CN109558839A - Adaptive face identification method and the equipment and system for realizing this method - Google Patents
Adaptive face identification method and the equipment and system for realizing this method Download PDFInfo
- Publication number
- CN109558839A CN109558839A CN201811444147.4A CN201811444147A CN109558839A CN 109558839 A CN109558839 A CN 109558839A CN 201811444147 A CN201811444147 A CN 201811444147A CN 109558839 A CN109558839 A CN 109558839A
- Authority
- CN
- China
- Prior art keywords
- face
- recognition
- facial image
- feature
- probability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 32
- 230000001815 facial effect Effects 0.000 claims abstract description 90
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 230000005540 biological transmission Effects 0.000 claims description 37
- 238000011144 upstream manufacturing Methods 0.000 claims description 4
- 238000012512 characterization method Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 abstract description 18
- 230000004044 response Effects 0.000 abstract description 3
- 238000012545 processing Methods 0.000 description 12
- 238000005286 illumination Methods 0.000 description 11
- 238000005259 measurement Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 7
- 238000003860 storage Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000001914 filtration Methods 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000003708 edge detection Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The present invention discloses adaptive face identification method, belongs to technical field of face recognition, comprising steps of acquisition facial image, executes feature extraction；According to the feature that the feature extraction is extracted, the probability of success of recognition of face is judged；It is preferential to upload the highest image of the recognition of face probability of success according to the probability of success of the recognition of face, it is used for recognition of face.The present invention can be under the conditions of communication bandwidth be limited, the facial image for being most suitable for executing recognition of face is found out to upload, and recognition of face is executed for the facial image, to realize that completing necessary data in the shortest time transmits and improve the accuracy identified for the first time, so as to improve the rapid response speed and subsequent applications of recognition of face and the user experience of service.
Description
Technical field
The present invention relates to the equipment and system of a kind of adaptive face identification method and realization this method, belong to recognition of face
Technical field.
Background technique
Wireless communication is widely used in the whole world, greatly facilitates the communication between people.Wireless communication can
To provide various services, including voice and video call, mobile flow medium play, phone network game, video real-time live broadcast, far
Range monitoring, industrial automatic control etc..The main difficulty of these business services is how time delay to be reduced in practical radio communication environment,
To promote the satisfaction of user service.
Recognition of face is of wide application, and the monitoring of airport, public domain, practical application are logged on to from gate inhibition, equipment
Static Human Face identification and dynamic human face identification can be divided into again.Dynamic human face identification refers to that identified people is in moving condition or step
Under the non-mated conditions such as row, its facial image is acquired, carries out recognition of face, it at this moment can be because of illumination, posture, camera focusing etc.
Problem causes accuracy of face identification to significantly reduce, and the problem under wireless environment is then even more serious.
Face identification system mainly includes man face image acquiring and detection, facial image pre-process, facial image feature mentions
It takes and matches and identification.The feature templates stored in the characteristic of the facial image of extraction and database scan for matching
When, database is bigger, then searches for that matched workload is bigger, and matching result is more accurate.In certain applications occasion, such as mobile machine
When artificial serve, need to realize recognition of face, but the usual processing capacity of mobile device and data by mobile device
Amount of storage is all little, is difficult to meet the requirement of dynamic human face identification, needs the method that can quickly and accurately identify face,
To be obviously improved user satisfaction.
Summary of the invention
In view of the above existing problems in the prior art, the present invention provides under a kind of wireless or bandwidth-constrained environment quickly from
It adapts to face identification method and realizes the equipment and system of this method.
To achieve the goals above, the adaptive face identification method that the present invention uses, comprising steps of
Facial image is acquired, feature extraction is executed；
According to the feature that the feature extraction is extracted, the probability of success of recognition of face is judged；
It is preferential to upload the highest image of the recognition of face probability of success according to the probability of success of the recognition of face, it is used for people
Face identification.
As an improvement, being compensated to the unconspicuous disadvantage of feature and/or in the feature extraction to the people acquired
Face image is handled.
As an improvement, utilizing the feature and face completed in advance according to the feature that the feature extraction is extracted
The relational model for identifying the probability of success, judges the probability of success of recognition of face.
As a further improvement, the relational model of the feature completed in advance and the recognition of face probability of success,
For described the case where compensating to the unconspicuous disadvantage of feature and/or handling the facial image acquired, need
Relational model is modified.
