CN107832158A - Face identification method and device - Google Patents
Face identification method and device Download PDFInfo
- Publication number
- CN107832158A CN107832158A CN201710957341.1A CN201710957341A CN107832158A CN 107832158 A CN107832158 A CN 107832158A CN 201710957341 A CN201710957341 A CN 201710957341A CN 107832158 A CN107832158 A CN 107832158A
- Authority
- CN
- China
- Prior art keywords
- face
- task
- recognition
- image data
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/54—Interprogram communication
- G06F9/546—Message passing systems or structures, e.g. queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/52—Program synchronisation; Mutual exclusion, e.g. by means of semaphores
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/54—Indexing scheme relating to G06F9/54
- G06F2209/548—Queue
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the present invention provides a kind of face identification method and device.In one embodiment, the face identification method includes：Progress data are asked to encapsulate to obtain task data bag the multiple recognition of face received；The task data bag is cached pending to a task stack etc.；When any thread free time in thread pool, the task data bag in the storehouse is obtained；And recognition of face is carried out to the multiple image data in the task data bag based on default neural network model, obtain the face recognition result of the multiple image data.
Description
Technical field
The present invention relates to image processing field, in particular to a kind of face identification method and device.
Background technology
Image procossing is presented as substantial amounts of floating-point operation, therefore very consumption calculations performance in a computer.Should in industry
In, usually there is the figure station of specialty as processing unit, supported for certain class image application.
Just because of above-mentioned reason, the real-time cloud service of image procossing becomes more difficult：In the use of large-scale concurrent
Under the request of family, the computer of high loaded process originally is difficult that to have redundancy overhead be multi-subscriber dispatching, while the difficulty of scheduling is more
It is to be upgraded to hardware level from application layer.
It is impossible to meet extensive real-time face identifying processing to require for simple load balancing, because single computer can be normal
High load operation often is in, in order to meet requirement, it is necessary to which the computer of magnanimity supports as calculating, but this requirement is not
It is in the cards.
The content of the invention
In view of this, the purpose of the embodiment of the present invention is to provide a kind of face identification method and device.
A kind of face identification method provided in an embodiment of the present invention, the face identification method include：
Progress data are asked to encapsulate to obtain task data bag the multiple recognition of face received；
The task data bag is cached pending to a task stack etc.；
When any thread free time in thread pool, the task data bag in the storehouse is obtained；And
Recognition of face is carried out to the multiple image data in the task data bag based on default neural network model, obtained
To the face recognition result of the multiple image data.
The embodiment of the present invention also provides a kind of face identification device, and the face identification device includes：
Package module, for asking progress data to encapsulate to obtain task data bag the multiple recognition of face received；
Cache module, it is pending to a task stack etc. for the task data bag to be cached；
Acquisition module, for when any thread free time in thread pool, obtaining the task data bag in the storehouse；And
Identification module, for being entered based on default neural network model to the multiple image data in the task data bag
Row recognition of face, obtain the face recognition result of the multiple image data.
Compared with prior art, the face identification method and device of the embodiment of the present invention, by thread pool scheduling caching
The packet of queuing, the nerual network technique of recycling, the disposable identification for completing multiple images.It is in addition, existing without disposing
The recognition of face of batch is can be realized as in the case of the complicated spark frameworks of method.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by embodiment it is required use it is attached
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, therefore be not construed as pair
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, can also be according to this
A little accompanying drawings obtain other related accompanying drawings.
Fig. 1 is the block diagram for the electric terminal that present pre-ferred embodiments provide.
Fig. 2 is the flow chart for the face identification method that present pre-ferred embodiments provide.
Fig. 3 is the data flow schematic diagram for the electric terminal that present pre-ferred embodiments provide.
Fig. 4 is the step S104 for the face identification method that present pre-ferred embodiments provide detail flowchart.
Fig. 5 is the step S104 for the face identification method that another preferred embodiment of the present invention provides detail flowchart.
Fig. 6 is the high-level schematic functional block diagram for the face identification device that present pre-ferred embodiments provide.
