aspect provides a kind of coprocessor for fingerprint recognition, comprising:
Processing unit is handled for the fingerprint image to finger, obtains the fisrt feature data of the fingerprint image;
Communication unit, for sending the fisrt feature data of the fingerprint image to the primary processor, wherein the finger
The fisrt feature data of print image carry out fingerprint recognition for the primary processor.
In one possible implementation, the processing unit is specifically used for: in the primary processor to the fingerprint
Image is handled to obtain the second feature data of the fingerprint image, and according to the second feature data of the fingerprint image into
During row fingerprint recognition, concurrently the fingerprint image is handled, obtains the fisrt feature number of the fingerprint image
According to.Wherein, the fisrt feature data of the fingerprint image are for the primary processor in the second spy according to the fingerprint image
Sign data carry out carrying out fingerprint recognition when fingerprint recognition failure.
In one possible implementation, the processing unit is also used to: if the primary processor is according to the fingerprint
The fisrt feature data of image carry out fingerprint recognition failure, then handle another fingerprint image of the finger, obtain institute
State the fisrt feature data of another frame image;
The communication unit is also used to: Xiang Suoshu primary processor sends the fisrt feature data of another fingerprint image,
The fisrt feature data of another fingerprint image carry out fingerprint recognition for the primary processor.
In one possible implementation, the second feature data of the fingerprint image include the fingerprint of coarse extraction
The characteristic of image, the fisrt feature data of the fingerprint image include the characteristic of the fingerprint image carefully extracted.
In one possible implementation, pass through application programming interfaces between the primary processor and the coprocessor
API connection.
In one possible implementation, the primary processor is CPU, and the coprocessor is DSP.
5th aspect, provides a kind of chip, for realizing above-mentioned first aspect or any possible reality of first aspect
Method in existing mode.Specifically, which includes primary processor, for computer program to be called and run from memory,
So that the equipment for being equipped with the chip is executed such as the side in any possible implementation of above-mentioned first aspect or first aspect
Method.
6th aspect, provides a kind of chip, for realizing above-mentioned second aspect or any possible reality of second aspect
Method in existing mode.Specifically, which includes coprocessor, for computer program to be called and run from memory,
So that the equipment for being equipped with the chip is executed such as the side in any possible implementation of above-mentioned first aspect or first aspect
Method.
7th aspect, provides a kind of computer readable storage medium, for storing computer program, the computer program
So that computer executes the method in any possible implementation of above-mentioned first aspect or first aspect.
Eighth aspect provides a kind of computer readable storage medium, for storing computer program, the computer program
So that computer executes the method in any possible implementation of above-mentioned second aspect or second aspect.
9th aspect, provides a kind of electronic equipment, any possible realization side including the third aspect or the third aspect
Coprocessor in any possible implementation of primary processor and fourth aspect or fourth aspect in formula.
In one possible implementation, the electronic equipment further includes fingerprint sensor, and the fingerprint sensor is used
In acquisition fingerprint image.
Based on the above-mentioned technical proposal, due to increasing coprocessor for carrying out the processing of fingerprint image, so that main process task
Device and coprocessor can cooperate with completion fingerprint recognition, improve the efficiency of fingerprint recognition.
Detailed description of the invention
Fig. 1 is the process interaction figure of the fingerprint identification method of the embodiment of the present application.
Fig. 2 is the process interaction figure of the fingerprint identification method of the embodiment of the present application.
Fig. 3 is that the CPU and DSP of the embodiment of the present application cooperate with the schematic block diagram handled fingerprint image.
Fig. 4 is the schematic block diagram of CPU and DSP interior processing unit.
Fig. 5 is CPU and DSP respectively to the schematic diagram of the processing mode of a frame fingerprint image.
Fig. 6 is the schematic diagram that CPU and DSP respectively execute data processing.
Fig. 7 is the process interaction figure of the fingerprint identification method of the embodiment of the present application.
Fig. 8 is a kind of process interaction figure of possible fingerprint identification method of the application.
Fig. 9 is the schematic flow chart that CPU is individually handled fingerprint image.
Figure 10 is that CPU and DSP cooperates with the schematic flow chart handled fingerprint.
Figure 11 is the schematic block diagram of the CPU for fingerprint recognition of the embodiment of the present application.
Figure 12 is the schematic block diagram of the DSP for fingerprint recognition of the embodiment of the present application.
Specific embodiment
Below in conjunction with attached drawing, technical solutions in the embodiments of the present application is described.
The embodiment of the present application is only illustrated by taking optical fingerprint systems as an example, but should not be constituted to the embodiment of the present application any
It limits, the embodiment of the present application is equally applicable to other fingerprint recognition systems such as capacitance type fingerprint system, ultrasonic fingerprint system
Deng as long as can be realized the acquisition of fingerprint image.
In optical finger print identification process, the light that fingerprint sensor (or being imaging sensor etc.) acquisition light source issues enters
The optical signal for being incident upon finger and being reflected by finger, to obtain the fingerprint image of the finger.The hand is carried in the fingerprint image
The finger print information of finger handle and match with pre-stored fingerprint template to the fingerprint image, it is possible thereby to realize
For the fingerprint recognition of the finger.
In general, CPU carries out fingerprint recognition according to the fingerprint image, for example, carrying out image preprocessing, figure to the fingerprint image
Image intensifying, image characteristics extraction and characteristic matching etc. finally determine whether the fingerprint is that user realizes the fingerprint for authenticating and storing.
The fingerprint algorithm that CPU is used is scalar operation, and is single thread operation, bit wide be usually 32 bits (bit) or
64bit, therefore the mass data in fingerprint image, processing speed are very slow.Especially in low temperature, finger is more dry and Qiang Guangzhi
The collected fingerprint image under cunning scene such as penetrate, CPU is decreased obviously the treatment effeciency of fingerprint image, and CPU is to fingerprint image
Processing be difficult to meet the needs of fingerprint recognition.
The embodiment of the present application proposes a kind of method of fingerprint recognition, can be improved the efficiency of fingerprint recognition.
