CN109863505B - Fingerprint identification method, processor and electronic equipment - Google Patents
Fingerprint identification method, processor and electronic equipment Download PDFInfo
- Publication number
- CN109863505B CN109863505B CN201980000111.3A CN201980000111A CN109863505B CN 109863505 B CN109863505 B CN 109863505B CN 201980000111 A CN201980000111 A CN 201980000111A CN 109863505 B CN109863505 B CN 109863505B
- Authority
- CN
- China
- Prior art keywords
- fingerprint image
- fingerprint
- characteristic data
- processor
- main processor
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
Abstract
The embodiment of the application provides a fingerprint identification method, which can improve fingerprint identification efficiency. The method comprises the following steps: the method comprises the steps that a main processor receives first characteristic data of fingerprint images, which are sent by a coprocessor and are obtained by the coprocessor in a processing mode, of the fingerprint images of fingers; and the main processor performs fingerprint identification according to the first characteristic data of the fingerprint image.
Description
Technical Field
The present application relates to the field of information technology, and more particularly, to a fingerprint identification method, a processor, and an electronic device.
Background
The fingerprint recognition technology is to collect fingerprint images of the finger through a fingerprint sensor and to carry out fingerprint recognition according to the fingerprint images through a central processing unit (Central Processing Unit, CPU). Because the fingerprint image is much disturbed in the acquisition process, and the data volume carried in the fingerprint image is huge, the algorithm complexity when the CPU processes the fingerprint image is increased, especially the fingerprint image acquired under the low temperature, the finger is drier, the strong light is directly irradiated and other drilling scenes, the processing efficiency of the CPU on the fingerprint image is obviously reduced, and the requirement of fingerprint identification is difficult to meet.
Disclosure of Invention
The embodiment of the application provides a fingerprint identification method, a processor and electronic equipment, which can improve fingerprint identification efficiency.
In a first aspect, a method for fingerprint identification is provided, including: the method comprises the steps that a main processor receives first characteristic data of fingerprint images, which are sent by a coprocessor and are obtained by the coprocessor in a processing mode, of the fingerprint images of fingers; and the main processor performs fingerprint identification according to the first characteristic data of the fingerprint image.
In one possible implementation, the method further includes: in the process of processing the fingerprint image by the coprocessor, the main processor processes the fingerprint image in parallel to obtain second characteristic data of the fingerprint image, and fingerprint identification is carried out according to the second characteristic data of the fingerprint image;
the main processor performs fingerprint identification according to the first feature data of the fingerprint image, and includes: and if the fingerprint identification is failed according to the second characteristic data of the fingerprint image, the main processor performs the fingerprint identification according to the first characteristic data of the fingerprint image.
In one possible implementation, the method further includes: if fingerprint identification fails according to the first characteristic data of the fingerprint image, the main processor receives the first characteristic data of another fingerprint image of the finger sent by the coprocessor; and the main processor performs fingerprint identification according to the first characteristic data of the other fingerprint image.
In one possible implementation, the method further includes: in the process of processing the fingerprint image by the coprocessor, the main processor processes the fingerprint image to obtain second characteristic data of the fingerprint image;
the main processor performs fingerprint identification according to the first feature data of the fingerprint image, and includes: and the main processor performs fingerprint identification according to the first characteristic data of the fingerprint image and the second characteristic data of the fingerprint image.
In one possible implementation, the second feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted coarsely, and the first feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted finely.
In one possible implementation, the main processor and the coprocessor are connected through an API interface.
In one possible implementation, the main processor is a CPU and the coprocessor is a DSP.
In a second aspect, a method for fingerprint identification is provided, including: the method comprises the steps that a coprocessor processes a fingerprint image of a finger to obtain first characteristic data of the fingerprint image; the coprocessor sends first characteristic data of the fingerprint image to a main processor, wherein the first characteristic data of the fingerprint image are used for fingerprint identification by the main processor.
In one possible implementation manner, the coprocessor processes a fingerprint image of a finger to obtain first feature data of the fingerprint image, including: and in the process that the main processor processes the fingerprint image to obtain second characteristic data of the fingerprint image and performs fingerprint identification according to the second characteristic data of the fingerprint image, the coprocessor processes the fingerprint image in parallel to obtain first characteristic data of the fingerprint image.
The first characteristic data of the fingerprint image are used for fingerprint identification when the main processor fails fingerprint identification according to the second characteristic data of the fingerprint image.
In one possible implementation, the method further includes: if the fingerprint identification of the main processor is failed according to the first characteristic data of the fingerprint image, the coprocessor processes the other fingerprint image of the finger to obtain the first characteristic data of the other frame image; the data processor sends first characteristic data of the other fingerprint image to the main processor, and the first characteristic data of the other fingerprint image is used for fingerprint identification by the main processor.
In one possible implementation manner, the coprocessor processes a fingerprint image of a finger to obtain first feature data of the fingerprint image, including: and in the process that the main processor processes the fingerprint image to obtain second characteristic data of the fingerprint image, the coprocessor processes the fingerprint image in parallel to obtain first characteristic data of the fingerprint image.
The first characteristic data of the fingerprint image and the second characteristic data of the fingerprint image are used for fingerprint identification loss by the main processor.
In one possible implementation, the second feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted coarsely, and the first feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted finely.
In one possible implementation, the host processor and the coprocessor are connected by an application programming interface API.
In one possible implementation, the main processor is a CPU and the coprocessor is a DSP.
In a third aspect, there is provided a main processor for fingerprint identification, comprising:
The communication unit is used for receiving first characteristic data of the fingerprint image, which is obtained by processing the fingerprint image of the finger by the coprocessor and is sent by the coprocessor;
and the fingerprint identification unit is used for carrying out fingerprint identification according to the first characteristic data of the fingerprint image.
In a possible implementation manner, the main processor further includes a processing unit, where the processing unit is configured to: processing the fingerprint image in parallel in the process of processing the fingerprint image by the coprocessor to obtain second characteristic data of the fingerprint image; the fingerprint identification unit is further configured to: fingerprint identification is carried out according to the second characteristic data of the fingerprint image; and if the fingerprint identification is failed according to the second characteristic data of the fingerprint image, the fingerprint identification is performed according to the first characteristic data of the fingerprint image.