Embodiment
Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.Generally exist
The component of the embodiment of the present invention described and illustrated in accompanying drawing can be configured to arrange and design with a variety of herein.Cause
This, the detailed description of the embodiments of the invention to providing in the accompanying drawings is not intended to limit claimed invention below
Scope, but it is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, those skilled in the art are not doing
The every other embodiment obtained on the premise of going out creative work, belongs to the scope of protection of the invention.
It should be noted that：Similar label and letter represents similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined, then it further need not be defined and explained in subsequent accompanying drawing in individual accompanying drawing.Meanwhile the present invention's
In description, term " first ", " second " etc. are only used for distinguishing description, and it is not intended that instruction or hint relative importance.
As shown in figure 1, it is the block diagram of an electric terminal 100.The electric terminal 100 includes face identification device
110th, memory 111, storage control 112, processor 113, Peripheral Interface 114, input-output unit 115, display unit
116.It will appreciated by the skilled person that the structure shown in Fig. 1 is only to illustrate, it is not to the knot of electric terminal 100
It is configured to limit.For example, electric terminal 100 may also include more either less components than shown in Fig. 1 or have and figure
Different configuration shown in 1.Electric terminal 100 described in the present embodiment can be personal computer, image processing server or
Mobile electronic device etc. has the computing device of image-capable.
The memory 111, storage control 112, processor 113, Peripheral Interface 114, input-output unit 115 and aobvious
Show that 116 each element of unit is directly or indirectly electrically connected between each other, to realize the transmission of data or interaction.For example, these
Element can be realized by one or more communication bus or signal wire be electrically connected between each other.The face identification device 110
The electricity can be stored in the memory 111 or be solidificated in including at least one in the form of software or firmware (Firmware)
Software function module in the operating system (Operating System, OS) of sub- terminal 100.The processor 113 is used to hold
The executable module stored in line storage, such as the software function module or computer that the face identification device 110 includes
Program.
Wherein, the memory 111 may be, but not limited to, random access memory (Random Access
Memory, RAM), read-only storage (Read Only Memory, ROM), programmable read only memory (Programmable
Read-Only Memory, PROM), erasable read-only memory (Erasable Programmable Read-Only
Memory, EPROM), electricallyerasable ROM (EEROM) (Electric Erasable Programmable Read-Only
Memory, EEPROM) etc..Wherein, memory 111 is used for storage program, the processor 113 after execute instruction is received,
Described program is performed, the method performed by electric terminal 100 that the process that any embodiment of the embodiment of the present invention discloses defines can
To be realized applied in processor 113, or by processor 113.
The processor 113 is probably a kind of IC chip, has the disposal ability of signal.Above-mentioned processor
113 can be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), network processes
Device (Network Processor, abbreviation NP) etc.；It can also be digital signal processor (DSP), application specific integrated circuit
(ASIC), field programmable gate array (FPGA) either other PLDs, discrete gate or transistor logic,
Discrete hardware components.It can realize or perform disclosed each method, step and the logic diagram in the embodiment of the present invention.It is general
Processor can be microprocessor or the processor can also be any conventional processor etc..
Various input/output devices are coupled to processor 113 and memory 111 by the Peripheral Interface 114.At some
In embodiment, Peripheral Interface 114, processor 113 and storage control 112 can be realized in one single chip.Other one
In a little examples, they can be realized by independent chip respectively.
The input-output unit 115 is used to be supplied to user input data.The input-output unit 115 can be,
But it is not limited to, mouse and keyboard etc..
The display unit 116 provided between the electric terminal 100 and user an interactive interface (such as user behaviour
Make interface) or for display image data give user reference.In the present embodiment, the display unit can be liquid crystal display
Or touch control display.If touch control display, it can be the capacitance type touch control screen or resistance for supporting single-point and multi-point touch operation
Formula touch screen etc..Single-point and multi-point touch operation is supported to refer to that touch control display can sense on the touch control display one
Or multiple opening positions touch control operation with caused by, and the touch control operation that this is sensed transfers to processor to be calculated and located
Reason.