Fig. 1 shows the process interaction figure of the method for the fingerprint recognition of the embodiment of the present application.This method can be by main process task
Device and coprocessor are realized.
As shown in Figure 1, all or part of the steps during this approach includes the following steps.
In 110, coprocessor handles the fingerprint image of finger, obtains the fisrt feature number of the fingerprint image
According to.
In 120, coprocessor sends the fisrt feature data of the fingerprint image to primary processor.
In 130, primary processor receives the fisrt feature data for the fingerprint image that coprocessor is sent.
In 140, primary processor carries out fingerprint recognition according to the fisrt feature data of the fingerprint image.
When the fingerprint image can be placed on fingerprint collecting region for the finger, by the collected fingerprint image of fingerprint sensor
Picture, the mode that the embodiment of the present application acquires the fingerprint image to fingerprint sensor do not do any restriction.
Fingerprint image is handled due to increasing coprocessor, primary processor is able to use coprocessor to fingerprint image
As the characteristic progress fingerprint recognition that processing obtains, host processor and coprocessor is allow to cooperate with completion fingerprint recognition,
Improve the efficiency of fingerprint recognition.
Optionally, the method also includes: coprocessor to the fingerprint image in the process of processing, primary processor
Concurrently the fingerprint image is handled, obtains the second feature data of the fingerprint image.
Wherein, in 140, primary processor carries out fingerprint recognition according to the fisrt feature data of the fingerprint image, comprising:
Primary processor carries out fingerprint recognition according to the fisrt feature data of the fingerprint image and the second feature data of the fingerprint image.
That is, primary processor and coprocessor can jointly be handled fingerprint image, the fingerprint is obtained respectively
The second feature data and fisrt feature data of image, so that primary processor is according to the second feature data of the fingerprint image and
One characteristic carries out fingerprint recognition.Further, primary processor and coprocessor can concurrently carry out the fingerprint image
Processing, to improve the speed of fingerprint recognition.
Wherein, primary processor can be for example the CPU of equipment or processing unit of fingerprint identification device etc., coprocessor
It such as can be digital signal processor (Digital Signal Processor, DSP), graphics processor (Graphics
Processing Unit, GPU), field programmable gate array (Field Programmable Gate Array, FPGA), specially
With integrated circuit (Application Specific Integrated Circuits, ASIC) etc., figure can be effectively realized
As processing.
Hereinafter, being CPU with primary processor, coprocessor is described for being DSP, but the application is not limited to this.
CPU is scalar operation in the fingerprint algorithm used in the process of processing fingerprint image, and is single thread
Operation, in addition the bit wide of CPU is usually 32bit or 64bit, so that speed of this monokaryon serial manner in finger prints processing
It spends slower.
And to fingerprint image, in the process of processing, the fingerprint algorithm that DSP is used is vector operation, and it can be multi-thread
Image data is handled to journey, and the width of single data processing can achieve 1024bit.Therefore, place of the DSP to fingerprint image
Reason speed is apparently higher than CPU to the processing speed of fingerprint image.It can will be used in fingerprint identification process in the embodiment of the present application
Fingerprint algorithm be divided into two parts, a part executes on CPU, and another part executes on DSP, thus by the powerful of DSP
Image-capable improve fingerprint recognition efficiency.It can will especially be suitable for the part of the big data quantity operation of DSP in DSP
Upper execution, preferably to utilize the image-capable of DSP.
Optionally, as shown in Fig. 2, this method further include with 110 execute parallel 150 and 160.
In 150, primary processor handles the fingerprint image of the finger, obtains the second feature number of the fingerprint image
According to.
In 160, primary processor carries out fingerprint recognition according to the second feature data of the fingerprint image.
Wherein, if primary processor carries out fingerprint recognition failure according to the second feature data of the fingerprint image, before executing
120 to 140 stated.
Wherein, 150 and 160, and 110, it is parallel to execute.That is, primary processor to fingerprint image carry out processing and according to
During the second feature data arrived carry out fingerprint recognition, coprocessor is concurrently handled to obtain the fingerprint image
The fisrt feature data of the fingerprint image;In other words, coprocessor is handled the fingerprint image to obtain the fingerprint image
Fisrt feature data during, primary processor concurrently to the fingerprint image carry out processing and according to obtained second feature
Data carry out fingerprint recognition.
Since primary processor and coprocessor are to handle simultaneously the collected fingerprint image of fingerprint sensor, fingerprint
Recognition strategy becomes double-core parallelism recognition abundant by simply serially identifying, further improves the efficiency of fingerprint recognition.
Particularly, for the fingerprint recognition under the extreme cases such as dry finger and strong light direct beam, user can significantly be promoted
Experience significantly reduces influence of the external environment variation to the data of fingerprint image, improves adaptation of the fingerprint recognition to crowd
Property.Using the method for fingerprint recognition provided by the embodiments of the present application, the bulk velocity of fingerprint recognition is able to ascend at least 30%.
In order to improve the speed of fingerprint recognition, primary processor can be carried out first according to the second feature data of the fingerprint image
Fingerprint recognition.For example, primary processor obtains the collected first frame fingerprint image of optical fingerprint sensor, if according to first frame fingerprint image
The second feature data of picture carry out fingerprint recognition success, then can the fisrt feature data no longer to first frame fingerprint image know
Not;If carrying out fingerprint recognition failure according to the second feature data of first frame fingerprint image, continue according to first frame fingerprint image
Fisrt feature data carry out fingerprint recognition.
Primary processor for example can by application programming interfaces (Application Programming Interface,
API it) is connected between coprocessor, to realize the communication between primary processor and coprocessor.The primary processor for example can be with
It is connect by Service Provider Interface (Service Provider Interface, SPI) with fingerprint sensor.
Such as shown in Fig. 3, primary processor CPU, coprocessor DSP.CPU can execute a part fingerprint algorithm with
And fingerprint identification process is controlled, DSP can execute the fingerprint algorithm of another part.And CPU can be to fingerprint recognition
It is controlled, and the operation of DSP is controlled by api interface.It is connected between CPU and fingerprint sensor by SPI interface,
The collected fingerprint image of fingerprint sensor can be obtained.The fingerprint image that CPU is got can be for example sent to by API
DSP。
Fig. 4 is the schematic block diagram of CPU and DSP interior processing unit.Wherein, CPU is carried out using single thread (Thread)
Scalar operation (Scalar Operation, SO), for example, operation is carried out to scalar S0 using thread 0, single treatment data
Width is 32bit or 64bit.