In a possible implementation manner, the communication unit is further configured to receive first feature data of another fingerprint image of the finger sent by the coprocessor; the fingerprint identification unit is also used for carrying out fingerprint identification according to the first characteristic data of the other fingerprint image.
In a possible implementation manner, the main processor further includes a processing unit, where the processing unit is configured to: processing the fingerprint image in the process of processing the fingerprint image by the coprocessor to obtain second characteristic data of the fingerprint image; the fingerprint identification unit is specifically configured to: and the main processor performs fingerprint identification according to the first characteristic data of the fingerprint image and the second characteristic data of the fingerprint image.
In one possible implementation, the second feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted coarsely, and the first feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted finely.
In one possible implementation, the main processor and the coprocessor are connected through an API interface.
In one possible implementation, the main processor is a CPU and the coprocessor is a DSP.
In a fourth aspect, there is provided a coprocessor for fingerprint recognition, comprising:
the processing unit is used for processing the fingerprint image of the finger to obtain first characteristic data of the fingerprint image;
And the communication unit is used for sending the first characteristic data of the fingerprint image to the main processor, wherein the first characteristic data of the fingerprint image is used for fingerprint identification by the main processor.
In a possible implementation manner, the processing unit is specifically configured to: and processing the fingerprint image in parallel to obtain first characteristic data of the fingerprint image in the process of processing the fingerprint image by the main processor to obtain second characteristic data of the fingerprint image and carrying out fingerprint identification according to the second characteristic data of the fingerprint image. The first characteristic data of the fingerprint image are used for fingerprint identification when the main processor fails fingerprint identification according to the second characteristic data of the fingerprint image.
In a possible implementation, the processing unit is further configured to: if the fingerprint identification of the main processor is failed according to the first characteristic data of the fingerprint image, processing the other fingerprint image of the finger to obtain the first characteristic data of the other frame image;
the communication unit is further configured to: and sending the first characteristic data of the other fingerprint image to the main processor, wherein the first characteristic data of the other fingerprint image is used for fingerprint identification by the main processor.
In one possible implementation, the second feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted coarsely, and the first feature data of the fingerprint image includes the feature data of the fingerprint image that is extracted finely.
In one possible implementation, the host processor and the coprocessor are connected by an application programming interface API.
In one possible implementation, the main processor is a CPU and the coprocessor is a DSP.
In a fifth aspect, a chip is provided for implementing the method of the first aspect or any possible implementation manner of the first aspect. In particular, the chip comprises a main processor for calling and running a computer program from a memory, such that a device on which the chip is installed performs the method as described above in the first aspect or any of the possible implementations of the first aspect.
In a sixth aspect, a chip is provided for implementing the method of the second aspect or any possible implementation manner of the second aspect. In particular, the chip comprises a coprocessor for invoking and running a computer program from memory, such that a device on which the chip is installed performs the method as described above in the first aspect or any possible implementation of the first aspect.
In a seventh aspect, a computer readable storage medium is provided for storing a computer program for causing a computer to perform the method of the first aspect or any possible implementation of the first aspect.
In an eighth aspect, a computer-readable storage medium is provided for storing a computer program for causing a computer to perform the method of the second aspect or any possible implementation of the second aspect.
A ninth aspect provides an electronic device comprising the main processor of the third aspect or any possible implementation of the third aspect, and the co-processor of the fourth aspect or any possible implementation of the fourth aspect.
In a possible implementation, the electronic device further comprises a fingerprint sensor for capturing a fingerprint image.
Based on the technical scheme, the coprocessor is added for processing fingerprint images, so that the main processor and the coprocessor can cooperatively finish fingerprint identification, and the fingerprint identification efficiency is improved.
Drawings
Fig. 1 is a flowchart of a fingerprint identification method according to an embodiment of the present application.
Fig. 2 is a flowchart of a fingerprint identification method according to an embodiment of the present application.
Fig. 3 is a schematic block diagram of processing fingerprint images in cooperation with a CPU and DSP according to an embodiment of the present application.
Fig. 4 is a schematic block diagram of the CPU and DSP internal processing units.
Fig. 5 is a schematic diagram of the processing manner of each of the CPU and DSP for one frame fingerprint image.
Fig. 6 is a schematic diagram of each of the CPU and DSP performing data processing.
Fig. 7 is a flowchart of a fingerprint identification method according to an embodiment of the present application.
Fig. 8 is a flow chart of one possible fingerprint recognition method of the present application.
Fig. 9 is a schematic flow chart of processing a fingerprint image by the CPU alone.
Fig. 10 is a schematic flow chart of processing fingerprints by cooperation of a CPU and a DSP.
Fig. 11 is a schematic block diagram of a CPU for fingerprint recognition according to an embodiment of the present application.
Fig. 12 is a schematic block diagram of a DSP for fingerprint identification according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings.
The embodiment of the application is only described by taking an optical fingerprint system as an example, but should not be limited to any way, and the embodiment of the application is equally applicable to other fingerprint identification systems, such as capacitive fingerprint systems, ultrasonic fingerprint systems, etc., as long as fingerprint image acquisition can be realized.
In the optical fingerprint recognition process, a fingerprint sensor (or referred to as an image sensor or the like) collects an optical signal that light emitted from a light source is incident on and reflected by a finger, thereby obtaining a fingerprint image of the finger. The fingerprint image carries fingerprint information of the finger, and the fingerprint image is processed and matched with a pre-stored fingerprint template, so that fingerprint identification of the finger can be realized.
In general, the CPU performs fingerprint recognition based on the fingerprint image, for example, performs image preprocessing, image enhancement, image feature extraction, feature matching, and the like on the fingerprint image, and finally determines whether the fingerprint is a fingerprint that the user has authenticated and stored. The fingerprint algorithm used by the CPU is a scalar operation and is a single threaded operation, its bit width is typically 32 bits (bits) or 64 bits, and thus processing speed is slow in the face of a large amount of data in a fingerprint image. Particularly, the fingerprint image collected under the low-temperature, dry finger, strong light direct irradiation and other drilling scenes is obviously reduced in processing efficiency of the CPU on the fingerprint image, and the processing of the CPU on the fingerprint image is difficult to meet the fingerprint identification requirement.
The embodiment of the application provides a fingerprint identification method, which can improve the fingerprint identification efficiency.
Fig. 1 shows a flowchart of a fingerprint identification method according to an embodiment of the present application. The method may be implemented by a host processor and a coprocessor.