The present invention also provides another adaptive face identification methods, comprising steps of
Obtain the prediction to uplink transmission rate；
Based on the prediction of the uplink transmission rate, the feature of optimal facial image is selected；
According to the feature of the optimal facial image, facial image is acquired；
Upload the facial image with the optimal facial image feature of the acquisition；
Recognition of face is executed to the facial image of the upload.
The present invention also provides a kind of adaptive man face image acquiring equipment for realizing adaptive face identification method, this sets
It is standby to include:
Uplink transmission rate obtains module, obtains the prediction result of uplink transmission rate；
Face characteristic extraction module, for obtaining the relevant key feature of recognition of face, or the configuration adaptive people
Face acquires the key characterization parameter that equipment executes man face image acquiring；
Face characteristic enhances module, executes image procossing to the facial image of acquisition；
Recognition of face probability of success prediction module determines how under the limitation of the uplink transmission rate and chooses face knowledge
Other key feature completes recognition of face to realize in most short time-delay, or is sentenced according to the relevant key feature of the recognition of face
Read the probability of success of recognition of face；
Recognition of face image transmitting sorting module, according to the probability of success of the recognition of face to the face figure of the acquisition
Sequence as executing transmission.
Finally, the present invention also provides the systems for realizing adaptive face identification method, comprising:
Adaptive face acquires equipment, for acquiring facial image and uploading；
Base station, for receiving facial image；
Face recognition device executes recognition of face；
The adaptive face acquisition equipment acquires facial image, and the prediction result of upstream bandwidth is obtained from base station, is being needed
The facial image is uploaded into face recognition device in the time delay to be guaranteed, and recognition of face is completed by face recognition device.
Compared with prior art, the present invention can find out under the conditions of communication bandwidth is limited and be most suitable for executing recognition of face
Facial image upload, and for the facial image execute recognition of face, thus realize complete in the shortest time it is necessary
The accuracy identified for the first time is transmitted and improved to data, so as to improve recognition of face rapid response speed and subsequent applications and
The user experience of service.
Detailed description of the invention
Fig. 1 is quick self-adapted face identification method of the one kind of the embodiment of the present invention wirelessly or under bandwidth-constrained environment
Flow diagram；
Fig. 2 is the flow diagram of the quick self-adapted face identification method of another kind provided in an embodiment of the present invention；
Fig. 3 is a kind of structural schematic diagram of quick self-adapted man face image acquiring equipment provided in an embodiment of the present invention；
Fig. 4 be quick self-adapted man face image acquiring equipment provided in an embodiment of the present invention apply in wireless environments be
System schematic diagram.
Specific embodiment
In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention is carried out below further detailed
It describes in detail bright.However, it should be understood that the specific embodiments described herein are merely illustrative of the present invention, it is not limited to this hair
Bright range.
Unless otherwise defined, all technical terms and scientific terms used herein are led with technology of the invention is belonged to
The normally understood meaning of the technical staff in domain is identical, and term as used herein in the specification of the present invention is intended merely to retouch
State the purpose of specific embodiment, it is not intended that in the limitation present invention.
Referring to Fig. 1, for a kind of quick self-adapted people wirelessly or under bandwidth-constrained environment provided in an embodiment of the present invention
The flow diagram of face recognition method, method includes the following steps:
Step S101, remote equipment acquire facial image, execute the feature extraction of the facial image.
The remote equipment can be mobile device, or the communication connection of Bandwidth-Constrained is only existed with face recognition device
Man face image acquiring equipment, the communication connection of Bandwidth-Constrained refers to that narrow band communication connects, or is easy to happen the communication link of congestion
It connects, such as ADSL (Asymmetric Digital Subscriber Line, Asymmetrical Digital Subscriber Line) connection, shares
LAN (local area network, local area network) connection etc..Mobile device can be mobile phone, tablet computer, laptop,
Or mobile camera etc..Man face image acquiring equipment can be the camera for the gate that reaches a standard, the camera etc. of airport security.Face
Identification equipment can be the face recognition software in cloud, or be mounted with the cluster computer of face recognition software, server or
Other computer equipments.The remote equipment acquires facial image using man face image acquirings equipment such as cameras, due to differentiating
Rate is different, and the data volume size of facial image collected is also different.The remote equipment can be held based on the facial image
Row recognition of face for the first time, but since the processing capacity of remote equipment is limited, facial image characteristic that remote equipment is stored
Database it is also limited, it is easy to there is a phenomenon where identification mistake or recognition failures.Therefore need will be collected for remote equipment
Face image data uploads to the face recognition device, using face recognition device, as cloud powerful calculating ability and
Complete recognition of face work in complete data library.Even under the premise of remote equipment complete independently recognition of face work,
Recognition of face is executed using cloud or remote server, it helps verifying or school are carried out to the recognition result of the remote equipment
Standard avoids subsequent work mistake occur when providing service for user.