And DSP carries out scalar operation using multithreading and carries out vector operation (Vector using the width of 1024bit
Operation, VO).It is, for example, possible to use threads 0 and thread 1 to handle scalar S0 to S3, and uses the width of 1024bit
Degree carries out vector operation to vector V0 to V3, and is handled using thread 2 and thread 3 scalar S4 to S7, and use
The width of 1024bit carries out vector operation to vector V4 to V7, and thread 0 to thread 3 can execute parallel.As can be seen that
Operational performance of the operational performance of DSP integrally than CPU is higher by 8-32 times.It is used in the process of processing when to the fingerprint image
Fingerprint algorithm a part realized on CPU and another part when being realized on DSP, can substantially accelerate processing speed
Degree.
By taking Fig. 5 as an example, Fig. 5 shows CPU and DSP respectively to the processing mode of a frame fingerprint image.Wherein, DSP can
By multiple threads, the multiple regions of the fingerprint image are handled simultaneously respectively, it is special with obtain the fingerprint image first
Levy data.As shown in figure 5, single thread, that is, thread 0 is used only to the fingerprint when CPU handles collected fingerprint image X
Image is handled.And DSP is when handling fingerprint image X, using multithreading, that is, thread 0 to thread 3 to fingerprint image X
In corresponding region handled.Assuming that the width fingerprint image is divided into totally 16 regions X0 to X15, wherein thread 0 is used for
The data in this four regions X0 to X3 are handled, thread 1 is used to handle the data in this four regions X4 to X7, and thread 2 is for handling
The data in this four regions X8 to X11, thread 3 are used to handle the data in this four regions X12 to X15, this four threads can be with
It is performed simultaneously fingerprint algorithm.Therefore, data, the speed of DSP multi-threading parallel process data are handled compared to CPU single-threaded serial
Much faster.
The high frequencies such as convolution, filtering used in fingerprint algorithm execute process can be realized by way of vector, therefore
The vector operation that can use width up to the 1024bit of DSP accelerates the processing of fingerprint image.Such as shown in Fig. 6, CPU is logical
It crosses single instrction to be handled, an instruction can only handle data 0.And DSP uses single-instruction multiple-data (Single
Instruction Multiple Data, SIMD) mode handled, realize through instruction treatmenting data 0 to data
N can make the big datas such as convolution, the filtering in processing and the speed of multi-cycle operation be doubled and redoubled.
The embodiment of the present application does not do any restriction to the treatment process of fingerprint image.For example, acquiring the fingerprint of finger to be measured
After image, image preprocessing can be carried out to collected fingerprint image, and image increasing is carried out to pretreated fingerprint image
By force, feature extraction is carried out to by the fingerprint image of image enhancement later, finally these characteristics and data based on extraction
Fingerprint template in library is matched, to complete fingerprint recognition.Also, image preprocessing or image are being carried out to fingerprint image
When enhancing, any mode such as convolution, high-pass filtering, median filtering can be used.
It should be understood that carrying out fingerprint recognition, Ke Yili according to the characteristic of fingerprint image described in the embodiment of the present application
Solution is to be according to the progress fingerprint image matching of the characteristic of the fingerprint image.
The second feature data that primary processor handles fingerprint image for example may include that this of coarse extraction refers to
The characteristic of print image, the fisrt feature data that coprocessor handles fingerprint image for example may include carefully mentioning
The fisrt feature data of the fingerprint image taken.
When using the stronger processor of image-capable as coprocessor, such as select DSP as coprocessor,
It since the image-capable of DSP is stronger, can handle more more complicated data, therefore taken the fingerprint figure using coprocessor
The complete data as in, and CPU is slower to the processing speed of fingerprint image, therefore in the image that can be used for roughly taking the fingerprint
Data.That is, the second feature data that CPU is obtained can be rough image data, i.e. coarse extraction;And DSP is obtained
Fisrt feature data can be comprehensive image data, i.e., it is thin to extract.
The application does not do any restriction, second feature data and fisrt feature to second feature data and fisrt feature data
Data are also possible to according to other modes division.For example, second feature data can be a part of region of fingerprint image
Image data, and fisrt feature data can be the image data in another part region of the fingerprint image.At this moment, primary processor
Concurrently fingerprint image is handled with coprocessor, and respectively obtains a part of region corresponding second of the fingerprint image
The corresponding fisrt feature data of characteristic and another part region.The fisrt feature data that coprocessor is obtained are sent
To primary processor, primary processor combines the second feature data of the fingerprint image and fisrt feature data to obtain entire fingerprint image
Characteristic and carry out fingerprint recognition.
Optionally, as shown in fig. 7, if primary processor is referred to according to the fisrt feature data of the fingerprint image in 140
Line recognition failures, this method can also include 710 to 740.
In 710, coprocessor handles another fingerprint image of the finger, obtains the of another fingerprint image
One characteristic.
In 720, coprocessor sends the fisrt feature data of another fingerprint image to primary processor.
Wherein, the fisrt feature data of another fingerprint image carry out fingerprint recognition for primary processor.
In 730, primary processor receives the fisrt feature data for another fingerprint image that coprocessor is sent.
740, primary processor carries out fingerprint recognition according to the fisrt feature data of another frame fingerprint image.
In the embodiment, in order to improve the success rate of fingerprint recognition, multiple fingerprint recognition chance can be provided.Primary processor
Fingerprint recognition can be carried out according to an at least width fingerprint image for the collected finger of the fingerprint sensor with coprocessor.Its
In, fingerprint sensor can be when the finger be placed on its fingerprint collecting region, the multiframe fingerprint image of the continuous acquisition finger.