As shown in fig. 1, the method includes all or part of the following steps.
At 110, the coprocessor processes a fingerprint image of the finger to obtain first feature data of the fingerprint image.
At 120, the coprocessor sends first characteristic data of the fingerprint image to the host processor.
At 130, the host processor receives first characteristic data of the fingerprint image sent by the coprocessor.
In 140, the main processor performs fingerprint recognition based on the first characteristic data of the fingerprint image.
This fingerprint image can be for this finger places when the fingerprint collection region, the fingerprint image that is gathered by fingerprint sensor, and this application embodiment does not do any limit to the mode that fingerprint sensor gathered this fingerprint image.
As the coprocessor is added to process the fingerprint image, the main processor can use the coprocessor to fingerprint the characteristic data obtained by processing the fingerprint image, so that the main processor and the coprocessor can cooperatively complete fingerprint identification, and the fingerprint identification efficiency is improved.
Optionally, the method further comprises: and in the process of processing the fingerprint image by the coprocessor, the main processor processes the fingerprint image in parallel to obtain second characteristic data of the fingerprint image.
Wherein, in 140, the main processor performs fingerprint identification according to the first feature data of the fingerprint image, including: and the main processor performs fingerprint identification according to the first characteristic data of the fingerprint image and the second characteristic data of the fingerprint image.
That is, the main processor and the coprocessor may process the fingerprint image together, and acquire the second feature data and the first feature data of the fingerprint image, respectively, so that the main processor performs fingerprint identification according to the second feature data and the first feature data of the fingerprint image. Further, the main processor and the coprocessor can process the fingerprint image in parallel, so that the fingerprint identification speed is improved.
The main processor may be, for example, a CPU of the device or a processing unit of the fingerprint recognition device, and the coprocessor may be, for example, a digital signal processor (Digital Signal Processor, DSP), a graphics processor (Graphics Processing Unit, GPU), a field programmable gate array (Field Programmable Gate Array, FPGA), an application specific integrated circuit (Application Specific Integrated Circuits, ASIC), or the like, so that the image processing may be effectively implemented.
Hereinafter, the main processor is taken as a CPU, and the coprocessor is taken as a DSP for example, but the present application is not limited thereto.
The fingerprint algorithm used by the CPU in the process of processing the fingerprint image is scalar operation and is single-thread operation, and the bit width of the CPU is usually 32 bits or 64 bits, so that the speed of the single-core serial mode in fingerprint processing is slower.
In the process of processing the fingerprint image, the fingerprint algorithm used by the DSP is vector operation, the image data can be processed in a multithreading way, and the width of single data processing can reach 1024 bits. Therefore, the processing speed of the DSP on the fingerprint image is obviously higher than that of the CPU on the fingerprint image. In the embodiment of the application, the fingerprint algorithm used in the fingerprint identification process can be divided into two parts, one part is executed on the CPU, and the other part is executed on the DSP, so that the fingerprint identification efficiency is improved by means of the powerful image processing capability of the DSP. In particular, the portion of the large data size operation that is suitable for the DSP may be performed on the DSP to better utilize the image processing capabilities of the DSP.
Optionally, as shown in fig. 2, the method further includes 150 and 160 performed in parallel with 110.
At 150, the main processor processes the fingerprint image of the finger to obtain second feature data of the fingerprint image.
At 160, the host processor performs fingerprint recognition based on the second characteristic data of the fingerprint image.
If the main processor fails to perform fingerprint identification according to the second feature data of the fingerprint image, the foregoing 120 to 140 are executed.
150 and 160, and 110, are performed in parallel. Namely, in the process that the main processor processes the fingerprint image and carries out fingerprint identification according to the obtained second characteristic data, the coprocessor processes the fingerprint image in parallel to obtain the first characteristic data of the fingerprint image; or in the process of processing the fingerprint image by the coprocessor to acquire the first characteristic data of the fingerprint image, the main processor processes the fingerprint image in parallel and performs fingerprint identification according to the acquired second characteristic data.
Because the main processor and the coprocessor process the fingerprint images acquired by the fingerprint sensor at the same time, the fingerprint identification strategy is changed from simple serial identification to rich dual-core parallel identification, and the fingerprint identification efficiency is further improved.
Particularly, for fingerprint identification under extreme conditions such as dry fingers, direct irradiation of strong light and the like, user experience can be obviously improved, influence of external environment change on data of fingerprint images is greatly reduced, and adaptability of fingerprint identification to people is improved. By using the fingerprint identification method provided by the embodiment of the application, the overall speed of fingerprint identification can be improved by at least 30%.
In order to increase the speed of fingerprint identification, the main processor can perform fingerprint identification according to the second characteristic data of the fingerprint image. For example, the main processor acquires a first frame fingerprint image acquired by the optical fingerprint sensor, and if fingerprint identification is successful according to second characteristic data of the first frame fingerprint image, the first characteristic data of the first frame fingerprint image can not be identified any more; if the fingerprint identification is failed according to the second characteristic data of the first frame fingerprint image, continuing to carry out the fingerprint identification according to the first characteristic data of the first frame fingerprint image.
The host processor may be connected to the co-processor, for example, via an application program interface (Application Programming Interface, API) to enable communication between the host processor and the co-processor. The main processor may be connected to the fingerprint sensor, for example, through a service provider interface (Service Provider Interface, SPI).
For example, as shown in fig. 3, the main processor is a CPU, and the coprocessor is a DSP. The CPU may execute one portion of the fingerprinting algorithm and control the fingerprinting process, and the DSP may execute another portion of the fingerprinting algorithm. And the CPU can control fingerprint identification and control the operation of the DSP through the API interface. The CPU is connected with the fingerprint sensor through an SPI interface, so that a fingerprint image acquired by the fingerprint sensor can be acquired. The fingerprint image acquired by the CPU may be sent to the DSP through an API, for example.
Fig. 4 is a schematic block diagram of the CPU and DSP internal processing units. The CPU performs Scalar Operation (SO) using a single Thread (Thread), for example, performs Operation on Scalar S0 using Thread 0, and the width of single-pass data is 32 bits or 64 bits.