In order to improve the accuracy of recognition of face, the remote equipment while reducing the data volume of data transmission
Feature extraction, including environment illumination intensity, noise size, edge detection etc. can be executed to the facial image.In the spy
On the basis of sign is extracted, the unconspicuous disadvantage of feature is compensated, such as the situation inadequate for intensity of illumination, be can use
The light source of remote equipment executes light compensation including flash lamp etc., and continues to acquire facial image under conditions of compensation, simultaneously
It can also be handled for the facial image acquired, be obtained for example, by using processing methods such as greyscale transformation, geometric correction, filtering
Take more facial images.
Step S102 judges the success of recognition of face according to the feature that the feature extraction of the facial image is extracted
Probability.
For the acquisition or the feature extracted by multiple facial images that calculation processing obtains, according to feature
Concrete condition, judge the probability of success of recognition of face.Such as the intensity of illumination for image, mathematics can be completed in advance and built
Mould finds the correlativity of image irradiation intensity Yu recognition of face success rate, in this way for multiple described facial images, compares it
Intensity of illumination can substantially judge the probability of success size comparative situation that recognition of face is executed according to each image.For image
The same available correlativity with recognition of face success rate of the fog-level of resolution ratio or image.
Described multiple facial images obtained by calculation processing need needle since calculation processing can introduce some errors
Processing image is counted, the error modeling of calculation processing is completed, is judging according to the face obtained by calculation processing
When image executes the probability of success of recognition of face, the judging result of the probability of success is modified.
Step S103 is preferentially uploaded in the facial image according to the probability of success of the recognition of face, recognition of face at
The highest image of function probability is used for recognition of face.
According to the probability of success of the recognition of face, to the acquisition or the multiple facial images obtained by calculation processing
Sequence is executed, the maximum facial image of the recognition of face probability of success therein is preferentially uploaded.For speed up processing, the people
The probability of success of face identification can not have to complete accurately calculate, and the identification probability of success is divided into several sections, such as it is divided into 0~
10%, 10%~70%, 70%~100% 3 section.It can also be just for a feature of most critical, such as the illumination of image
Intensity carries out the estimation of the recognition of face probability of success, ignores the influence of other feature, as long as the same corresponding identification probability of success
Section where determining.When the recognition of face probability of success of multiple images is located at the same probability interval, these figures
As that can be uploaded in any order when uploading, to mitigate calculating pressure, speed up processing.
After the facial image uploads, recognition of face is completed in the face recognition device, recognition result can return
To the remote equipment, follow-up service is completed, such as provides guide service by mobile robot for guest or takes parcel services, or
Person's face recognition result is directly sent to next service point beyond the clouds, and reminding the service point in advance is that user prepares required thing
Product.
The above method can be applicable in for various wireless communications environments, equally applicable for narrowband wired communication environment,
To realize under general communication condition, reduces communication to the greatest extent and successfully complete the time delay of recognition of face, improve user's clothes
Business experience.
As it can be seen that quick self-adapted face of the one kind provided according to embodiments of the present invention wirelessly or under bandwidth-constrained environment is known
Other method can find out the facial image for being most suitable for executing recognition of face and upload and be directed under the conditions of communication bandwidth is limited
The facial image executes recognition of face, completes necessary data in the shortest time and transmit and improve to identify for the first time to realize
Accuracy, so as to improve the rapid response speed and subsequent applications of recognition of face and the user experience of service.
Referring to Fig. 2, the process for the quick self-adapted face identification method of another kind provided in an embodiment of the present invention is illustrated
Figure, method includes the following steps:
Step S201, remote equipment obtain the prediction to uplink transmission rate.