After the frame fingerprint image to finger to be measured carries out fingerprint recognition failure, another frame fingerprint image of the finger can be used again
Secondary carry out fingerprint recognition.
For example, fingerprint sensor acquires two frame fingerprint images, i.e., first frame fingerprint image and time frame fingerprint image.When main place
Manage device and fingerprint recognition failure carried out according to the characteristic of first frame fingerprint image, then according to the characteristic of time frame fingerprint image into
Row fingerprint recognition.Wherein, primary processor can carry out fingerprint recognition according to the second feature data of first frame fingerprint image, and lose
Continue to carry out fingerprint recognition according to the fisrt feature data of first frame fingerprint image when losing.In primary processor according to first frame fingerprint image
Fisrt feature data also recognition failures when, can according to the second feature data of secondary frame fingerprint image carry out fingerprint recognition, and
Continue to carry out fingerprint recognition according to time fisrt feature data of frame fingerprint image in failure;Alternatively, primary processor refers in first frame
After line recognition failures, fingerprint recognition directly is carried out according to time fisrt feature data of frame fingerprint image.
Below with reference to Fig. 8 to Figure 10, the process of the fingerprint recognition of citing description the embodiment of the present application.Fig. 8 shows this Shen
Please embodiment a kind of possible fingerprint identification method.Primary processor shown in fig. 8 is CPU, coprocessor DSP.The party
Method includes:
In 801, CPU handles the first frame fingerprint image of finger to be measured, obtains the second feature of the fingerprint image
Data.
In 802, CPU carries out fingerprint recognition according to the second feature data of first frame fingerprint image.
In 830, DSP handles first frame fingerprint image, obtains the fisrt feature data of first frame fingerprint image.
Wherein, 801 and 802, and 803, it is parallel to execute.
If CPU carries out fingerprint recognition success according to the second feature data of first frame fingerprint image in 802, according to finger
The result of line identification executes corresponding operating, such as judges whether finger is finger, execution user's operation of user's registration etc..If
CPU carries out fingerprint recognition failure according to the second feature data of first frame fingerprint image, then executes 804 to 806.
In 804, DSP sends the fisrt feature data of first frame fingerprint image to CPU.
In 805, CPU receives the fisrt feature data for the first frame fingerprint image that DSP is sent.
CPU for example can send control instruction to DSP, to indicate that DSP is sent to it the fisrt feature of first frame fingerprint image
Data.DSP sends the fisrt feature data of first frame fingerprint image to CPU according to the control instruction received.If CPU base
Fingerprint recognition success is carried out in the second feature data of first frame fingerprint image, then DSP can not also send first frame fingerprint image to CPU
The fisrt feature data of picture, to save signaling overheads.
In 806, CPU carries out fingerprint recognition according to the fisrt feature data of first frame fingerprint image.
If CPU carries out fingerprint recognition success, basis according to the fisrt feature data of the head frame fingerprint image in 806
The result of fingerprint recognition executes corresponding operating.If CPU carries out fingerprint knowledge according to the fisrt feature data of the head frame fingerprint image
Do not fail, then can carry out fingerprint recognition for the secondary frame fingerprint image of the finger, that is, execute 807 to 810.
In 807, DSP handles the secondary frame fingerprint image of the finger, obtains time fisrt feature of frame fingerprint image
Data.
In 808, DSP sends time fisrt feature data of frame fingerprint image to CPU.
In 809, CPU receives the fisrt feature data for the secondary frame fingerprint image that DSP is sent.
CPU for example can send control instruction to DSP, to indicate that DSP is sent to it the first spy of the secondary frame fingerprint image
Levy data.DSP sends time fisrt feature data of frame fingerprint image according to the control instruction received, to CPU.
In 810, CPU carries out fingerprint recognition according to the fisrt feature data of the secondary frame fingerprint image.
If CPU carries out fingerprint recognition success according to time fisrt feature data of frame fingerprint image in 810, according to finger
The result of line identification executes corresponding operating.If CPU carries out fingerprint recognition mistake according to time fisrt feature data of frame fingerprint image
It loses, then can terminate fingerprint recognition.
Certainly, if allowing more fingerprint recognition, CPU can also use similar fashion to the third of the finger
Frame fingerprint image carries out fingerprint recognition.
It should be understood that magnitude of the sequence numbers of the above procedures are not meant to execute suitable in the various embodiments of the application
Sequence it is successive, the execution of each process sequence should be determined by its function and internal logic, the implementation without coping with the embodiment of the present application
Process constitutes any restriction.
For example, CPU and DSP is when carrying out feature extraction to fingerprint image, Fig. 8 is with DSP and CPU concurrently to fingerprint image
As being described for being handled, i.e., 801 and 802, it is parallel with 803;Alternatively, CPU is in the second feature of image that takes the fingerprint
Data and according to the second feature data carry out fingerprint recognition failure after, DSP extracts the fisrt feature data of the fingerprint image again,
And the fisrt feature data are sent to CPU and are used to carry out fingerprint recognition, i.e., 801 and 802 execute prior to 803.
In another example and then 807 in Fig. 8 can execute after 804, i.e. DSP sends the first of first frame fingerprint image
After characteristic, and then secondary frame fingerprint image is handled;Alternatively, DSP can also be executed after 806, i.e. DSP is determined
After CPU carries out fingerprint recognition failure according to the fisrt feature data of first frame fingerprint image, just secondary frame fingerprint image is handled.
As long as being handled in the method for fingerprint recognition, by means of image processors such as DSP fingerprint image
The protection scope of the application should be fallen into.
Below with reference to Fig. 9 and Figure 10, compares CPU and processing and CPU and DSP individually are carried out parallel to fingerprint to fingerprint image
The time-consuming that image is handled.
Fig. 9 show the process that CPU was individually handled fingerprint image and carried out fingerprint recognition.Wherein second feature number
According to the data for coarse extraction, fisrt feature data are the data carefully extracted.CPU extracts the second feature number of first frame fingerprint image
According to, and fingerprint recognition is carried out according to the second feature data of first frame fingerprint image;If fingerprint recognition fails, first frame fingerprint is extracted
The fisrt feature data of image, and fingerprint recognition is carried out according to the fisrt feature data of first frame fingerprint image;If recognition failures,
Time fisrt feature data of frame fingerprint image are extracted, and carry out fingerprint recognition according to the fisrt feature data of secondary frame fingerprint image.