While DSPs use multithreading for scalar operations and 1024bit width for Vector Operations (VO). For example, scalar S0 to S3 may be processed using thread 0 and thread 1, vector operations may be performed on vectors V0 to V3 using 1024bit width, scalar S4 to S7 may be processed using thread 2 and thread 3, vector operations may be performed on vectors V4 to V7 using 1024bit width, and threads 0 to 3 may be executed in parallel. It can be seen that the overall operational performance of the DSP is 8-32 times higher than that of the CPU. When a part of the fingerprint algorithm used in processing the fingerprint image is implemented on the CPU and the other part is implemented on the DSP, the processing speed can be greatly increased.
Taking fig. 5 as an example, fig. 5 shows a processing manner of a CPU and a DSP for one frame fingerprint image, respectively. The DSP can process a plurality of areas of the fingerprint image simultaneously through a plurality of threads respectively to obtain first characteristic data of the fingerprint image. As shown in fig. 5, when the CPU processes the acquired fingerprint image X, the fingerprint image is processed using only a single thread, that is, thread 0. When the DSP processes the fingerprint image X, the corresponding region in the fingerprint image X is processed using the multithreading, i.e., the threads 0 to 3. Assuming that the fingerprint image is divided into 16 regions X0 to X15, where thread 0 is used to process data of four regions X0 to X3, thread 1 is used to process data of four regions X4 to X7, thread 2 is used to process data of four regions X8 to X11, thread 3 is used to process data of four regions X12 to X15, and the four threads can simultaneously execute the fingerprint algorithm. Thus, the speed of DSP multithreading parallel processing of data is much faster than the CPU single line Cheng Chuanhang processing the data.
The high-frequency execution processes such as convolution, filtering and the like used in the fingerprint algorithm can be realized in a vector mode, so that the processing of the fingerprint image can be accelerated by using the vector operation with the width of 1024 bits of the DSP. For example, as shown in fig. 6, the CPU performs processing by a single instruction, and only data 0 can be processed by one instruction. While DSP processes with single instruction multiple data (Single Instruction Multiple Data, SIMD) to process data 0 through data n with one instruction, the speed of convolution, filtering, etc. large data and multi-cycle operation in processing can be multiplied.
The processing procedure of the fingerprint image is not limited in this embodiment. For example, after collecting the fingerprint image of the finger to be detected, the collected fingerprint image may be subjected to image preprocessing, the preprocessed fingerprint image is subjected to image enhancement, then the fingerprint image subjected to image enhancement is subjected to feature extraction, and finally the fingerprint image is matched with the fingerprint template in the database based on the extracted feature data, so that fingerprint identification is completed. Also, any means such as convolution, high-pass filtering, median filtering, etc. may be used in image preprocessing or image enhancement of the fingerprint image.
It should be understood that, in the embodiments of the present application, fingerprint identification is performed according to the feature data of the fingerprint image, which may be understood as performing fingerprint image matching according to the feature data of the fingerprint image.
The second feature data obtained by processing the fingerprint image by the main processor may for example comprise the feature data of the fingerprint image which is extracted coarsely, and the first feature data obtained by processing the fingerprint image by the coprocessor may for example comprise the first feature data of the fingerprint image which is extracted finely.
When a processor with stronger image processing capability is used as the coprocessor, for example, a DSP is selected as the coprocessor, more and more complex data can be processed due to the stronger image processing capability of the DSP, so that the coprocessor is used for extracting the complete data in the fingerprint image, and the CPU has slower processing speed on the fingerprint image, so that the DSP can be used for roughly extracting the data in the fingerprint image. That is, the second feature data acquired by the CPU may be rough image data, i.e., rough extraction; the first feature data acquired by the DSP may be comprehensive image data, i.e. fine extraction.
The second feature data and the first feature data are not limited in any way, and may be divided according to other ways. For example, the second feature data may be image data of a partial region of a fingerprint image, and the first feature data may be image data of another partial region of the fingerprint image. At this time, the main processor processes the fingerprint image in parallel with the coprocessor, and obtains second feature data corresponding to a part of the region of the fingerprint image and first feature data corresponding to another part of the region of the fingerprint image, respectively. The coprocessor sends the obtained first characteristic data to the main processor, and the main processor combines the second characteristic data and the first characteristic data of the fingerprint image to obtain the characteristic data of the whole fingerprint image and performs fingerprint identification.
Optionally, as shown in fig. 7, if the main processor fails to perform fingerprint identification according to the first feature data of the fingerprint image in 140, the method may further include 710 to 740.
At 710, the co-processor processes another fingerprint image of the finger to obtain first feature data for the other fingerprint image.
At 720, the coprocessor sends first feature data of the other fingerprint image to the host processor.
Wherein the first characteristic data of the other fingerprint image is used for fingerprint identification by the main processor.
At 730, the host processor receives first feature data of the other fingerprint image sent by the co-processor.
At 740, the host processor performs fingerprint recognition based on the first feature data of the other frame of fingerprint image.
In this embodiment, in order to improve the success rate of fingerprint recognition, multiple fingerprint recognition opportunities may be provided. The main processor and the coprocessor can perform fingerprint identification according to at least one fingerprint image of the finger acquired by the fingerprint sensor. The fingerprint sensor can continuously collect multi-frame fingerprint images of the finger when the finger is placed in the fingerprint collecting area. When fingerprint identification of one frame of fingerprint image of the finger to be detected fails, fingerprint identification can be carried out again by using the other frame of fingerprint image of the finger.
For example, the fingerprint sensor captures two frames of fingerprint images, namely a first frame of fingerprint image and a second frame of fingerprint image. And when the main processor fails to perform fingerprint identification according to the characteristic data of the first frame fingerprint image, performing fingerprint identification according to the characteristic data of the second frame fingerprint image. The main processor can conduct fingerprint identification according to the second characteristic data of the first frame fingerprint image, and continues to conduct fingerprint identification according to the first characteristic data of the first frame fingerprint image when failure occurs. When the main processor fails to identify according to the first characteristic data of the first frame fingerprint image, fingerprint identification can be carried out according to the second characteristic data of the sub frame fingerprint image, and fingerprint identification is continuously carried out according to the first characteristic data of the sub frame fingerprint image when the main processor fails; or after the first frame fingerprint identification fails, the main processor directly performs fingerprint identification according to the first characteristic data of the sub-frame fingerprint image.
The following describes the fingerprint recognition process according to the embodiment of the present application by way of example with reference to fig. 8 to 10. Fig. 8 illustrates one possible fingerprint identification method according to an embodiment of the present application. The main processor shown in fig. 8 is a CPU and the coprocessor is a DSP. The method comprises the following steps:
in 801, the CPU processes a first frame fingerprint image of a finger to be detected to obtain second feature data of the fingerprint image.
In 802, the CPU performs fingerprint recognition based on second feature data of the first frame fingerprint image.
In 830, the DSP processes the first frame fingerprint image to obtain first feature data of the first frame fingerprint image.
Wherein 801 and 802, and 803, are performed in parallel.
If the CPU performs fingerprint recognition based on the second feature data of the first frame fingerprint image successfully in 802, a corresponding operation is performed based on the result of fingerprint recognition, for example, whether the finger is a registered finger of the user, performing a user operation, or the like. If the CPU fails to fingerprint based on the second feature data of the first frame fingerprint image, then execution proceeds to 804 through 806.
In 804, the DSP sends first feature data of the first frame fingerprint image to the CPU.
In 805, the CPU receives first feature data of a first frame fingerprint image transmitted by the DSP.
The CPU may, for example, send a control instruction to the DSP instructing the DSP to send first characteristic data of the first frame fingerprint image thereto. And the DSP sends first characteristic data of the first frame fingerprint image to the CPU according to the received control instruction. If the CPU performs fingerprint identification successfully based on the second characteristic data of the first frame fingerprint image, the DSP can also not send the first characteristic data of the first frame fingerprint image to the CPU so as to save signaling overhead.
At 806, the CPU performs fingerprint recognition based on the first feature data of the first frame fingerprint image.
If the CPU performs fingerprint identification successfully according to the first feature data of the first frame fingerprint image at 806, a corresponding operation is performed according to the result of fingerprint identification. If the CPU fails to perform fingerprint recognition based on the first feature data of the first frame fingerprint image, fingerprint recognition may be performed for the sub-frame fingerprint image of the finger, i.e., execution 807 to 810.
In 807, the DSP processes the sub-frame fingerprint image of the finger to obtain first feature data for the sub-frame fingerprint image.
At 808, the DSP sends the first characteristic data of the sub-frame fingerprint image to the CPU.
In 809, the CPU receives first feature data of the sub-frame fingerprint image transmitted by the DSP.
The CPU may, for example, send a control instruction to the DSP instructing the DSP to send the first characteristic data of the sub-frame fingerprint image thereto. And the DSP sends the first characteristic data of the sub-frame fingerprint image to the CPU according to the received control instruction.
In 810, the CPU performs fingerprint recognition based on the first feature data of the sub-frame fingerprint image.
If the CPU performs fingerprint recognition based on the first feature data of the sub-frame fingerprint image successfully at 810, a corresponding operation is performed based on the result of the fingerprint recognition. If the CPU fails to perform fingerprint identification according to the first characteristic data of the sub-frame fingerprint image, the fingerprint identification can be terminated.
Of course, if more fingerprinting is allowed, the CPU may also fingerprint the third frame of fingerprint image of the finger in a similar manner.
It should be understood that, in various embodiments of the present application, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present application.
For example, when the CPU and the DSP perform feature extraction on the fingerprint image, fig. 8 is a description taking the example that the DSP and the CPU perform processing on the fingerprint image in parallel, i.e., 801 and 802, and 803 are parallel; or, after the CPU extracts the second feature data of the fingerprint image and performs fingerprint recognition according to the second feature data, the DSP re-extracts the first feature data of the fingerprint image and sends the first feature data to the CPU for fingerprint recognition, that is, 801 and 802 are performed before 803.
As another example, 807 in fig. 8 may be performed immediately after 804, i.e., the DSP sends the first feature data of the first frame fingerprint image and then processes the second frame fingerprint image; alternatively, the DSP may be executed after 806, that is, the DSP determines that the CPU fails to perform fingerprint recognition according to the first feature data of the first frame fingerprint image, and then processes the second frame fingerprint image.
Any method for processing a fingerprint image for fingerprint recognition by means of an image processor such as a DSP should fall within the scope of the present application.
In the following, with reference to fig. 9 and 10, the time consumed by the CPU processing the fingerprint image alone and the CPU and DSP processing the fingerprint image in parallel is compared.
Fig. 9 shows a process in which the CPU processes the fingerprint image alone and performs fingerprint recognition. Wherein the second characteristic data is coarse extracted data and the first characteristic data is fine extracted data. The CPU extracts second characteristic data of the first frame fingerprint image and performs fingerprint identification according to the second characteristic data of the first frame fingerprint image; if the fingerprint identification fails, extracting first characteristic data of the first frame fingerprint image, and carrying out fingerprint identification according to the first characteristic data of the first frame fingerprint image; if the identification fails, extracting first characteristic data of the sub-frame fingerprint image, and carrying out fingerprint identification according to the first characteristic data of the sub-frame fingerprint image.
The second characteristic data of the first frame fingerprint image is acquired, and fingerprint identification based on the second characteristic data is called as fast identification of the first frame fingerprint image; the first characteristic data of the first frame fingerprint image is acquired, and fingerprint identification based on the first characteristic data is called complete identification of the first frame fingerprint image; the first feature data of the sub-frame fingerprint image is acquired and fingerprint identification based on the first feature data is called complete identification of the sub-frame fingerprint image.
As shown in table one, when the CPU processes and performs fingerprint recognition on the fingerprint image alone, the duration of time for the CPU to perform the fast recognition of the first frame fingerprint image is 237ms, that is, the time consumed for performing 901 and 902 is 237ms; the duration that the CPU has experienced to perform the complete recognition of the first frame image is 143ms, i.e. the time taken to perform 903 and 904 is 143ms; the duration that the CPU has experienced to perform a complete identification of the sub-frame fingerprint image is 330ms, i.e. the time taken to perform 905 and 906 is 330ms.
The time for quickly identifying the first frame fingerprint image in the first table comprises the time for extracting the second characteristic and the time spent in the processes of image preprocessing and the like; the time of complete recognition of the first frame fingerprint image only comprises the time of first feature extraction, because the results obtained by the processes of image preprocessing and the like in the rapid recognition process can be directly used.
List one
Fig. 10 shows a process of processing a fingerprint image and performing fingerprint recognition by cooperation of a CPU and a DSP. When the CPU extracts the second characteristic data of the first frame fingerprint image and performs fingerprint identification according to the second characteristic data, the DSP extracts the first characteristic data of the first frame fingerprint image in parallel, so that when the CPU fails to perform fingerprint identification according to the second characteristic data of the first frame fingerprint image, the CPU can perform fingerprint identification according to the first characteristic data of the first frame fingerprint image. If the CPU fails to perform fingerprint identification according to the first characteristic data of the first frame fingerprint image, the DSP extracts the complete characteristics of the sub-frame fingerprint image, and the CPU performs fingerprint identification on the sub-frame fingerprint image according to the complete characteristics of the sub-frame fingerprint image.
As shown in table one, the time consumed by the CPU and DSP for performing the fast recognition of the first frame fingerprint image is 237ms, that is, the time consumed by the processes for performing 1001 and 1002 is 237ms, wherein the process for acquiring the first feature data of the first frame fingerprint image in 1003 is parallel to the processes for performing 1001 and 1002, and no extra time is consumed; the time taken by the CPU and the DSP to perform the complete recognition of the first frame fingerprint image is 92ms, i.e. the time taken by the CPU and the DSP to perform 1004 and 1005 is 92ms; the CPU and DSP take 250ms to perform the complete recognition of the sub-frame image, i.e. the time taken to perform 1006 to 1008 is 250ms.
It can be seen that when the CPU and the DSP execute fingerprint identification in parallel, the time consumption for complete identification of the first frame fingerprint image is saved by 35%, and the time consumption for complete identification of the second frame fingerprint image is saved by 24%.
Therefore, the DSP also participates in the fingerprint identification process, and the fingerprint identification efficiency is obviously improved by means of the powerful image processing capability of the DSP. And the CPU and the DSP process the fingerprint image acquired by the fingerprint sensor in parallel to obtain the characteristic data of the fingerprint image for fingerprint identification, so that the fingerprint identification efficiency is further improved.
Optionally, the fingerprint sensor in the embodiment of the present application may be an optical fingerprint sensor, but is not limited thereto, and any fingerprint sensor capable of capturing a fingerprint image falls within the scope of the present application.
The embodiment of the application also provides that the image processing capability of the DSP can be applied to the identification process of living fingerprints (or called true fingerprints) and non-living fingerprints (or called false fingerprints, false fingerprints and the like), for example, the image processing capability of the DSP can be applied to a deep learning network for identifying true fingerprints.
For example, various types of true and false fingerprint images can be collected as training samples, the training samples are input into a deep learning network, the deep learning network is adjusted according to the difference between the result output by the deep learning network and the actually expected result, and the deep learning network can effectively distinguish the true and false fingerprint images through a large amount of training learning.
When the received true and false fingerprint images are processed by the deep learning network, the DSP can be used for processing the fingerprint images of the true and false fingerprints, so that the speed of processing the fingerprint images is improved, and the training efficiency of the deep learning network is improved.
Alternatively, the method described in the embodiments of the present application may also be used for biometric identification, i.e. replacing the aforementioned fingerprint image with biometric information, such as facial identification, iris identification, etc.
Fig. 11 is a schematic block diagram of a processor 1100 for fingerprint identification according to an embodiment of the present application. The processor 1200 is the main processor. As shown in fig. 11, the main processor 1100 includes:
a communication unit 1110, configured to receive first feature data of a fingerprint image of a finger sent by the coprocessor;
and a fingerprint identification unit 1120, configured to perform fingerprint identification according to the first feature data of the fingerprint image.
As the coprocessor is additionally used for processing fingerprint images, the main processor and the coprocessor can cooperatively finish fingerprint identification, and the fingerprint identification efficiency is improved.
Optionally, the main processor further includes a processing unit 1130, where the processing unit 1130 is configured to: in the process that the coprocessor processes the fingerprint image to obtain first characteristic data of the fingerprint image, the fingerprint image is processed in parallel to obtain second characteristic data of the fingerprint image; the fingerprint recognition unit 1120 is further configured to: fingerprint identification is carried out according to the second characteristic data of the fingerprint image; and if the fingerprint identification is failed according to the second characteristic data of the fingerprint image, the fingerprint identification is performed according to the first characteristic data of the fingerprint image.
Optionally, the communication unit 1110 is further configured to receive first feature data of another fingerprint image of the finger sent by the coprocessor; the fingerprint identification unit 1120 is further configured to perform fingerprint identification according to the first feature data of the fingerprint image of the other frame.
Optionally, the processing unit 1130 is configured to: processing the fingerprint image in the process of processing the fingerprint image by the coprocessor to obtain second characteristic data of the fingerprint image; the fingerprint identification unit 1120 is specifically configured to: and the main processor performs fingerprint identification according to the first characteristic data of the fingerprint image and the second characteristic data of the fingerprint image.
Optionally, the second feature data of the fingerprint image includes feature data of the fingerprint image that is extracted coarsely, and the first feature data of the fingerprint image includes feature data of the fingerprint image that is extracted finely.
Optionally, the main processor and the coprocessor are connected through an API interface.
Optionally, the main processor is a CPU, and the coprocessor is a DSP.
Fig. 12 is a schematic block diagram of a processor 1200 for fingerprint identification according to an embodiment of the present application. The processor 1200 is a coprocessor. As shown in fig. 12, the main processor 1200 includes:
A processing unit 1210, configured to process a fingerprint image of a finger to obtain first feature data of the fingerprint image;
and a communication unit 1220, configured to send the first feature data of the fingerprint image to a main processor, where the first feature data of the fingerprint image is used for fingerprint identification by the main processor.
As the coprocessor is additionally used for processing fingerprint images, the main processor and the coprocessor can cooperatively finish fingerprint identification, and the fingerprint identification efficiency is improved.
Optionally, the processing unit 1210 is specifically configured to: and processing the fingerprint image in parallel to obtain first characteristic data of the fingerprint image in the process of processing the fingerprint image by the main processor to obtain second characteristic data of the fingerprint image and carrying out fingerprint identification according to the second characteristic data of the fingerprint image, wherein if the fingerprint identification of the main processor according to the second characteristic data of the fingerprint image fails, the first characteristic data of the fingerprint image is used for the main processor to carry out fingerprint identification.
Optionally, the processing unit 1210 is further configured to: if the fingerprint identification of the main processor is failed according to the first characteristic data of the fingerprint image, processing the other fingerprint image of the finger to obtain the first characteristic data of the other frame image; the communication unit 1220 is further configured to: and sending the first characteristic data of the other fingerprint image to the main processor, wherein the first characteristic data of the other fingerprint image is used for fingerprint identification by the main processor.
Optionally, the second feature data of the fingerprint image includes feature data of the fingerprint image that is extracted coarsely, and the first feature data of the fingerprint image includes feature data of the fingerprint image that is extracted finely.
Optionally, the main processor and the coprocessor are connected through an API interface.
Optionally, the main processor is a CPU, and the coprocessor is a DSP.
The embodiment of the application also provides electronic equipment, which comprises the main processor and the coprocessor in the various embodiments of the application.
Alternatively, the display screen may employ the display screen described above, such as an LCD display screen or an OLED display screen. When the display screen is an OLED display screen, the light-emitting layer of the display screen comprises a plurality of organic light-emitting diode light sources, and at least part of the organic light-emitting diode light sources are adopted as excitation light sources for fingerprint identification by the fingerprint identification device.
Optionally, the electronic device further comprises a fingerprint sensor for capturing a fingerprint image.
By way of example, and not limitation, the electronic device may be a portable or mobile computing device such as a terminal device, a cell phone, a tablet computer, a notebook computer, a desktop computer, a gaming device, an in-vehicle electronic device, or a wearable smart device, as well as other electronic devices such as an electronic database, an automobile, a bank automated teller machine (Automated Teller Machine, ATM), and the like. The wearable intelligent device comprises full functions, large size and complete or partial functions which can be realized independent of the intelligent mobile phone, for example: smart watches or smart glasses, etc., and are only focused on certain application functions, and need to be used in combination with other devices, such as smart phones, as well as devices for monitoring physical signs, such as smart bracelets, smart jewelry, etc.
It should be understood that the specific examples in the embodiments of the present application are only for helping those skilled in the art to better understand the embodiments of the present application, and not limit the scope of the embodiments of the present application, and those skilled in the art may make various improvements and modifications based on the above embodiments, and these improvements or modifications fall within the protection scope of the present application.
The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (18)
1. A method of fingerprint identification, the method being applied to an electronic device comprising an optical fingerprint identification system disposed below a display screen, the optical fingerprint identification system comprising an optical fingerprint sensor for capturing a fingerprint image of a finger above the display screen, the method comprising:
the method comprises the steps that a main processor receives first characteristic data of a fingerprint image, which is sent by a coprocessor and is obtained by processing the fingerprint image by the coprocessor;
The main processor performs fingerprint identification according to the first characteristic data of the fingerprint image;
the method further comprises the steps of:
in the process of processing the fingerprint image by the coprocessor, the main processor processes the fingerprint image in parallel to obtain second characteristic data of the fingerprint image, and fingerprint identification is carried out according to the second characteristic data of the fingerprint image;
the main processor performs fingerprint identification according to the first feature data of the fingerprint image, and includes:
if the main processor fails to perform fingerprint identification according to the second characteristic data of the fingerprint image, the main processor performs fingerprint identification according to the first characteristic data of the fingerprint image;
the second characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a rough mode, and the first characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a fine mode.
2. The method according to claim 1, wherein the method further comprises:
if fingerprint identification fails according to the first characteristic data of the fingerprint image, the main processor receives the first characteristic data of another fingerprint image of the finger sent by the coprocessor;
And the main processor performs fingerprint identification according to the first characteristic data of the other fingerprint image.
3. A method according to claim 1 or 2, wherein the host processor and the coprocessor are connected by an application programming interface API.
4. A method according to claim 1 or 2, wherein the main processor is a central processing unit CPU and the co-processor is a digital signal processor DSP.
5. A method of fingerprint identification, the method being applied to an electronic device comprising an optical fingerprint identification system disposed below a display screen, the optical fingerprint identification system comprising an optical fingerprint sensor for capturing a fingerprint image of a finger above the display screen, the method comprising:
the coprocessor processes the fingerprint image to obtain first characteristic data of the fingerprint image;
the coprocessor sends first characteristic data of the fingerprint image to a main processor, wherein the first characteristic data of the fingerprint image are used for fingerprint identification by the main processor;
the coprocessor processes a fingerprint image of a finger to obtain first characteristic data of the fingerprint image, and the method comprises the following steps:
In the process that the main processor processes the fingerprint image to obtain second characteristic data of the fingerprint image and performs fingerprint identification according to the second characteristic data of the fingerprint image, the coprocessor processes the fingerprint image in parallel to obtain first characteristic data of the fingerprint image,
the first characteristic data of the fingerprint image is used for fingerprint identification when the main processor fails fingerprint identification according to the second characteristic data of the fingerprint image;
the second characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a rough mode, and the first characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a fine mode.
6. The method of claim 5, wherein the method further comprises:
if the fingerprint identification of the main processor is failed according to the first characteristic data of the fingerprint image, the coprocessor processes the other fingerprint image of the finger to obtain the first characteristic data of the other fingerprint image;
the coprocessor sends first characteristic data of the other fingerprint image to the main processor, and the first characteristic data of the other fingerprint image are used for fingerprint identification of the main processor.
7. The method of claim 5 or 6, wherein the host processor and the coprocessor are connected by an application programming interface API.
8. The method according to claim 5 or 6, wherein the main processor is a central processing unit CPU and the co-processor is a digital signal processor DSP.
9. A processor for fingerprint identification, wherein the processor is applied to electronic equipment, the electronic equipment includes the optical fingerprint identification system that sets up in the display screen below, the optical fingerprint identification system includes optical fingerprint sensor, optical fingerprint sensor is used for gathering the fingerprint image of finger above the display screen, the processor is main processor, main processor includes:
the communication unit is used for receiving first characteristic data of the fingerprint image, which is sent by the coprocessor and is obtained by processing the fingerprint image by the coprocessor;
the fingerprint identification unit is used for carrying out fingerprint identification according to the first characteristic data of the fingerprint image;
the main processor further comprises a processing unit,
the processing unit is used for: processing the fingerprint image in parallel in the process of processing the fingerprint image by the coprocessor to obtain second characteristic data of the fingerprint image;
The fingerprint identification unit is further configured to:
fingerprint identification is carried out according to the second characteristic data of the fingerprint image;
if the fingerprint identification is failed according to the second characteristic data of the fingerprint image, the fingerprint identification is performed according to the first characteristic data of the fingerprint image;
the second characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a rough mode, and the first characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a fine mode.
10. The processor of claim 9, wherein the processor further comprises a processor controller,
the communication unit is further used for receiving first characteristic data of another fingerprint image of the finger sent by the coprocessor;
the fingerprint identification unit is also used for carrying out fingerprint identification according to the first characteristic data of the other fingerprint image.
11. The processor according to claim 9 or 10, wherein the host processor and the coprocessor are connected by an application programming interface API.
12. The processor according to claim 9 or 10, wherein the main processor is a central processing unit CPU and the co-processor is a digital signal processor DSP.
13. A processor for fingerprint recognition, the processor being applied to an electronic device, the electronic device comprising an optical fingerprint recognition system arranged below a display screen, the optical fingerprint recognition system comprising an optical fingerprint sensor for capturing a fingerprint image of a finger above the display screen, the processor being a co-processor comprising:
the processing unit is used for processing the fingerprint image to obtain first characteristic data of the fingerprint image;
the communication unit is used for sending the first characteristic data of the fingerprint image to the main processor, wherein the first characteristic data of the fingerprint image are used for fingerprint identification by the main processor;
the processing unit is specifically configured to:
processing the fingerprint image in parallel to obtain first characteristic data of the fingerprint image in the process of processing the fingerprint image by the main processor to obtain second characteristic data of the fingerprint image and carrying out fingerprint identification according to the second characteristic data of the fingerprint image,
the first characteristic data of the fingerprint image is used for fingerprint identification when the main processor fails fingerprint identification according to the second characteristic data of the fingerprint image;
The second characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a rough mode, and the first characteristic data of the fingerprint image comprise characteristic data of the fingerprint image which are extracted in a fine mode.
14. The processor of claim 13, wherein the processing unit is further configured to:
if the fingerprint identification of the main processor is failed according to the first characteristic data of the fingerprint image, processing the other fingerprint image of the finger to obtain the first characteristic data of the other fingerprint image;
the communication unit is further configured to:
and sending the first characteristic data of the other fingerprint image to the main processor, wherein the first characteristic data of the other fingerprint image is used for fingerprint identification by the main processor.
15. The processor according to claim 13 or 14, wherein the host processor and the coprocessor are connected by an application programming interface API.
16. The processor according to claim 13 or 14, wherein the main processor is a central processing unit CPU and the co-processor is a digital signal processor DSP.
17. An electronic device comprising a main processor according to any of claims 9 to 12, and a co-processor according to any of claims 13 to 16.
18. The electronic device of claim 17, further comprising a fingerprint sensor for capturing a 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 CN109863505A (en) | 2019-06-07 |
| CN109863505B true 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) |
Families Citing this family (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 |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9135646B2 (en) * | 2013-10-09 | 2015-09-15 | Ricoh Company, Ltd. | Associating advertising content with a channel |
| US10169637B2 (en) * | 2015-10-19 | 2019-01-01 | Qualcomm Incorporated | On-screen optical fingerprint capture for user authentication |
| CN106022069B (en) * | 2016-05-30 | 2019-08-20 | Oppo广东移动通信有限公司 | A kind of solution lock 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 |
| CN107066961B (en) * | 2017-03-31 | 2019-08-09 | 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 |
| CN107545166B (en) * | 2017-08-24 | 2021-04-16 | Oppo广东移动通信有限公司 | Biometric identification method and related product |
| 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 WO PCT/CN2019/072655 patent/WO2020150891A1/en active Application Filing
- 2019-01-22 CN CN201980000111.3A patent/CN109863505B/en active Active
Non-Patent Citations (2)
| Title |
|---|
| 基于可重构微服务器的高能效指纹比对方法;钱磊 等;计算机研究与发展;全文 * |
| 基于嵌入式GPU的汗孔识别算法并行设计;刘义鹏 等;传感器与微系统;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109863505A (en) | 2019-06-07 |
| WO2020150891A1 (en) | 2020-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102324697B1 (en) | Biometric detection method and device, electronic device, computer readable storage medium | |
| WO2017202196A1 (en) | Method and device for fingerprint unlocking and user terminal | |
| JP2015513754A (en) | Face recognition method and device | |
| CN111626243B (en) | Mask face shielding identity recognition method and device and storage medium | |
| CN111274977A (en) | Multitask convolution neural network model, using method, device and storage medium | |
| JP5361524B2 (en) | Pattern recognition system and pattern recognition method | |
| CN111242273B (en) | Neural network model training method and electronic equipment | |
| JP2022524833A (en) | Fingerprint image enhancement, fingerprint recognition, application program startup method, fingerprint detection system, electronic devices and storage media | |
| CN112149570A (en) | Multi-person living body detection method and device, electronic equipment and storage medium | |
| JP2023015228A (en) | Information processing system, information processing device, information processing method, and computer program | |
| CN109863505B (en) | Fingerprint identification method, processor and electronic equipment | |
| CN110175500B (en) | Finger vein comparison method, device, computer equipment and storage medium | |
| CN113486738A (en) | Fingerprint identification method and device, electronic equipment and readable storage medium | |
| CN111144374B (en) | Facial expression recognition method and device, storage medium and electronic equipment | |
| CN110750673A (en) | Image processing method, device, equipment and storage medium | |
| US20220309782A1 (en) | Using smartphone camera and application to capture, analyze, and evaluate latent fingerprints in real-time | |
| Zhou et al. | Iris biometic processor enhanced module FPGA-based design | |
| CN106156568B (en) | Biological information identification module and electronic equipment | |
| CN111405175B (en) | Camera control method, device, computer equipment and storage medium | |
| Ouanan et al. | Implementation and optimization of face detection frameworkbased on OpenCV library on mobile platforms using Davinci’s technology | |
| CN114596638A (en) | Face living body detection method, device and storage medium | |
| CN113268769A (en) | Method, system and storage medium for processing biological characteristic data based on random confusion | |
| CN105139254A (en) | Earprint recognition-based bank remote identity authentication method and system | |
| CN110390353A (en) | A kind of biometric discrimination method and system based on image procossing | |
| CN113536953B (en) | Face recognition method and device, electronic equipment and 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 |