In order to which the facial image of suitable size is uploaded to face recognition device with rate appropriate, remote equipment is needed
The information of uplink bandwidth is obtained first.This can be provided by the face recognition device, can also be by uplink line
Some equipment in road, as base station provides.The face recognition device can be in the message for receiving the remote equipment transmission, packet
When including the facial image, receiving velocity completion is measured and recorded, executes filtering for the measurement result of the receiving velocity
Algorithm to obtain the prediction to following uplink transmission rate, and notifies the remote equipment.The filtering algorithm can be
To the simple average of the measurement result of the receiving velocity, it is flat to be also possible to the weighting to the measurement result of the receiving velocity
, i.e. va=w1*v (t-1)+w2*v (t-2)+w3*v (t-3)+...+wn*v (t-n), wherein va is described to following uplink
The prediction result of transmission rate, w1, w2 ..., wn are n weighted values, and n is random natural number ,+the wn=1 that meets w1+w2+ ..., and
V (t-1) is then the measurement result record numerical value at previous receiving velocity measurement moment, and other v (t-i) are similar, 1 < i < n, v (t-
N) be then n-th receiving velocity in the front measurement moment measurement result record numerical value.As w1=w2=...=wn=1/n when, institute
Result be simple average calculated value, when wn < ... when < w3 < w2 < w1 < 1, calculating resulting va then is with time characteristic of oblivion
Filter result.Above-mentioned calculating can also be executed by the equipment of some in upstream transmission line, generally setting by communication performance bottleneck
It is standby, as there is the base station equipment execution in the route of wireless transmission bottleneck, and timing or the request for answering remote equipment, will calculate
The prediction result to uplink bandwidth arrived notifies remote equipment.Or by the face recognition device by the receiving velocity
Measurement result notify the remote equipment, filtering algorithm is voluntarily executed by remote equipment, is obtained to following uplink speed
The prediction result of rate.
Step S202, prediction of the remote equipment based on the uplink transmission rate, selects the spy of optimal facial image
Sign.
According to the prediction of the uplink transmission rate, the remote equipment, which needs to find, to be most suitable for completing under the present conditions
The facial image of recognition of face configures.This can pass through the allocation optimum that is determined in advance under different transmission rates, such as face figure
As resolution sizes, then size of intensity of illumination etc. is directed to the prediction of the uplink transmission rate, and tabling look-up, it is corresponding to obtain
Facial image resolution sizes, the parameters such as intensity of illumination, by setting described for the corresponding configuration parameter for acquiring facial image
It tables look-up the parameter of acquisition, so that the optimal facial image under the uplink transmission rate is collected, the optimal face
Image, parameter such as resolution ratio, intensity of illumination etc. are the relevant parameters of the acquisition of tabling look-up, and can guarantee pass in the uplink in this way
Under conditions of defeated rate, the collected facial image, it is highest for executing the success rate of recognition of face.
In the actual environment, since some parameters can not be controlled as required, such as facial orientation etc., in dynamic
The variation that not can control the facial orientation parameter in the application of recognition of face needs according to the actual situation, to consider corresponding
Parameter limitation under, how remaining parameter influences the success rate of recognition of face, passes through and chooses corresponding uplink transmission rate, people
Most fast recognition of face is realized in the allocation optimum of remaining parameter under the restrictive condition of face direction.
Step S203, the remote equipment is according to the collection apparatus facial image of the facial image.
After the remote equipment determines the feature of the optimal facial image, according to the feature configuration, equipment is completed
Parameter adjustment, acquire facial image.Such as the case where for camera face face, it is complete to face that flash lamp can be set
Region executes illumination enhancing, while focusing is also multipoint focalizing, guarantees that face's complete image is as clear as possible, image resolution ratio is then
It can suitably reduce, guarantee under conditions of the uplink transmission rate, the facial image being passed on allows within 1 second
The success rate of recognition of face reaches 80%；Or when transmission rate is lower, consider that 2 seconds facial images being passed on can
So that the success rate of recognition of face reaches 80%, at this moment image resolution ratio can be set some higher.
The case where side face is only directed at for camera can be set focusing and there was only single point focalizing, while resolution ratio is then arranged
It is more relatively higher, and facial image collected can only pass through the image comprising face that edge detection determines, keep away
Exempt from the waste of transmission bandwidth.Meanwhile man face image acquiring be also more concerned about can collected face characteristic, for this Partial Feature
Image enhancement is executed, and chooses the higher image configuration parameter of the recognition of face probability of success for being directed to the feature, executes face
Image Acquisition.
Step S204 uploads the facial image with optimal facial image feature.
In the case where collection apparatus facial image of the remote equipment according to the facial image, face collected
Image can guarantee under conditions of the uplink transmission rate, the facial image be uploaded in regulation time delay, according to described
The probability of success highest of facial image completion recognition of face.
Step S205 executes recognition of face to the facial image of the upload, it is ensured that correctly hold with most fast rate
Row recognition of face, to guarantee the user experience of follow-up service.