Hereinafter, will acquire the second feature data of first frame fingerprint image and the fingerprint knowledge carried out based on the second feature data
The quick identification of frame fingerprint image headed by nickname;It will acquire the fisrt feature data of first frame fingerprint image and based on the fisrt feature
The fingerprint recognition that data carry out is known as the complete identification of first frame fingerprint image;Obtain time fisrt feature data of frame fingerprint image simultaneously
The fingerprint recognition carried out based on the fisrt feature data is known as time complete identification of frame fingerprint image.
As shown in Table 1, when CPU individually handles fingerprint image and carries out fingerprint recognition, CPU executes first frame fingerprint
A length of 237ms when the quickly identification of image is experienced, that is, the time-consuming for executing 901 and 902 is 237ms;CPU executes first frame image
Completely identification it is experienced when a length of 143ms, that is, the time-consuming for executing 903 and 904 is 143ms;CPU executes time frame fingerprint image
Completely identification it is experienced when a length of 330ms, that is, the time-consuming for executing 905 and 906 is 330ms.
Wherein, the time of first frame fingerprint image quickly identified not only includes time that second feature is extracted in table one, also
The processes the time it takes such as including image preprocessing;And the time of first frame fingerprint image completely identified only includes fisrt feature
The time of extraction, because the result that quickly processes such as image preprocessing in identification process obtain can be used directly.
Table one
Figure 10 show the process that CPU and DSP collaboration was handled fingerprint image and carried out fingerprint recognition.CPU is being mentioned
When taking the second feature data of first frame fingerprint image and carrying out fingerprint recognition according to the second feature data, DSP is concurrently extracted
The fisrt feature data of first frame fingerprint image, to carry out fingerprint knowledge according to the second feature data of first frame fingerprint image in CPU
Not Shi Bai when, can according to the fisrt feature data of first frame fingerprint image carry out fingerprint recognition.If CPU is according to first frame fingerprint image
The fisrt feature data of picture carry out fingerprint recognition and also fail, then DSP extracts time complete characterization of frame fingerprint image, and CPU is according to secondary
The complete characterization of frame fingerprint image carries out fingerprint recognition to secondary frame fingerprint image.
As shown in Table 1, it is 237ms that CPU and DSP, which executes the time-consuming of first frame fingerprint image quickly identified, that is, executes 1001
Time-consuming with 1002 is 237ms, wherein obtained in 1003 the processes of the fisrt feature data of first frame fingerprint image with 1001 and
1002 is parallel, no longer expends the extra time;The time-consuming completely identified that CPU and DSP executes first frame fingerprint image is 92ms, i.e.,
The time-consuming for executing 1004 and 1005 is 92ms;It is 250ms that CPU and DSP, which executes time time-consuming of frame image completely identified, that is, is executed
1006 to 1008 time-consuming is 250ms.
As can be seen that when CPU and DSP execute fingerprint recognition parallel, the time-consuming saving of first frame fingerprint image completely identified
35%, the time-consuming of secondary frame fingerprint image completely identified saves 24%.
Therefore, because DSP also assists in the process of fingerprint recognition, by the powerful image-capable of DSP, so that fingerprint
The efficiency of identification is significantly improved.Also, CPU and DSP passes through concurrently to the collected fingerprint image of fingerprint sensor
It is handled, the characteristic to obtain the fingerprint image further improves the efficiency of fingerprint recognition for fingerprint recognition.
Optionally, the fingerprint sensor in the embodiment of the present application can be optical fingerprint sensor, and but it is not limited to this, energy
Any fingerprint sensor for enough collecting fingerprint image is all fallen in the protection scope of the application.
The embodiment of the present application, which has been proposed that, can apply the image-capable of DSP at living body finger print (or true fingerprint)
In non-living body fingerprint (or pseudo- fingerprint, false fingerprint etc.) identification process, such as applied to the deep learning of true and false fingerprint recognition
In network.
For example, various types of true and false fingerprint images can be acquired to input as training sample, and by training sample
Deep learning network, according to the difference between the result of deep learning network output and practical desired result, to this
Deep learning network is adjusted, and so that the deep learning network is effectively differentiated fingerprint image by largely training study
The true and false.
Wherein, deep learning network, can be using DSP to the true and false when handling the true and false fingerprint image received
The fingerprint image of fingerprint is handled, to improve the speed handled fingerprint image, improves the instruction of deep learning network
Practice efficiency.
Optionally, method described in the embodiment of the present application can be used for living things feature recognition, i.e., by fingerprint image above-mentioned
As replacing with biological information, such as face recognition, iris recognition etc..
Figure 11 is the schematic block diagram according to the processor 1100 for fingerprint recognition of the embodiment of the present application.The processor
1200 be primary processor.As shown in figure 11, which includes:
Communication unit 1110, the fisrt feature data of the fingerprint image of the finger for receiving coprocessor transmission;
Fingerprint identification unit 1120 carries out fingerprint recognition for the fisrt feature data according to the fingerprint image.
Due to increasing coprocessor for carrying out the processing of fingerprint image, assist host processor and coprocessor
With fingerprint recognition is completed, the efficiency of fingerprint recognition is improved.
Optionally, the primary processor further includes processing unit 1130, and the processing unit 1130 is used for: at the association
Reason device handles during obtaining the fisrt feature data of the fingerprint image fingerprint image, concurrently to described
Fingerprint image is handled, and the second feature data of the fingerprint image are obtained;The fingerprint identification unit 1120 is also used to: root
Fingerprint recognition is carried out according to the second feature data of the fingerprint image;If being carried out according to the second feature data of the fingerprint image
Fingerprint recognition failure, then carry out fingerprint recognition according to the fisrt feature data of the fingerprint image.