In order to realize the recognition of face request in large-scale concurrent, in the prior art, closest solution is
Spark Stream Processings.The spark Stream Processings are the frameworks by complete set, and hardware bottom layer adapter is come, and then will
Streaming computing resolves into a series of short and small batch processing jobs, is calculated by specifically handling engine.Wherein, the number of each link
According to RDD (Resilient Distributed Dataset) form is all converted into, RDD is become into intermediate result by operation and protected
Exist in internal memory.Whole streaming computing can be overlapped according to the demand of business to the result of centre, or outside is arrived in storage
Equipment.Fig. 2 shows Spark Streaming whole flow process.
Prior art is a kind of very effective real-time processing framework.But inventor is carried out carefully to above-mentioned processing framework
Research, the recognition of face request of above-mentioned processing framework processing batch, that is, for extensive real-time face identification mission, still
The defects of in the presence of at least two aspects：
1. batch processing Define defects：Recognition of face is more by a variety of pretreatments, Face datection, feature extraction, Characteristic Contrast etc.
Individual algorithm composition, according to whetheing there is face, face how many algorithm combination have certain change, and algorithm complex and differ, complete
Last synthetic determination is remake after being handled more than.Therefore, it is necessary to define multiple Stream Processings, finally also once to integrate and sentence
Fixed, multiple parallel reduces the efficiency of Stream Processing.
2. algorithm migrates defect：In order that with Stream Processing, it is necessary to use scala multiple programmings.And false distinguishing algorithm often relates to
And complicated pattern-recognition, substantial amounts of bottom computing is difficult by Small Groups complete independently.Moved for the bottom of image machine learning
Move, and parameter learning is even more to be difficult to complete.
The application can be realized as the recognition of face of batch in the case of without using complicated spark frameworks.Utilize nerve
The structural particularity of network, without using complicated spark frameworks, the backstage design for changing batch images processing can be realized as effectively
Image procossing.For example, in primary network computing, while calculate multiple image data (a collection of view data).In conjunction with task
Fractionation mechanism is gentle to deposit mechanism so that backstage oncurrent processing ability greatly increases, and also neutral net is fully utilized.Under
Face is described in detail by several specifically embodiments.
Referring to Fig. 2, it is the recognition of face side for being applied to the electric terminal shown in Fig. 1 that present pre-ferred embodiments provide
The flow chart of method.The idiographic flow shown in Fig. 2 will be described in detail below.
Step S101, progress data are asked to encapsulate to obtain task data bag the multiple recognition of face received.
In the present embodiment, it may include multiple image data in each recognition of face request, described image data can be figure
Piece, image etc. have the data of picture.
In one embodiment, step S101 includes：Face image data during the recognition of face received is asked
Data are carried out to encapsulate to obtain task data bag；Task label is added for the task data bag, the task label includes task
Content and task amount.
In one embodiment, as shown in figure 3, step S101 can perform in the application layer of electric terminal.The electricity
Sub- terminal is when carrying out data encapsulation, it would be desirable to which the task and task amount of execution are bundled in the packet header of the packet.One
In individual example, based on recognition of face request need task to be processed may include the face in detection image, face normalization and
Face characteristic extracts, then the task definition of the task label in the packet header of the packet may include face in detection image,
Face normalization and face characteristic extraction.For another example the task definition of task label can also include face characteristic matching etc..It is described
Also include task amount in task label, the task amount is used to mark the data to include needing the facial image handled.For example,
It may include multiple recognition of face request in each packet, multiple image data can be included in each recognition of face request, often
It may include multiple face figures in width view data.Have it is above-mentioned understand, each packet include multiple face figures need to handle, it is described
The quantity of face figure corresponds to task amount.In present embodiment, by adding task label in the packet, increase thread pool
Thread parallel processing capability, raising efficiency, complete the task of recognition of face.
Step S102, the task data bag is cached pending to a task stack etc..
In one embodiment, the storehouse can be a heap, and the heap meets that queue is preferential, first in first out.Example
Such as, data A is stored into the heap prior to data B, then data A is taken out from the heap prior to data B and handled.
In another embodiment, the storehouse can be a stack, and the stack meets first-in last-out.For example, data A
It is stored into prior to data B in the heap, then data A is taken out from the heap after data B and handled.