Referring to Fig. 3, being that a kind of structure of quick self-adapted man face image acquiring equipment provided in an embodiment of the present invention is shown
It is intended to, adaptive face acquisition equipment 301 includes:
Uplink transmission rate obtains module 302, and the receiving ends such as available face recognition device, base station are to uplink speed
The prediction result of rate can also voluntarily estimate uplink transmission rate.The prediction result of the uplink transmission rate is incoming
Recognition of face probability of success prediction module 305 determines how under the limitation of the uplink transmission rate and chooses recognition of face
Key feature and the key feature should value what can guarantee the success rate of recognition of face reaches most within the scope of
Height, while facial image propagation delay time collected is most short under the configuration of the key feature.The key characterization parameter passes
Face characteristic extraction module 303 is passed, face characteristic extraction module 303 configures adaptive face acquisition equipment 301 in the pass
The configuration of key characteristic parameter is lower to execute man face image acquiring, and facial image collected executes figure by face characteristic enhancing module 304
As processing, to improve the success rate of recognition of face.
When uplink transmission rate can not obtain, adaptive face acquisition equipment 301 voluntarily acquires facial image, passes through people
Face characteristic extracting module 303 obtains the relevant key feature of recognition of face, and is sentenced by recognition of face probability of success prediction module 305
The facial image collected that breaks completes the success rate of recognition of face.The success rate of recognition of face is completed according to the facial image,
Recognition of face image transmitting sorting module 306 determines the highest facial image of prioritised transmission success rate.
Referring to Fig. 4, in wireless environments for quick self-adapted man face image acquiring equipment provided in an embodiment of the present invention
The system schematic of application, comprising:
Adaptive face acquires equipment 401, and the prediction result of upstream bandwidth, such as 2Mbps (million are obtained from base station 402
Bits per second).In order to guarantee user experience, recognition of face needs are completed in 2 seconds, therefore facial image collected needs
Face recognition device 403 is uploaded in 1 second, and certainly by the completion recognition of face of face recognition device 403 and by result notice
It adapts to face and acquires equipment 401.Based on this, the recognition of face probability of success prediction module of adaptive face acquisition equipment 401 is true
The size of the fixed facial image should control within 2Mb (megabit), and thereby determine that optimal image resolution ratio, illumination are strong
Degree etc. configuration, or even consider with side face image replace face image, with improve small data quantity transmission limitation under recognition of face at
Power.On this basis, the face characteristic extraction module of adaptive face acquisition equipment 401 is according to the man face image acquiring
Configuration drives adaptive face acquisition equipment 401 to complete man face image acquiring and be uploaded to face recognition device 403, guarantor
Face identification is completed in 2 seconds and notifies the probability of adaptive face acquisition equipment 401 that can guarantee to reach 99%, is left 1%
The case where may then need adaptive face acquisition equipment 401 to resurvey facial image and upload, cause to successfully complete face
The time delay of identification is more than 2 seconds.
It should be noted that for the various method embodiments described above, for simple description, therefore, it is stated as a series of
Combination of actions, but those skilled in the art should understand that, the present invention is not limited by the sequence of acts described because
According to the present invention, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art should also know
It knows, the embodiments described in the specification are all preferred embodiments, and related actions and modules is not necessarily of the invention
It is necessary.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, there is no the portion being described in detail in some embodiment
Point, reference can be made to the related descriptions of other embodiments.
Through the above description of the embodiments, it is apparent to those skilled in the art that the present invention can be with
It is realized with hardware realization or firmware realization or their combination mode.It when implemented in software, can be by above-mentioned function
Storage in computer-readable medium or as on computer-readable medium one or more instructions or code transmitted.Meter
Calculation machine readable medium includes computer storage media and communication media, and wherein communication media includes convenient for from a place to another
Any medium of a place transmission computer program.Storage medium can be any usable medium that computer can access.With
For this but be not limited to: computer-readable medium may include random access memory (RandomAccess Memory, RAM),
Read-only memory (Read-Only Memory, ROM), Electrically Erasable Programmable Read-Only Memory (Electrically Erasable
Programmable Read-Only Memory, EEPROM), CD-ROM (Compact Disc Read-Only Memory,
CD-ROM) or other optical disc storages, magnetic disk storage medium or other magnetic storage apparatus or can be used in carry or store tool
There is the desired program code of instruction or data structure form and can be by any other medium of computer access.Furthermore.Appoint
What connection appropriate can become computer-readable medium.For example, if software is using coaxial cable, optical fiber cable, multiple twin
The nothing of line, Digital Subscriber Line (Digital Subscriber Line, DSL) or such as infrared ray, radio and microwave etc
Line technology from website, server perhaps other remote sources transmit so coaxial cable, optical fiber cable, twisted pair, DSL or
The wireless technology of such as infrared ray, wireless and microwave etc includes in the definition of affiliated medium.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modification, equivalent replacement or improvement etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (7)
1. adaptive face identification method, which is characterized in that comprising steps of
Facial image is acquired, feature extraction is executed；
According to the feature that the feature extraction is extracted, the probability of success of recognition of face is judged；
It is preferential to upload the highest image of the recognition of face probability of success according to the probability of success of the recognition of face, know for face
Not.
2. adaptive face identification method according to claim 1, which is characterized in that in the feature extraction, to spy
Unconspicuous disadvantage is levied to compensate and/or handle the facial image acquired.
3. adaptive face identification method according to claim 1, which is characterized in that extracted according to the feature extraction
Feature out, using the relational model of the feature completed in advance and the recognition of face probability of success, judge recognition of face at
Function probability.
4. the adaptive face identification method according to claim 2 and 3, which is characterized in that it is described complete in advance it is described
The relational model of feature and the recognition of face probability of success compensates and/or to the unconspicuous disadvantage of feature for described
The case where facial image of acquisition is handled needs to be modified relational model.
5. adaptive face identification method, which is characterized in that comprising steps of
Obtain the prediction to uplink transmission rate；
Based on the prediction of the uplink transmission rate, the feature of optimal facial image is selected；
According to the feature of the optimal facial image, facial image is acquired；
Upload the facial image with the optimal facial image feature of the acquisition；
Recognition of face is executed to the facial image of the upload.
6. a kind of adaptive man face image acquiring equipment for realizing adaptive face identification method, which is characterized in that the equipment packet
It includes:
Uplink transmission rate obtains module, obtains the prediction result of uplink transmission rate；
Face characteristic extraction module, for obtaining the relevant key feature of recognition of face, or the configuration adaptive face is adopted
Collect the key characterization parameter that equipment executes man face image acquiring；
Face characteristic enhances module, executes image procossing to the facial image of acquisition；
Recognition of face probability of success prediction module determines how under the limitation of the uplink transmission rate and chooses recognition of face
Key feature completes recognition of face to realize in most short time-delay, or according to the relevant key feature interpretation people of the recognition of face
The probability of success of face identification；
Recognition of face image transmitting sorting module is held according to facial image of the probability of success of the recognition of face to the acquisition
The sequence of row transmission.
7. the system for realizing adaptive face identification method characterized by comprising
Adaptive face acquires equipment, for acquiring facial image and uploading；
Base station, for receiving facial image；
Face recognition device executes recognition of face；
The adaptive face acquisition equipment acquires facial image, and the prediction result of upstream bandwidth is obtained from base station, is needing to protect
The facial image is uploaded into face recognition device in the time delay of card, and recognition of face is completed by face recognition device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811444147.4A CN109558839A (en) | 2018-11-29 | 2018-11-29 | Adaptive face identification method and the equipment and system for realizing this method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811444147.4A CN109558839A (en) | 2018-11-29 | 2018-11-29 | Adaptive face identification method and the equipment and system for realizing this method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109558839A true CN109558839A (en) | 2019-04-02 |
Family
ID=65868050
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811444147.4A Pending CN109558839A (en) | 2018-11-29 | 2018-11-29 | Adaptive face identification method and the equipment and system for realizing this method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109558839A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110472788A (en) * | 2019-08-13 | 2019-11-19 | 天津高创安邦技术有限公司 | A kind of personal information prediction technique and device |