Optionally, the communication unit 1110 is also used to, and receives another finger for the finger that the coprocessor is sent
The fisrt feature data of print image;The fingerprint identification unit 1120 is also used to, according to the first of another frame fingerprint image
Characteristic carries out fingerprint recognition.
Optionally, processing unit 1130 is used for: the coprocessor to the fingerprint image in the process of processing,
The fingerprint image is handled, the second feature data of the fingerprint image are obtained;The fingerprint identification unit 1120 has
Body is used for: the primary processor is according to the fisrt feature data of the fingerprint image and the second feature number of the fingerprint image
According to progress fingerprint recognition.
Optionally, the second feature data of the fingerprint image include the characteristic of the fingerprint image of coarse extraction,
The fisrt feature data of the fingerprint image include the characteristic of the fingerprint image carefully extracted.
Optionally, it is connected between the primary processor and the coprocessor by api interface.
Optionally, the primary processor is CPU, and the coprocessor is DSP.
Figure 12 is the schematic block diagram according to the processor 1200 for fingerprint recognition of the embodiment of the present application.The processor
1200 be coprocessor.As shown in figure 12, which includes:
Processing unit 1210 is handled for the fingerprint image to finger, obtains the fisrt feature of the fingerprint image
Data;
Communication unit 1220, for sending the fisrt feature data of the fingerprint image to primary processor, wherein the finger
The fisrt feature data of print image carry out fingerprint recognition for the primary processor.
Due to increasing coprocessor for carrying out the processing of fingerprint image, assist host processor and coprocessor
With fingerprint recognition is completed, the efficiency of fingerprint recognition is improved.
Optionally, the processing unit 1210 is specifically used for: handling in the primary processor the fingerprint image
The second feature data of the fingerprint image are obtained, and carry out fingerprint recognition according to the second feature data of the fingerprint image
In the process, concurrently the fingerprint image is handled, obtains the fisrt feature data of the fingerprint image, wherein if institute
It states primary processor and carries out fingerprint recognition failure according to the second feature data of the fingerprint image, the first of the fingerprint image is special
It levies data and carries out fingerprint recognition for the primary processor.
Optionally, the processing unit 1210 is also used to: if the primary processor is according to the first spy of the fingerprint image
It levies data and carries out fingerprint recognition failure, then another fingerprint image of the finger is handled, obtain another frame image
Fisrt feature data;The communication unit 1220 is also used to: Xiang Suoshu primary processor sends the of another fingerprint image
The fisrt feature data of one characteristic, another fingerprint image carry out fingerprint recognition for the primary processor.
Optionally, the second feature data of the fingerprint image include the characteristic of the fingerprint image of coarse extraction,
The fisrt feature data of the fingerprint image include the characteristic of the fingerprint image carefully extracted.
Optionally, it is connected between the primary processor and the coprocessor by api interface.
Optionally, the primary processor is CPU, and the coprocessor is DSP.
The embodiment of the present application also provides a kind of electronic equipment, which includes in the various embodiments of above-mentioned the application
Primary processor and coprocessor.
Optionally, which can be shown using the display screen in above description, such as LCD display or OLED
Screen.Wherein, when which is OLED display screen, the luminescent layer of the display screen includes multiple organic light-emitting diode light sources,
In the fingerprint identification device using at least partly excitation light source of the organic light-emitting diode light source as fingerprint recognition.
Optionally, the electronic equipment further includes fingerprint sensor, and the fingerprint sensor is for acquiring fingerprint image.
It is non-limiting as example, the electronic equipment can for terminal device, mobile phone, tablet computer, laptop,
The portable or mobiles such as desktop computer, game station, vehicle electronic device or wearable intelligent equipment calculate equipment, Yi Ji electricity
Other electronic equipments such as subdata base, automobile, ATM (automatic teller machine) (Automated Teller Machine, ATM).This is worn
The formula smart machine of wearing, which includes that function is complete, size is big, can not depend on smart phone, realizes complete or partial function, such as: intelligence
Wrist-watch or intelligent glasses etc., and only it is absorbed in certain a kind of application function, it needs to be used cooperatively with other equipment such as smart phone,
The Intelligent bracelet of such as all kinds of carry out sign monitorings, intelligent jewellery equipment.
It should be understood that the specific example in the embodiment of the present application is intended merely to that those skilled in the art is helped to more fully understand
The embodiment of the present application, rather than the range of the embodiment of the present application is limited, those skilled in the art can be on the basis of above-described embodiment
It is upper to carry out various improvement and deformations, and these are improved or deformation is all fallen in the protection scope of the application.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application should be based on the protection scope of the described claims.
Claims (26)
1. a kind of method of fingerprint recognition, which is characterized in that the described method includes:
Primary processor receives described in the coprocessor that coprocessor is sent handles the fingerprint image of finger
The fisrt feature data of fingerprint image;
The primary processor carries out fingerprint recognition according to the fisrt feature data of the fingerprint image.
2. the method according to claim 1, wherein the method also includes:
The coprocessor to the fingerprint image in the process of processing, primary processor is concurrently to the fingerprint image
It is handled, obtains the second feature data of the fingerprint image, and carry out according to the second feature data of the fingerprint image
Fingerprint recognition;
Wherein, the primary processor carries out fingerprint recognition according to the fisrt feature data of the fingerprint image, comprising:
If carrying out fingerprint recognition failure according to the second feature data of the fingerprint image, the primary processor is according to the fingerprint
The fisrt feature data of image carry out fingerprint recognition.
3. method according to claim 1 or 2, which is characterized in that the method also includes:
If carrying out fingerprint recognition failure according to the fisrt feature data of the fingerprint image, the primary processor is received at the association
Manage the fisrt feature data of another fingerprint image for the finger that device is sent;
The primary processor carries out fingerprint recognition according to the fisrt feature data of another fingerprint image.
4. according to claim 1 to method described in 3, which is characterized in that the second feature data of the fingerprint image include thick
The characteristic of the fingerprint image extracted, the fisrt feature data of the fingerprint image include the fingerprint image carefully extracted
The characteristic of picture.
5. according to claim 1 to method described in 4, which is characterized in that lead between the primary processor and the coprocessor
Cross application programming interfaces API connection.