In an example, as shown in figure 3, the step S102 can also be performed in the application layer of electric terminal.
Step S103, when any thread free time in thread pool, obtain the task data bag in the storehouse.
Step S104, pedestrian is entered to the multiple image data in the task data bag based on default neural network model
Face identifies, obtains the face recognition result of the multiple image data.
In the present embodiment, the face recognition result can be the face characteristic for the face figure that the packet includes；
The face recognition result can be whether the face characteristic of the face figure in the packet matches with the compare feature to prestore
Result.
In one embodiment, as shown in figure 3, the step S104 can be held in the computation layer of the electric terminal
OK.
Face identification method according to embodiments of the present invention, pass through the packet being lined up in thread pool scheduling caching, then profit
Nerual network technique, the disposable identification for completing multiple images.In addition, the complicated spark framves without disposing existing method
The recognition of face of batch is can be realized as in the case of structure.
In one embodiment, as shown in figure 4, step S104 may include：Step S1041- steps S1044.
Step S1041, the task data bag is decapsulated to obtain multiple image data.
Step S1042, detect the facial image in the multiple image data.
In the present embodiment, described image data can be picture, image etc..In an example, described image data are
During image, the picture of the image can be intercepted, yet further detects the facial image in the picture of interception.
In the present embodiment, the electric terminal can use the detection methods such as dlib, OpenCV detection described image data
In facial image.
Step S1043, the facial image is corrected.
In the present embodiment, the facial image come out from the multiple image Data Detection is probably inclined, or not
Clearly image.For example, when the facial image tilts, it is suspicious to be corrected facial image by affine transformation.Example again
Such as, when the facial image is unintelligible, by improving the processing such as picture contrast, denoising image can be made apparent.
Step S1044, the facial image after correction is inputted in the model of neutral net and is trained to obtain face characteristic,
The face recognition result that the recognition of face as corresponding to of face characteristic corresponding to each facial image is asked.
In the present embodiment, made by the vector that the neutral net is trained to obtain to last in convolutional layer layer output
It is vectorial for face, and using face vector as the face characteristic.
In the present embodiment, the face vector can represent the feature of a face, the same person shape of face into face to
The Euclidean distance of amount is small relative to the Euclidean distance of face vector corresponding to different facial images.Two face vectors
Euclidean distance it is smaller represent face vector corresponding to facial image more like.
The face characteristic in view data can be identified in bulk by above-mentioned steps, in addition, the method in the present embodiment,
Without disposing the complicated spark frameworks of existing method, processing batch can be realized under the various systems such as windows, linux
The recognition of face request of amount.
In the present embodiment, methods described also includes：Step S105, after the face recognition result is obtained, start a line
Journey stores the face recognition result into a database.One line can be notified after result is identified by the above method
The face recognition result is stored in database by journey, prevents thread is shelved from causing to block.
In one embodiment, as shown in figure 5, step S104 may include：Step S1045- steps S1047.
Step S1045, detect multiple facial images in the multiple image data.
Step S1046, the multiple facial image is inputted in the model of neutral net and is trained to obtain multigroup face
Feature.
Step S1047, multigroup face characteristic is matched with default compare feature respectively, according to matching result
Obtain face recognition result corresponding to each facial image.
In an application scenarios, the method in the present embodiment can be used for safety to deploy to ensure effective monitoring and control of illegal activities, and safety management is with taking precautions against.At one
It can include acquisition terminal, data processing terminal and database in system of deploying to ensure effective monitoring and control of illegal activities safely.The acquisition terminal is public for gathering
The image in place, and the image of collection is sent to the data processing terminal.The data processing terminal performs the present embodiment
In method in each step.
In an example, the compare feature can steal face vector corresponding to the image of suspect.Pass through meter
The vectorial face vector corresponding with the image of the pilferage suspect of face corresponding to facial image in the image collected
Euclidean distance, when the distance is less than limit value, represent the facial image with it is described pilferage suspect images match
Success.In this application scene, if the image that the recognition result that the data processing terminal obtains is collection includes the control
Corresponding to feature during facial image, the result for recognizing suspect, further, the acquisition terminal are sent to the acquisition terminal
Warning message can also be sent to designated terminal, the designated terminal can be the mobile device of the staff such as police, security personnel.
In another example, the method in the present embodiment is used in the places such as market, bank, office building, the compare feature
Can be face vector corresponding to the VIP member of the market or bank, the compare feature can also be the work people of office building
Face vector corresponding to member.The acquisition terminal can be arranged on market, bank, the doorway of office building, when multigroup face
For feature when it fails to match with default compare feature respectively, the acquisition terminal sends warning message, prompts to there may be non-meeting
Member or non-working person enter market, bank or office building.
Method in the present embodiment is applied in several scenes, can be with the case where not needing complicated spark frameworks
Realize that quickly and efficiently face is matched, and the condition of favourable recognition of face is provided for public places such as market, campuses.
Referring to Fig. 6, it is the functional module of the face identification device 110 shown in Fig. 1 of present pre-ferred embodiments offer
Schematic diagram.Modules in face identification device 110 and unit in the present embodiment are used to perform in above method embodiment
Each step.The face identification device 110 includes package module 1101, cache module 1102, acquisition module 1103 and known
Other module 1104.
The package module 1101, for asking progress data to encapsulate to obtain number of tasks the multiple recognition of face received
According to bag.
The cache module 1102, it is pending to a task stack etc. for the task data bag to be cached.
The acquisition module 1103, for when any thread free time in thread pool, obtaining the number of tasks in the storehouse
According to bag.
The identification module 1104, for based on default neural network model to several figures in the task data bag
As data progress recognition of face, the face recognition result of the multiple image data is obtained.
In one embodiment, the identification module 1104 includes：Decapsulation unit, detection unit, correction unit and
Feature identification unit.
The decapsulation unit, for being decapsulated to obtain multiple image data the task data bag.
The detection unit, for detecting the facial image in the multiple image data.
The correction unit, for the facial image to be corrected.
The feature identification unit, for will be trained in the model of the facial image input neutral net after correction
To face characteristic, the face recognition result that the recognition of face as corresponding to of face characteristic corresponding to each facial image is asked.
In one embodiment, the face identification device 110 also includes：Memory module 1105, for described when obtaining
After face recognition result, start a thread and store the face recognition result into a database.
In one embodiment, the package module 1101 includes：Data packaging unit and label adding device.
The data packaging unit, data envelope is carried out for the face image data during the recognition of face received is asked
Dress obtains task data bag.
The label adding device, for adding task label for the task data bag, the task label includes appointing
Content of being engaged in and task amount.
In one embodiment, the identification module 1104 includes：Detection unit, training unit and matching unit.
The detection unit, for detecting multiple facial images in the multiple image data.
The training unit is more for that will be trained to obtain in the model of the multiple facial image input neutral net
Group face characteristic.
The matching unit, for multigroup face characteristic to be matched with default compare feature respectively, according to
Matching result obtains face recognition result corresponding to each facial image.
Other details on the device in the present embodiment can further refer to the description in above method embodiment,
It will not be repeated here.
Face identification device according to embodiments of the present invention, pass through the packet being lined up in thread pool scheduling caching, then profit
Nerual network technique, the disposable identification for completing multiple images.In addition, the complicated spark framves without disposing existing method
The recognition of face of batch is can be realized as in the case of structure.
In several embodiments provided herein, it should be understood that disclosed apparatus and method, can also pass through
Other modes are realized.Device embodiment described above is only schematical, for example, flow chart and block diagram in accompanying drawing
Show the device of multiple embodiments according to the present invention, method and computer program product architectural framework in the cards,
Function and operation.At this point, each square frame in flow chart or block diagram can represent the one of a module, program segment or code
Part, a part for the module, program segment or code include one or more and are used to realize holding for defined logic function
Row instruction.It should also be noted that at some as in the implementation replaced, the function that is marked in square frame can also with different from
The order marked in accompanying drawing occurs.For example, two continuous square frames can essentially perform substantially in parallel, they are sometimes
It can perform in the opposite order, this is depending on involved function.It is it is also noted that every in block diagram and/or flow chart
The combination of individual square frame and block diagram and/or the square frame in flow chart, function or the special base of action as defined in performing can be used
Realize, or can be realized with the combination of specialized hardware and computer instruction in the system of hardware.
In addition, each functional module in each embodiment of the present invention can integrate to form an independent portion
Point or modules individualism, can also two or more modules be integrated to form an independent part.
If the function is realized in the form of software function module and is used as independent production marketing or in use, can be with
It is stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially in other words
The part to be contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are causing a computer equipment (can be
People's computer, server, or network equipment etc.) perform all or part of step of each embodiment methods described of the present invention.
And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.Need
Illustrate, herein, such as first and second or the like relational terms be used merely to by an entity or operation with
Another entity or operation make a distinction, and not necessarily require or imply between these entities or operation any this reality be present
The relation or order on border.Moreover, term " comprising ", "comprising" or its any other variant are intended to the bag of nonexcludability
Contain, so that process, method, article or equipment including a series of elements not only include those key elements, but also including
The other element being not expressly set out, or also include for this process, method, article or the intrinsic key element of equipment.
In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including the key element
Process, method, other identical element also be present in article or equipment.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies
Change, equivalent substitution, improvement etc., should be included in the scope of the protection.It should be noted that：Similar label and letter exists
Similar terms is represented in following accompanying drawing, therefore, once being defined in a certain Xiang Yi accompanying drawing, is then not required in subsequent accompanying drawing
It is further defined and explained.
The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all be contained
Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.
Claims (10)
1. a kind of face identification method, it is characterised in that the face identification method includes：
Progress data are asked to encapsulate to obtain task data bag the multiple recognition of face received；
The task data bag is cached pending to a task stack etc.；
When any thread free time in thread pool, the task data bag in the storehouse is obtained；And
Recognition of face is carried out to the multiple image data in the task data bag based on default neural network model, obtains institute
State the face recognition result of multiple image data.
2. face identification method as claimed in claim 1, it is characterised in that described to be based on default neural network model to institute
The multiple image data stated in task data bag carry out recognition of face, obtain the face recognition result of the multiple image data
Step includes：
The task data bag is decapsulated to obtain multiple image data；
Detect the facial image in the multiple image data；
The facial image is corrected；
It will be trained to obtain face characteristic in the model of the facial image input neutral net after correction, by each facial image
The face recognition result that the recognition of face as corresponding to of corresponding face characteristic is asked.
3. face identification method as claimed in claim 1, it is characterised in that methods described also includes：
After the face recognition result is obtained, start a thread and store the face recognition result into a database.
4. face identification method as claimed in claim 1, it is characterised in that described to ask the multiple recognition of face received
Data are carried out to encapsulate to include the step of obtaining task data bag：
Face image data during the recognition of face received is asked carries out data and encapsulates to obtain task data bag；
Task label is added for the task data bag, the task label includes task definition and task amount.
5. the face identification method as described in claim 1-4 any one, it is characterised in that described to be based on default nerve net
Network model carries out recognition of face to the multiple image data in the task data bag, obtains the face of the multiple image data
The step of recognition result, includes：
Detect multiple facial images in the multiple image data；
It will be trained to obtain multigroup face characteristic in the model of the multiple facial image input neutral net；
Multigroup face characteristic is matched with default compare feature respectively, each face figure is obtained according to matching result
The face recognition result as corresponding to.
6. a kind of face identification device, it is characterised in that the face identification device includes：
Package module, for asking progress data to encapsulate to obtain task data bag the multiple recognition of face received；
Cache module, it is pending to a task stack etc. for the task data bag to be cached；
Acquisition module, for when any thread free time in thread pool, obtaining the task data bag in the storehouse；And
Identification module, for entering pedestrian to the multiple image data in the task data bag based on default neural network model
Face identifies, obtains the face recognition result of the multiple image data.
7. face identification device as claimed in claim 6, it is characterised in that the identification module includes：
Decapsulation unit, for being decapsulated to obtain multiple image data the task data bag；
Detection unit, for detecting the facial image in the multiple image data；
Unit is corrected, for the facial image to be corrected；
Feature identification unit, for that will be trained to obtain face spy in the model of the facial image input neutral net after correction
Sign, the face recognition result that the recognition of face as corresponding to of face characteristic corresponding to each facial image is asked.
8. face identification device as claimed in claim 6, it is characterised in that described device also includes：
Memory module, the face recognition result storage is arrived for after the face recognition result is obtained, starting a thread
In one database.
9. face identification device as claimed in claim 6, it is characterised in that the package module includes：
Data packaging unit, carry out data for the face image data during the recognition of face received is asked and encapsulate and must take office
Business packet；
Label adding device, for for the task data bag add task label, the task label include task definition and
Task amount.
10. the face identification device as described in claim 6-9 any one, it is characterised in that the identification module includes：
Detection unit, for detecting multiple facial images in the multiple image data；
Training unit, for that will be trained to obtain multigroup face spy in the model of the multiple facial image input neutral net
Sign；
Matching unit, for multigroup face characteristic to be matched with default compare feature respectively, according to matching result
Obtain face recognition result corresponding to each facial image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710957341.1A CN107832158A (en) | 2017-10-16 | 2017-10-16 | Face identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710957341.1A CN107832158A (en) | 2017-10-16 | 2017-10-16 | Face identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107832158A true CN107832158A (en) | 2018-03-23 |
Family
ID=61648198
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710957341.1A Pending CN107832158A (en) | 2017-10-16 | 2017-10-16 | Face identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107832158A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108875328A (en) * | 2018-05-30 | 2018-11-23 | 腾讯科技（深圳）有限公司 | A kind of auth method, device and storage medium |
| CN109086132A (en) * | 2018-06-26 | 2018-12-25 | 深圳市买买提信息科技有限公司 | A kind of recognition of face task balance call method, device and terminal device |
| CN109388501A (en) * | 2018-08-31 | 2019-02-26 | 平安科技（深圳）有限公司 | Communication matching process, device, equipment and medium based on recognition of face request |
| CN110276331A (en) * | 2019-06-28 | 2019-09-24 | 四川长虹电器股份有限公司 | The inspection modification method of Identification of Images result |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020076088A1 (en) * | 2000-12-15 | 2002-06-20 | Kun-Cheng Tsai | Method of multi-level facial image recognition and system using the same |
| CN102289898A (en) * | 2011-09-08 | 2011-12-21 | 广州广电运通金融电子股份有限公司 | Automatic emergency hedge system for self-service transaction |
| WO2014138288A1 (en) * | 2013-03-05 | 2014-09-12 | Tyco Fire & Security Gmbh | Facial recognition controlled access areas utilizing electronic article surveillance (east) system |
| CN106209682A (en) * | 2016-07-08 | 2016-12-07 | 北京百度网讯科技有限公司 | Business scheduling method, device and system |
| CN106447625A (en) * | 2016-09-05 | 2017-02-22 | 北京中科奥森数据科技有限公司 | Facial image series-based attribute identification method and device |
| CN106453834A (en) * | 2016-09-07 | 2017-02-22 | 努比亚技术有限公司 | Mobile terminal and camera shooting method |
| CN106504540A (en) * | 2016-12-12 | 2017-03-15 | 浙江宇视科技有限公司 | A kind of analysis method of information of vehicles and device |
| CN106547890A (en) * | 2016-11-04 | 2017-03-29 | 深圳云天励飞技术有限公司 | Quick clustering preprocess method in large nuber of images characteristic vector |
| CN106548151A (en) * | 2016-11-03 | 2017-03-29 | 北京光年无限科技有限公司 | Towards the target analyte detection track identification method and system of intelligent robot |
-
2017
- 2017-10-16 CN CN201710957341.1A patent/CN107832158A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020076088A1 (en) * | 2000-12-15 | 2002-06-20 | Kun-Cheng Tsai | Method of multi-level facial image recognition and system using the same |
| CN102289898A (en) * | 2011-09-08 | 2011-12-21 | 广州广电运通金融电子股份有限公司 | Automatic emergency hedge system for self-service transaction |
| WO2014138288A1 (en) * | 2013-03-05 | 2014-09-12 | Tyco Fire & Security Gmbh | Facial recognition controlled access areas utilizing electronic article surveillance (east) system |
| CN106209682A (en) * | 2016-07-08 | 2016-12-07 | 北京百度网讯科技有限公司 | Business scheduling method, device and system |
| CN106447625A (en) * | 2016-09-05 | 2017-02-22 | 北京中科奥森数据科技有限公司 | Facial image series-based attribute identification method and device |
| CN106453834A (en) * | 2016-09-07 | 2017-02-22 | 努比亚技术有限公司 | Mobile terminal and camera shooting method |
| CN106548151A (en) * | 2016-11-03 | 2017-03-29 | 北京光年无限科技有限公司 | Towards the target analyte detection track identification method and system of intelligent robot |
| CN106547890A (en) * | 2016-11-04 | 2017-03-29 | 深圳云天励飞技术有限公司 | Quick clustering preprocess method in large nuber of images characteristic vector |
| CN106504540A (en) * | 2016-12-12 | 2017-03-15 | 浙江宇视科技有限公司 | A kind of analysis method of information of vehicles and device |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108875328A (en) * | 2018-05-30 | 2018-11-23 | 腾讯科技（深圳）有限公司 | A kind of auth method, device and storage medium |
| CN109086132A (en) * | 2018-06-26 | 2018-12-25 | 深圳市买买提信息科技有限公司 | A kind of recognition of face task balance call method, device and terminal device |
| CN109388501A (en) * | 2018-08-31 | 2019-02-26 | 平安科技（深圳）有限公司 | Communication matching process, device, equipment and medium based on recognition of face request |
| CN109388501B (en) * | 2018-08-31 | 2024-03-05 | 平安科技（深圳）有限公司 | Communication matching method, device, equipment and medium based on face recognition request |
| CN110276331A (en) * | 2019-06-28 | 2019-09-24 | 四川长虹电器股份有限公司 | The inspection modification method of Identification of Images result |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107832158A (en) | Face identification method and device | |
| US10956753B2 (en) | Image processing system and image processing method | |
| EP3540635B1 (en) | Method for identifying an object within an image and mobile device for executing the method | |
| CN111666857A (en) | Human behavior recognition method and device based on environment semantic understanding and storage medium | |
| CN112052781A (en) | Feature extraction model training method, face recognition device, face recognition equipment and medium | |
| WO2019178501A1 (en) | Fraudulent transaction identification method and apparatus, server, and storage medium | |
| CN109522790A (en) | Human body attribute recognition approach, device, storage medium and electronic equipment | |
| CN109903474A (en) | A kind of intelligence based on recognition of face opens cabinet method and device | |
| CN108961387A (en) | A kind of display methods and terminal device of house virtual three-dimensional model | |
| US20190378171A1 (en) | Targeted advertisement system | |
| CN108389230A (en) | Refrigerator capacity automatic testing method, system, equipment and storage medium | |
| US11348415B2 (en) | Cognitive automation platform for providing enhanced automated teller machine (ATM) security | |
| CN109002776B (en) | Face recognition method, system, computer device and computer-readable storage medium | |
| CN114187561A (en) | Abnormal behavior identification method and device, terminal equipment and storage medium | |
| CN111931707A (en) | Face image prediction method, device, equipment and medium based on countercheck patch | |
| CN113902574A (en) | Protocol data processing method, device, computer equipment and storage medium | |
| US20230123879A1 (en) | Method and apparatus for positioning express parcel | |
| CN117152844A (en) | High-integrity worker construction attitude detection method and system based on computer vision | |
| CN114758286B (en) | Intelligent edge safety monitoring method and device based on work ticket event | |
| CN111814613A (en) | Face recognition method, face recognition equipment and computer readable storage medium | |
| Adaickalam et al. | A Vibrant Multiple Object Detection Using Machine Learning Techniques | |
| CN112419637A (en) | Security image data processing method and device | |
| CN110866500A (en) | Face detection alignment system, method, device, platform, mobile terminal and storage medium | |
| CN202404694U (en) | Adaptive disturbance signal identification module of distributing type optical fiber sensing application system | |
| CN113762221B (en) | Human body detection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180323 |