| CN119420699A (en) * | 2025-01-03 | 2025-02-11 | 深圳市泰维精密技术有限公司 | A remote intelligent control method, system and medium for a barrier gate machine |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20030003584A (en) * | 2001-07-03 | 2003-01-10 | 엘지전자 주식회사 | Method to institute resolution of dynamic picture image |
| CN101140620A (en) * | 2007-10-16 | 2008-03-12 | 上海博航信息科技有限公司 | Human face recognition system |
| CN102970510A (en) * | 2012-11-23 | 2013-03-13 | 清华大学 | Method for transmitting human face video |
| CN103310179A (en) * | 2012-03-06 | 2013-09-18 | 上海骏聿数码科技有限公司 | Method and system for optimal attitude detection based on face recognition technology |
| CN103324912A (en) * | 2013-05-30 | 2013-09-25 | 苏州福丰科技有限公司 | Face recognition system and method for ATM |
| US20130258113A1 (en) * | 2009-05-21 | 2013-10-03 | May Patents Ltd. | System and method for control based on face or hand gesture detection |
| CN104254109A (en) * | 2013-06-25 | 2014-12-31 | 华为技术有限公司 | User equipment, base station, streaming media adaptive transmission system and method |
| CN105262982A (en) * | 2014-07-09 | 2016-01-20 | 韩华泰科株式会社 | robot control system |
| CN105812711A (en) * | 2016-05-05 | 2016-07-27 | 广东小天才科技有限公司 | Method and system for optimizing image quality in video call process |
| CN105809125A (en) * | 2016-03-06 | 2016-07-27 | 北京工业大学 | Multi-core ARM platform based human face recognition system |
| CN105979194A (en) * | 2016-05-26 | 2016-09-28 | 努比亚技术有限公司 | Video image processing apparatus and method |
| CN105975961A (en) * | 2016-06-28 | 2016-09-28 | 北京小米移动软件有限公司 | Human face recognition method, device and terminal |
| CN106682650A (en) * | 2017-01-26 | 2017-05-17 | 北京中科神探科技有限公司 | Mobile terminal face recognition method and system based on technology of embedded deep learning |
| CN107292272A (en) * | 2017-06-27 | 2017-10-24 | 广东工业大学 | A kind of method and system of the recognition of face in the video of real-time Transmission |
| CN107346426A (en) * | 2017-07-10 | 2017-11-14 | 深圳市海清视讯科技有限公司 | A kind of face information collection method based on video camera recognition of face |
| CN107871011A (en) * | 2017-11-21 | 2018-04-03 | 广东欧珀移动通信有限公司 | Image processing method, device, mobile terminal, and computer-readable storage medium |
| CN107968944A (en) * | 2017-12-04 | 2018-04-27 | 深圳市瑞科慧联科技有限公司 | A kind of image transfer method and system |
| CN108229323A (en) * | 2017-11-30 | 2018-06-29 | 深圳市商汤科技有限公司 | Supervision method and device, electronic equipment, computer storage media |
| CN108280399A (en) * | 2017-12-27 | 2018-07-13 | 武汉普利商用机器有限公司 | A kind of scene adaptive face identification method |
| CN108399665A (en) * | 2018-01-03 | 2018-08-14 | 平安科技（深圳）有限公司 | Method for safety monitoring, device based on recognition of face and storage medium |
| CN108429701A (en) * | 2018-02-08 | 2018-08-21 | 四川速宝网络科技有限公司 | network acceleration system |
-
2018
- 2018-11-29 CN CN201811444147.4A patent/CN109558839A/en active Pending
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20030003584A (en) * | 2001-07-03 | 2003-01-10 | 엘지전자 주식회사 | Method to institute resolution of dynamic picture image |
| CN101140620A (en) * | 2007-10-16 | 2008-03-12 | 上海博航信息科技有限公司 | Human face recognition system |
| US20130258113A1 (en) * | 2009-05-21 | 2013-10-03 | May Patents Ltd. | System and method for control based on face or hand gesture detection |
| CN103310179A (en) * | 2012-03-06 | 2013-09-18 | 上海骏聿数码科技有限公司 | Method and system for optimal attitude detection based on face recognition technology |
| CN102970510A (en) * | 2012-11-23 | 2013-03-13 | 清华大学 | Method for transmitting human face video |
| CN103324912A (en) * | 2013-05-30 | 2013-09-25 | 苏州福丰科技有限公司 | Face recognition system and method for ATM |
| CN104254109A (en) * | 2013-06-25 | 2014-12-31 | 华为技术有限公司 | User equipment, base station, streaming media adaptive transmission system and method |
| CN105262982A (en) * | 2014-07-09 | 2016-01-20 | 韩华泰科株式会社 | robot control system |
| CN105809125A (en) * | 2016-03-06 | 2016-07-27 | 北京工业大学 | Multi-core ARM platform based human face recognition system |
| CN105812711A (en) * | 2016-05-05 | 2016-07-27 | 广东小天才科技有限公司 | Method and system for optimizing image quality in video call process |
| CN105979194A (en) * | 2016-05-26 | 2016-09-28 | 努比亚技术有限公司 | Video image processing apparatus and method |
| CN105975961A (en) * | 2016-06-28 | 2016-09-28 | 北京小米移动软件有限公司 | Human face recognition method, device and terminal |
| CN106682650A (en) * | 2017-01-26 | 2017-05-17 | 北京中科神探科技有限公司 | Mobile terminal face recognition method and system based on technology of embedded deep learning |
| CN107292272A (en) * | 2017-06-27 | 2017-10-24 | 广东工业大学 | A kind of method and system of the recognition of face in the video of real-time Transmission |
| CN107346426A (en) * | 2017-07-10 | 2017-11-14 | 深圳市海清视讯科技有限公司 | A kind of face information collection method based on video camera recognition of face |
| CN107871011A (en) * | 2017-11-21 | 2018-04-03 | 广东欧珀移动通信有限公司 | Image processing method, device, mobile terminal, and computer-readable storage medium |
| CN108229323A (en) * | 2017-11-30 | 2018-06-29 | 深圳市商汤科技有限公司 | Supervision method and device, electronic equipment, computer storage media |
| CN107968944A (en) * | 2017-12-04 | 2018-04-27 | 深圳市瑞科慧联科技有限公司 | A kind of image transfer method and system |
| CN108280399A (en) * | 2017-12-27 | 2018-07-13 | 武汉普利商用机器有限公司 | A kind of scene adaptive face identification method |
| CN108399665A (en) * | 2018-01-03 | 2018-08-14 | 平安科技（深圳）有限公司 | Method for safety monitoring, device based on recognition of face and storage medium |
| CN108429701A (en) * | 2018-02-08 | 2018-08-21 | 四川速宝网络科技有限公司 | network acceleration system |
Non-Patent Citations (2)
| Title |
|---|
| 季丽萍: "基于RFID和人脸识别考勤系统的设计与实现", 《佳木斯职业学院学报》 * |
| 季丽萍: "基于RFID和人脸识别考勤系统的设计与实现", 《佳木斯职业学院学报》, no. 04, 15 April 2018 (2018-04-15), pages 425 - 426 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110472788A (en) * | 2019-08-13 | 2019-11-19 | 天津高创安邦技术有限公司 | A kind of personal information prediction technique and device |
| CN119420699A (en) * | 2025-01-03 | 2025-02-11 | 深圳市泰维精密技术有限公司 | A remote intelligent control method, system and medium for a barrier gate machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108885698B (en) | Face recognition method and device and server | |
| US10796685B2 (en) | Method and device for image recognition | |
| US20200322237A1 (en) | Traffic detection method and traffic detection device | |
| CN108156519B (en) | Image classification method, television device and computer-readable storage medium | |
| EP2523145A1 (en) | Method for dynamically adapting video image parameters for facilitating subsequent applications | |
| CN110661727B (en) | Data transmission optimization method and device, computer equipment and storage medium | |
| US10055669B2 (en) | Methods and systems of determining a minimum blob size in video analytics | |
| CN107480699A (en) | A kind of intrusion detection method based on channel condition information and SVMs | |
| US20240394556A1 (en) | Machine learning model training method, service data processing method, apparatus, and system | |
| CN110247819B (en) | Wi-Fi video acquisition equipment detection method and system based on encrypted stream identification | |
| CN115661615A (en) | Training method and device of image recognition model and electronic equipment | |
| CN110234015A (en) | Live broadcast control method and device, storage medium and terminal | |
| CN109558839A (en) | Adaptive face identification method and the equipment and system for realizing this method | |
| CN117395705A (en) | 5G network bandwidth prediction method and device based on self-adaptive lifting | |
| CN113660687B (en) | Network difference cell processing method, device, equipment and storage medium | |
| JP6613900B2 (en) | Remote server, communication system, and communication control method | |
| CN109359689A (en) | A kind of data identification method and device | |
| Fang et al. | PIB: Prioritized information bottleneck framework for collaborative edge video analytics | |
| CN113347659B (en) | Flow prediction method and device | |
| CN111817902B (en) | Method and system for controlling bandwidth | |
| KR102274913B1 (en) | Apparatus for Bounding Box Redundancy Removal and Driving Method Thereof | |
| WO2022082742A1 (en) | Model training method and device, server, terminal, and storage medium | |
| WO2024093799A1 (en) | Data labeling method and apparatus, terminal device, and network side device | |
| CN115175233B (en) | Voice quality evaluation method, device, electronic equipment and storage medium | |
| CN116800382B (en) | Index modulation method and device based on visual sensor |
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 |