6. according to claim 1 to method described in 5, which is characterized in that the primary processor is central processor CPU, described
Coprocessor is digital signal processor DSP.
7. a kind of method of fingerprint recognition, which is characterized in that the described method includes:
Coprocessor handles the fingerprint image of finger, obtains the fisrt feature data of the fingerprint image;
The coprocessor sends the fisrt feature data of the fingerprint image to primary processor, wherein the fingerprint image
Fisrt feature data carry out fingerprint recognition for the primary processor.
8. the method according to the description of claim 7 is characterized in that the coprocessor to the fingerprint image of finger at
Reason, obtains the fisrt feature data of the fingerprint image, comprising:
Handled to obtain the second feature data of the fingerprint image to the fingerprint image in the primary processor, and according to
During the second feature data of the fingerprint image carry out fingerprint recognition, the coprocessor is concurrently to the fingerprint image
Picture is handled, and the fisrt feature data of the fingerprint image are obtained,
Wherein, the fisrt feature data of the fingerprint image are for the primary processor in the second spy according to the fingerprint image
Sign data carry out carrying out fingerprint recognition when fingerprint recognition failure.
9. method according to claim 7 or 8, which is characterized in that the method also includes:
If the primary processor carries out fingerprint recognition failure, the coprocessor according to the fisrt feature data of the fingerprint image
Another fingerprint image of the finger is handled, the fisrt feature data of another frame image are obtained;
The data processor sends the fisrt feature data of another fingerprint image, another finger to the primary processor
The fisrt feature data of print image carry out fingerprint recognition for the primary processor.
10. method according to any one of claims 7 to 9, which is characterized in that the second feature number of the fingerprint image
According to the characteristic for the fingerprint image for including coarse extraction, the fisrt feature data of the fingerprint image include the institute carefully extracted
State the characteristic of fingerprint image.
11. method according to any one of claims 7 to 10, it is characterised in that the primary processor and the association are handled
It is connected between device by application programming interfaces API.
12. the method according to claim 7 to 11, which is characterized in that the primary processor is central processor CPU, institute
Stating coprocessor is digital signal processor DSP.
13. a kind of processor for fingerprint recognition, which is characterized in that the processor is primary processor, the primary processor
Include:
Communication unit, what the coprocessor for receiving coprocessor transmission handled the fingerprint image of finger
The fisrt feature data of the fingerprint image;
Fingerprint identification unit carries out fingerprint recognition for the fisrt feature data according to the fingerprint image.
14. processor according to claim 13, which is characterized in that the primary processor further includes processing unit,
The processing unit is used for: the coprocessor to the fingerprint image in the process of processing, concurrently to institute
It states fingerprint image to be handled, obtains the second feature data of the fingerprint image;
The fingerprint identification unit is also used to:
Fingerprint recognition is carried out according to the second feature data of the fingerprint image;
If fingerprint recognition failure is carried out according to the second feature data of the fingerprint image, according to the first of the fingerprint image
Characteristic carries out fingerprint recognition.
15. processor described in 3 or 14 according to claim 1, which is characterized in that
The communication unit is also used to, and receives the fisrt feature of another fingerprint image for the finger that the coprocessor is sent
Data;
The fingerprint identification unit is also used to, and according to the fisrt feature data of another fingerprint image, carries out fingerprint recognition.
16. processor described in 3 to 15 according to claim 1, which is characterized in that the second feature data packet of the fingerprint image
The characteristic of the fingerprint image of coarse extraction is included, the fisrt feature data of the fingerprint image include the finger carefully extracted
The characteristic of print image.
17. processor described in 3 to 16 according to claim 1, which is characterized in that the primary processor and the coprocessor it
Between connected by application programming interfaces API.
18. processor described in 3 to 17 according to claim 1, which is characterized in that the primary processor is central processor CPU,
The coprocessor is digital signal processor DSP.
19. a kind of processor for fingerprint recognition, which is characterized in that the processor is coprocessor, the coprocessor
Include:
Processing unit is handled for the fingerprint image to finger, obtains the fisrt feature data of the fingerprint image;
Communication unit, for sending the fisrt feature data of the fingerprint image to primary processor, wherein the fingerprint image
Fisrt feature data carry out fingerprint recognition for the primary processor.
20. processor according to claim 19, which is characterized in that the processing unit is specifically used for:
Handled to obtain the second feature data of the fingerprint image to the fingerprint image in the primary processor, and according to
During the second feature data of the fingerprint image carry out fingerprint recognition, concurrently the fingerprint image is handled,
The fisrt feature data of the fingerprint image are obtained,
Wherein, the fisrt feature data of the fingerprint image are for the primary processor in the second spy according to the fingerprint image
Sign data carry out carrying out fingerprint recognition when fingerprint recognition failure.
21. processor described in 9 or 20 according to claim 1, which is characterized in that the processing unit is also used to:
If the primary processor carries out fingerprint recognition failure according to the fisrt feature data of the fingerprint image, to the finger
Another fingerprint image handled, obtain the fisrt feature data of another frame image;
The communication unit is also used to:
The fisrt feature data of another fingerprint image are sent to the primary processor, the first of another fingerprint image is special
It levies data and carries out fingerprint recognition for the primary processor.
22. processor described in any one of 9 to 21 according to claim 1, which is characterized in that the second of the fingerprint image is special
Sign data include the characteristic of the fingerprint image of coarse extraction, and the fisrt feature data of the fingerprint image include thin extract
The fingerprint image characteristic.
23. processor described in any one of 9 to 22 according to claim 1, which is characterized in that the primary processor and the association
It is connected between processor by application programming interfaces API.
24. processor described in 9 to 23 according to claim 1, which is characterized in that the primary processor is central processor CPU,
The coprocessor is digital signal processor DSP.
25. a kind of electronic equipment, which is characterized in that including primary processor described according to claim 1 any one of 3 to 18,
And coprocessor described in any one of claim 19 to 24.
26. electronic equipment according to claim 25, which is characterized in that the electronic equipment further includes fingerprint sensor,
The fingerprint sensor is for acquiring fingerprint image.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2019/072655 WO2020150891A1 (en) | 2019-01-22 | 2019-01-22 | Fingerprint identification method, processor, and electronic device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109863505A true CN109863505A (en) | 2019-06-07 |
CN109863505B CN109863505B (en) | 2023-05-23 |
Family
ID=66889360
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201980000111.3A Active CN109863505B (en) | 2019-01-22 | 2019-01-22 | Fingerprint identification method, processor and electronic equipment |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109863505B (en) |
WO (1) | WO2020150891A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110928579A (en) * | 2019-09-29 | 2020-03-27 | 天博电子信息科技有限公司 | Instruction set acceleration method for biological recognition and processor |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150100427A1 (en) * | 2013-10-09 | 2015-04-09 | Ricoh Co., Ltd. | Associating Advertising Content with a Channel |
CN106022069A (en) * | 2016-05-30 | 2016-10-12 | 广东欧珀移动通信有限公司 | Unlocking control method and mobile terminal |
CN106295539A (en) * | 2016-08-03 | 2017-01-04 | 乐视控股(北京)有限公司 | Fingerprint identification method, fingerprint identification device and terminal unit |
US20170109564A1 (en) * | 2015-10-19 | 2017-04-20 | Qualcomm Incorporated | On-screen optical fingerprint capture for user authentication |
CN107066961A (en) * | 2017-03-31 | 2017-08-18 | 清华大学 | Fingerprint method for registering and device |
CN107358144A (en) * | 2017-05-20 | 2017-11-17 | 深圳信炜科技有限公司 | Image identification system and electronic installation |
CN107545166A (en) * | 2017-08-24 | 2018-01-05 | 广东欧珀移动通信有限公司 | Biometric discrimination method and Related product |
CN107688770A (en) * | 2016-08-05 | 2018-02-13 | 金佶科技股份有限公司 | Fingerprint identification module and fingerprint identification method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108256415A (en) * | 2017-11-30 | 2018-07-06 | 北京集创北方科技股份有限公司 | Fingerprint identification method, device and system, electronic equipment |
-
2019
- 2019-01-22 CN CN201980000111.3A patent/CN109863505B/en active Active
- 2019-01-22 WO PCT/CN2019/072655 patent/WO2020150891A1/en active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150100427A1 (en) * | 2013-10-09 | 2015-04-09 | Ricoh Co., Ltd. | Associating Advertising Content with a Channel |
US20170109564A1 (en) * | 2015-10-19 | 2017-04-20 | Qualcomm Incorporated | On-screen optical fingerprint capture for user authentication |
CN106022069A (en) * | 2016-05-30 | 2016-10-12 | 广东欧珀移动通信有限公司 | Unlocking control method and mobile terminal |
CN106295539A (en) * | 2016-08-03 | 2017-01-04 | 乐视控股(北京)有限公司 | Fingerprint identification method, fingerprint identification device and terminal unit |
CN107688770A (en) * | 2016-08-05 | 2018-02-13 | 金佶科技股份有限公司 | Fingerprint identification module and fingerprint identification method |
CN107066961A (en) * | 2017-03-31 | 2017-08-18 | 清华大学 | Fingerprint method for registering and device |
CN107358144A (en) * | 2017-05-20 | 2017-11-17 | 深圳信炜科技有限公司 | Image identification system and electronic installation |
CN107545166A (en) * | 2017-08-24 | 2018-01-05 | 广东欧珀移动通信有限公司 | Biometric discrimination method and Related product |
Non-Patent Citations (2)
Title |
---|
刘义鹏 等: "基于嵌入式GPU的汗孔识别算法并行设计", 传感器与微*** * |
钱磊 等: "基于可重构微服务器的高能效指纹比对方法", 计算机研究与发展 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110928579A (en) * | 2019-09-29 | 2020-03-27 | 天博电子信息科技有限公司 | Instruction set acceleration method for biological recognition and processor |
Also Published As
Publication number | Publication date |
---|---|
CN109863505B (en) | 2023-05-23 |
WO2020150891A1 (en) | 2020-07-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yap et al. | Deep learning in diabetic foot ulcers detection: A comprehensive evaluation | |
CN105631798B (en) | Low Power Consumption Portable realtime graphic object detecting and tracking system and method | |
KR102324697B1 (en) | Biometric detection method and device, electronic device, computer readable storage medium | |
Wang et al. | Smartphone-based wound assessment system for patients with diabetes | |
JP2020507850A (en) | Method, apparatus, equipment, and storage medium for determining the shape of an object in an image | |
CN104007817B (en) | Wearable information system at least one video camera | |
CN111914812B (en) | Image processing model training method, device, equipment and storage medium | |
CN112036331B (en) | Living body detection model training method, device, equipment and storage medium | |
Conti et al. | Accelerated visual context classification on a low-power smartwatch | |
Singh et al. | Human pose estimation using convolutional neural networks | |
US11556784B2 (en) | Multi-task fusion neural network architecture | |
CN110610191A (en) | Elevator floor identification method and device and terminal equipment | |
CN109583364A (en) | Image-recognizing method and equipment | |
CN108476296A (en) | Device and method for making the data of electronic device synchronize | |
Garcia-Martin et al. | Vision transformers for vein biometric recognition | |
WO2022143314A1 (en) | Object registration method and apparatus | |
WO2022161011A1 (en) | Method for generating image and electronic device | |
CN109863505A (en) | Method, processor and the electronic equipment of fingerprint recognition | |
CN114924950A (en) | Test method, electronic device, and computer-readable medium | |
CN112037305B (en) | Method, device and storage medium for reconstructing tree-like organization in image | |
Noureldeen et al. | Deep learning model for Cutaneous leishmaniasis detection and classification using Yolov5 | |
Reale et al. | Facial action unit analysis through 3d point cloud neural networks | |
CN111931544B (en) | Living body detection method, living body detection device, computing equipment and computer storage medium | |
CN113033305B (en) | Living body detection method, living body detection device, terminal equipment and storage medium | |
CN114970576A (en) | Identification code identification method, related electronic equipment and computer readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |