CN103984941B - Face recognition checking-in method and device thereof - Google Patents
Face recognition checking-in method and device thereof Download PDFInfo
- Publication number
- CN103984941B CN103984941B CN201410255700.5A CN201410255700A CN103984941B CN 103984941 B CN103984941 B CN 103984941B CN 201410255700 A CN201410255700 A CN 201410255700A CN 103984941 B CN103984941 B CN 103984941B
- Authority
- CN
- China
- Prior art keywords
- face image
- face
- front face
- image
- attendance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The invention provides a face recognition checking-in method and a device thereof. Front and side face images of a person are obtained in the face checking-in process, then collected face images in two directions of the front and the side are matched, and the method can be switched into a normal identification process only if the collected pictures of two angles are successfully matched, so as to effectively overcome the problem of false attendance by using the picture. By adopting the scheme, not only is the recognition accuracy rate and efficiency of a face attendance machine not affected, but also the problem of false attendance by using the picture can be quickly and effectively overcome.
Description
Technical field
The present invention relates to field of image recognition, refers in particular to a kind of human face identification work-attendance checking method and its device.
Background technology
The staff attendance management strictly regulated is the important guarantee that management benefit improves in modern enterprises and institutions, and traditional
To check card, block the work attendance form of Zhong Wei representatives, exist generation beat phenomenon, calculating speed slowly, magnetic card be easily damaged or leave behind, equipment
The drawbacks such as maintenance cost height, have increasingly been not suitable for the needs of modern enterprises and institutions' development.
Biological identification technology is by high-techs such as computer and optics, acoustics, biosensor and biostatisticss' principles
Skill means are intimately associated, using the intrinsic physiological property of human body (such as fingerprint, face picture, iris etc.) and behavior characteristicss (such as person's handwriting, sound
Sound, gait etc.) carrying out the identification of personal identification.Nowadays bio-identification is constantly being popularized, and not only becomes examining for enterprise's first-selection
Diligent management system, with the continuous upgrading of biological identification technology, its application is also constantly extending.Enterprise's work attendance is enterprise
A big vital point, the continuous upgrading of its equipment also represents the continuous upgrading of biological identification technology.Biometric apparatus it is emerging
Rise overcome be replicated, it is stolen, a series of problems, such as pass into silence, the life with fingerprint recognition, recognition of face, iris identification as representative
Thing identification equipment is obtained in every field and is widely applied.Fingerprint attendance system is most ripe at present and price is relatively inexpensive
Biological Time Attendance Device, but requirement of the fingerprint attendance machine to environment and work attendance personnel's skin is all very high, when being air-dried, skin
It is dirty, situations such as cast off a skin with regard to None- identified, and read head easily weares and teares.These reasons cause attendance recorder short life, and maintenance cost is high.
The reliability of iris identification equipment is fine, but cost is high, it is impossible to promote on a large scale.
Recognition of face relative to other biometric discrimination methods, with obvious advantage, so as to rapidly become in recent years
One market focus in the whole world.The advantage of face Time Attendance Device is as follows：
1) user is easily accepted by, and transports simply to use, to user without particular/special requirement.
2) anti-counterfeiting performance is good, is difficult to forge or stolen.
3) can carry with, without worrying to omit or losing, can use whenever and wherever possible.
4) noncontact, it is much cleaner, it is not afraid of spread of germs.
5) convenient and swift, recognition time is less than 1 second.
6) a large amount of popularizations of photographic head, it is easy to promote the use of.
7) it is safe and reliable, it is not related to individual privacy.
8) communicated by network or USB flash disk.
It can be seen that, the attendance management mode of enterprise, specification staff attendance behaviour can be effectively improved using human face identification work-attendance checking system
Make, prevent for the behavior checked card, practise fraud, be also convenient for effectively improving work attendance efficiency.TCP/IP networkings are supported simultaneously
Mode, attendance data upload administration section automatically, manage attendance data.It is widely used in enterprises and institutions, education of middle and primary schools machine
Structure, hotel, club, hospital etc..
However, traditional two dimensional image face attendance recorder generally gathers facial image from front, if cribber uses him
The photo of people carries out work attendance, it is possible to play the purpose for replacing other people work attendances, and this allows for two dimensional image face attendance recorder and loses
The meaning for existing.And 3-D view face attendance recorder cost is high, computation complexity high, the general consumer group there is no need choosing
Select big, the slow-footed Time Attendance Device of this cost.
The content of the invention
The technical problem to be solved is：Avoid common face work attendance reliability in the past low and three-dimensional face work attendance
The problem of high cost, there is provided a kind of to pass through front face image and Side Face Image, by increasing face front and side
Matching flow process so that realize low cost under high reliability face work attendance.
In order to solve above-mentioned technical problem, the technical solution used in the present invention is：A kind of human face identification work-attendance checking method, including
Matching flow process and identification process；
The matching flow process includes step,
S21 the front face image and Side Face Image of destination object) are gathered in synchronization；
S22 front face image and Side Face Image are compared using Elastic Matching method), identification is gone to if the match is successful
Flow process, it fails to match then returns error message；
The identification process includes the front face image of collection is compared with data in face database, if comparing consistent
Then work attendance success, fail the step of then returning error message.
The invention further relates to a kind of human face identification work-attendance checking device, including at least two groups photographic head and a central processing unit；Two
Group photographic head is respectively arranged at the both sides of face pickup area and is connected with the central processing unit；
Two groups of photographic head are respectively used to the front face image and Side Face Image for gathering destination object；
The central processing unit is for comparing front face image and Side Face Image using Elastic Matching method, if matching
Successful then go to identification process, it fails to match then returns error message；Perform the front face image by collection of identification process
Compare with data in face database, the work attendance success if comparing unanimously, failure then return error message.
The beneficial effects of the present invention is：By in face work attendance obtain people front and Side Face Image, then
The facial image of the front that collects and side both direction is matched, only when the picture of two angles for collecting
With successfully can just go to normal identification process, so as to effectively overcome the problem using photo falseness work attendance.The program is not only not
The recognition accuracy and efficiency of face attendance recorder are affected, and can fast and effectively overcome asking using photo falseness work attendance
Topic.
Description of the drawings
The concrete structure of the present invention is described in detail in detail below in conjunction with the accompanying drawings
Fig. 1 is the specific example flow chart of the present invention.
Specific embodiment
By describing technology contents of the invention, structural features in detail, realizing purpose and effect, below in conjunction with embodiment
And coordinate accompanying drawing to be explained in detail.
The design of most critical of the present invention is:Before recognition of face, increase a match cognization, it is therefore an objective to check front
Whether the facial image that the facial image that photographic head is obtained is obtained with side photographic head is same person.Needed checking by collection respectively
Match cognization is carried out after diligent people front and the facial image of side, is only put when such a match occurs and is identified, mismatched then straight
Connect return error message.If cribber carries out work attendance using photo, it is clear that facial image and side scanning that front scan is arrived
Facial image be not same person, matching stage cannot pass through, face attendance recorder rejection.This has just been effectively accomplished prevents from making
The purpose practised fraud with photo.
The invention provides a kind of human face identification work-attendance checking method, including matching flow process and identification process；
The matching flow process includes step,
S21 the front face image and Side Face Image of destination object) are gathered in synchronization；
S22 front face image and Side Face Image are compared using Elastic Matching method), identification is gone to if the match is successful
Flow process, it fails to match then returns error message；
The identification process includes the front face image of collection is compared with data in face database, if comparing consistent
Then work attendance success, fail the step of then returning error message.
Knowable to foregoing description, the beneficial effects of the present invention is：By increasing matching flow process before recognition of face, in people
Front and the Side Face Image of people are obtained during face work attendance, then to the front that collects and the facial image of side both direction
Matched, only when the picture match of two angles for collecting successfully can just go to normal identification process, so as to effective
Overcome the problem using photo falseness work attendance.The program does not only affect the recognition accuracy and efficiency of face attendance recorder, and
The problem using photo falseness work attendance can fast and effectively be overcome.
Embodiment 1：
Register flow path is included also in a kind of above-mentioned human face identification work-attendance checking method, the register flow path includes step：
S11) distribute one No. ID for destination object；
S12) gather the facial image of multiple destination objects；
S13) facial image to gathering carries out the laggard pedestrian's face feature extraction of pretreatment；
S14) face database is stored in by the face characteristic of extraction corresponding with No. ID.
Embodiment 2：
Further, step S13 is realized by Adaboost algorithm in above-mentioned register flow path.
Adaboost algorithm is that Paul Viola and Michael Jones proposes [10] in calendar year 2001.It is that one kind changes
For method, its basic thought is to train same grader (Weak Classifier) for different training sets, then that these are different
The grader for obtaining in training set is gathered, and constitutes a final strong classifier.
Embodiment 3：
In the matching flow process, step S22 specifically includes step,
S221 the two-dimensional grid F of face template) is defined to front face image.
S222) use feature vector, XiRepresent the information near two-dimensional grid F interior joint i.
S223 the two-dimensional Gabor filter that mid frequency is different, bandwidth is different, direction is different) is defined, G=is expressed as
(g1,g2,...,gm)T。
S224 two-dimensional grid F' is defined to Side Face Image).
S225) use feature vector, Xi' represent information near two-dimensional grid F' interior joint i, and Xi' and XiIt is special for same type
Levy vector
S226 matching value E (f) between vector is calculated using Euclidean distance)：
In formula, P (i) is grid F interior joint i coordinates in front face image, and Q (j) exists for each node in grid F'
Coordinate in Side Face Image, K represent the number of taken key point, P (Ik),Q(Jk) represent kth key point in positive dough figurine respectively
Coordinate in face image and Side Face Image, λ1And λ2For weight coefficient；
S227) given threshold thresholdd；
S228), as E (f) >=thresholdd, the match is successful goes to identification process；As E (f)<During thresholdd,
It fails to match returns error message.
The method that Elastic Matching is used in the matching flow process of the present embodiment.The method is by face sparse matrix chart
Show, the Gabor wavelet of the stage picture position in figure decomposes the characteristic vector labelling for obtaining, image border connecting node
Distance vector labelling, as Elastic Matching method is insensitive to illumination, displacement, dimensional variation, stablizes with certain to deformation
Property, it is highly suitable for the positive face of this programme and matching for side face.The stage with Elastic Matching method to front face image and
Side Face Image is matched, only successful match, can just carry out the cognitive phase of next step；Otherwise, attendance recorder rejection.
Embodiment 4：
The identification process specifically includes step,
S31) front face image to collecting carries out pretreatment and feature extraction；
The pretreatment includes carrying out image the adjustment of illumination, the removal of noise, the unification of dimension；
The feature extraction is PAC feature extractions, and which is mainly extracted to face characteristic, all spies of each object
Levy and an one-dimensional vector is stored as a training sample, and all training samples are stored in a two-dimensional matrix.
S32) adopt in the comparison sample and face database in the front face image for collecting based on presentation class method
Test sample is compared, and obtains comparing score value；
S33 the absolute difference for comparing sample and any two test specimens in face database in front face image) is compared,
If no more than first threshold, recognition failures return error message, otherwise continue step；
S34) judge that the difference of two comparison score values, whether less than Second Threshold, is that then recognition failures return error message,
Otherwise continue step；
S35) judge whether comparison result is consistent with No. ID, is, continue step, otherwise recognition failures return error message；
In this step, No. ID is by during information gathering, having collected corresponding people with associating for result
The id number of face.During work attendance, matching stage success is identified.Identification process is used based on presentation class
Method, judges when S33 and S34 has been passed through, also judges whether recognition result consistent with No. ID for collecting, if it is inconsistent still
Work attendance cannot be passed through.
S36) work attendance success.
Embodiment 5：
It is as follows based on presentation class method in step S32 in the matching flow process,
It is described based on presentation class method to be：First with all training samples a linear combination representing test specimens
This.Hypothesis has C class, the training sample for having n column vector form per class.Make x1,...,xNFor all N number of training sample (N=
Cn)。x(i-1)n+kRepresent k-th training sample of i-th object, i=1,2 ..., C.Column vector z is made to represent test sample.Will
Representing test sample, test sample with approximate representation can be training sample from all classes
Y=XB ... (4)
Wherein X=[x1...xN] represent training sample, B=[b1...bN]TRepresent coefficient matrix.(4) Xie Tongwei of formula
μ is a little positive constant, and I is unit matrix.OrderIf, it is clear that XTX be it is nonsingular,
The solution of formula (3) can be expressed as following formula
The deviation that CRC is calculated between the i-th class training sample and test sample (compares score value ri) be：
Wherein Xi=[x(i-1)*n+1...xi*n],.IfSo CRC will be tested
Sample assigns to kth class.
Embodiment 6：
In the matching flow process, in step S33, concrete operations are as follows,
Work as riDuring >=thresholdd1, system rejection.Because when deviation score is more than or equal to thresholdd1 (the first thresholds
Value) when, illustrating the test sample and differ too big with the object existed in training sample database, the object of the test sample is not
The personnel of registration；
Embodiment 7：
In the matching flow process, in step S34, concrete operations are as follows,
P test sample is provided with, r is madei1,ri2,...,ripRepresent respectively i-th training sample respectively with p test sample
Difference.So, as | ria-rib| (1≤a≤p, 1≤b≤p, thresholdd2 are Second Threshold to≤thresholdd2, ria
And ribRepresent the difference of i-th training sample and any two test sample) when, system rejection.Because working as test sample and two
When deviation between individual object obtains phase-splitting difference very little, explanation system can not distinguish the two objects.We arrange two threshold values
Thresholdd1 (first threshold) and thresholdd2 (Second Threshold) is unsatisfactory for the testee rejection of condition to survey.By increasing
Plus rejection function, our face telltale system is more safe and reliable.
It is embodied as example：
Below, a concrete application example is provided with regard to the technical scheme of this patent said method.
This programme is improved to face attendance recorder, to overcome attempt that sending out for the cheating of work attendance is carried out using photo
It is raw.Attendance recorder after improvement is using three parts of process point：Register flow path, matching flow process and identification process.It is specifically described below every
The ins and outs of individual flow process：
Register flow path
The flow process is (to be for example usually the information of all measurands (name, No. ID, department etc.) and facial image
Everyone 10 width) it is stored in attendance recorder in advance, to carry out work attendance and record to measurand afterwards.This stage mainly includes people
Face detection, Image semantic classification, feature extraction, face characteristic are stored in face database.We are using conventional based on adaboost algorithms
Carry out Face datection.Adaboost algorithm is that Paul Viola and Michael Jones was proposed in calendar year 2001.It is that one kind changes
For method, its basic thought is to train same grader (Weak Classifier) for different training sets, then that these are different
The grader for obtaining in training set is gathered, and constitutes a final strong classifier.
Matching flow process
Using the method for Elastic Matching.Face sparse matrix figure is represented by the method, then the stage in the figure is schemed
The Gabor wavelet of image position decomposes the characteristic vector labelling for obtaining, the distance vector labelling of image border connecting node.Elasticity
Matching process is insensitive to illumination, displacement, dimensional variation, to deformation with certain stability, is highly suitable for this programme
Positive face and side face match.The stage is matched to front face image and Side Face Image with Elastic Matching method,
Only successful match, can just carry out the cognitive phase of next step；Otherwise, attendance recorder rejection.
Idiographic flow is：
1) No. ID of user input oneself；
2) front face image and Side Face Image of destination object are gathered in synchronization；
3) judge whether face detects successfully, i.e., whether achievement collects the front face image and side face figure of object
Picture, otherwise returns error message, is, continue；
4) front face image and Side Face Image are compared using Elastic Matching method, identification stream is gone to if the match is successful
Journey, it fails to match then returns error message.
Identification process
Using being also referred to as based on the method for classification (RBC) method for representing, conventional RBC, " rarefaction representation is classified
(SRC) method ".In the middle of numerous face identification methods, SRC receives extensive concern.SRC methods are one based on entirety
The method of sample, its basic thought are to combine to represent given test sample using the sparse linear of whole training samples, dilute
Thin non-zero represents that coefficient assumes to concentrate at the true class label of test sample.Verified SRC methods are in terms of identification face
It is largely effective, and expression shape change, illumination condition and face are blocked all there is suitable robustness.Ask in recognition of face
In topic, conventional RBC include based on be the classification of norm minimum, based on the classification of norm minimum, based on norm minimum
Classification.Wherein, the RBC for being constrained based on norm minimum has two types.A kind of is using the training sample from all classes
To represent test sample, such as：Cooperation presentation class (CRC)；Another kind of training sample using from every class is respectively representing survey
Sample sheet, such as：Linear regression classifies (LRC).Both approaches are all finally using representing result being classified.Based on norm
The classification of minimum with its computation complexity it is little, degree of accuracy is high the advantages of be widely used.
Idiographic flow is：
5) front face image to collecting carries out pretreatment and feature extraction；
6) compare with face database, obtain comparing score value；
7) judge to compare that score value, whether not less than first threshold, is that then recognition failures return error message, otherwise continues step
Suddenly；
8) judge to compare that the difference of score value, whether less than Second Threshold, is that then recognition failures return error message, otherwise after
Continuous step；
9) judge whether comparison result is consistent with No. ID, is, continue step, otherwise recognition failures return error message；
10) work attendance success.
The invention further relates to a kind of human face identification work-attendance checking device, including at least two groups photographic head and a central processing unit；Two
Group photographic head is respectively arranged at the both sides of face pickup area and is connected with the central processing unit；
Two groups of photographic head are respectively used to the front face image and Side Face Image for gathering destination object；
The central processing unit is for comparing front face image and Side Face Image using Elastic Matching method, if matching
Successful then go to identification process, it fails to match then returns error message；Perform the front face image by collection of identification process
Compare with data in face database, the work attendance success if comparing unanimously, failure then return error message.
Said apparatus can also be equipped with a display screen if necessary, for showing the facial image in front and side and processing one
Sequence of maneuvers is instructed.A kind of typical scheme, the face attendance recorder after improvement may be designed as having installation in inclined flat board side
One flat board perpendicular to ground, and in photographic head installed above.The people in front and side both direction is gathered simultaneously can
Face image, and show on a display screen.
Face Work attendance device after improvement has following several features：
(1) while gathering the facial image of front and side both direction, and matched.Prevent from examining by photo deception
Diligent machine is so as to the phenomenon of cheating.
(2) display screen can show the facial image of front and side simultaneously, make testee see whether photographic head is clear
The face-image for collecting oneself.
(3) face recognition result is displayed on screen, and checking-in result is recorded in data base.
(4) flat board equipped with infrared camera and screen is inclined, no matter it is low that user is height, is required for bowing slightly
Look down photographic head, you can complete collects facial image.
(5) single-chip microcomputer is housed, for carrying out view data storage and image real time transfer, convenient maintenance and liter in column
Level.
(6) manager can be preserved to processing information by way of wireless network or USB flash disk, convenient to manage.
In sum, the invention provides a kind of human face identification work-attendance checking method and dress for overcoming the cheating of face Time Attendance Device
Put, by increasing match cognization function in the existing technology of face attendance recorder so that cribber cannot carry out void using photo
False work attendance.The improvement project increases a photographic head in the side of attendance recorder, while gathering the image of work attendance person front and side.
First, attendance recorder will carry out face matching, it is ensured that the front face image and Side Face Image for collecting is same person.
After success, the facial image that front is collected by attendance recorder is compared with face database.Not only to non-registered in comparison process
Personnel carry out rejection, also carry out rejection to the object that the deviation between two objects obtains phase-splitting difference very little, so as to ensure that people
Face attendance recorder is safe and efficient to carry out work attendance.Improved recognition of face is not only safe and reliable, and computational efficiency is high, more
It is convenient and practical.
Embodiments of the invention are the foregoing is only, the scope of the claims of the present invention is not thereby limited, it is every using this
Equivalent structure or equivalent flow conversion that bright description and accompanying drawing content are made, or directly or indirectly it is used in other related skills
Art field, is included within the scope of the present invention.
Claims (5)
1. a kind of human face identification work-attendance checking method, it is characterised in that：Including matching flow process and identification process；
The matching flow process includes step,
S21 the front face image and Side Face Image of destination object) are gathered in synchronization；
S22 front face image and Side Face Image are compared using Elastic Matching method), identification stream is gone to if the match is successful
Journey, it fails to match then returns error message；
The identification process includes the front face image of collection is compared with data in face database, examines if comparing unanimously
Diligent success, fail the step of then returning error message；
In the matching flow process, step S22 specifically includes step,
S221 the two-dimensional grid F of face template) is defined to front face image；
S222) use feature vector, XiRepresent the information near two-dimensional grid F interior joint i；
S223 the two-dimensional Gabor filter that mid frequency is different, bandwidth is different, direction is different) is defined, G=(g are expressed as1,
g2,...,gm)T；
S224 two-dimensional grid F' is defined to Side Face Image),
S225) use feature vector, Xi' represent information near two-dimensional grid F' interior joint i, and Xi' and XiFor same type feature to
Amount
S226 matching value E (f) between vector is calculated using Euclidean distance)：
In formula, P (i) is grid F interior joint i coordinates in front face image, and Q (j) is each node in grid F' in side
Coordinate in facial image, K represent the number of taken key point, P (Ik),Q(Jk) represent kth key point in front face figure respectively
Coordinate on picture and Side Face Image, λ1And λ2For weight coefficient；
S227) given threshold thresholdd；
S228), as E (f) >=thresholdd, the match is successful goes to identification process；As E (f) ＜ thresholdd, matching
Failure returns error message.
2. human face identification work-attendance checking method as claimed in claim 1, it is characterised in that：Also include register flow path；The registration flow
Journey includes step,
S11) distribute one No. ID for destination object；
S12) gather the facial image of multiple destination objects；
S13) facial image to gathering carries out the laggard pedestrian's face feature extraction of pretreatment；
S14) face database is stored in by the face characteristic of extraction corresponding with No. ID.
3. human face identification work-attendance checking method as claimed in claim 2, it is characterised in that：Pass through Adaboost in the register flow path
Algorithm realizes step S13.
4. human face identification work-attendance checking method as claimed in claim 1, it is characterised in that：The identification process specifically includes step,
S31) front face image to collecting carries out pretreatment and feature extraction；
S32) using based on presentation class method by the test compared in sample and face database in the front face image for collecting
Sample is compared, and obtains comparing score value；
S33 the absolute difference for comparing sample and any two test specimens in face database in front face image) is compared, if not
More than first threshold, then recognition failures return error message, otherwise continue step；
S34) judge that the difference of two comparison score values, whether less than Second Threshold, is that then recognition failures return error message, otherwise
Continue step；
S35) judge whether comparison result is consistent with No. ID, is, continue step, otherwise recognition failures return error message；
S36) work attendance success.
5. a kind of human face identification work-attendance checking device of the human face identification work-attendance checking method based on described in any one of claim 1-4, which is special
Levy and be：Including at least two groups photographic head and a central processing unit；Two groups of photographic head are respectively arranged at the two of face pickup area
Side is simultaneously connected with the central processing unit；
Two groups of photographic head are respectively used to the front face image and Side Face Image for gathering destination object；
The central processing unit for comparing front face image and Side Face Image using Elastic Matching method, if the match is successful
Identification process is gone to then, it fails to match then returns error message；Perform identification process by collection front face image and people
In face storehouse, data are compared, and the work attendance success if comparing unanimously, failure then return error message.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410255700.5A CN103984941B (en) | 2014-06-10 | 2014-06-10 | Face recognition checking-in method and device thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410255700.5A CN103984941B (en) | 2014-06-10 | 2014-06-10 | Face recognition checking-in method and device thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103984941A CN103984941A (en) | 2014-08-13 |
| CN103984941B true CN103984941B (en) | 2017-04-12 |
Family
ID=51276903
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410255700.5A Active CN103984941B (en) | 2014-06-10 | 2014-06-10 | Face recognition checking-in method and device thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103984941B (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104851140A (en) * | 2014-12-12 | 2015-08-19 | 重庆凯泽科技有限公司 | Face recognition-based attendance access control system |
| CN105139470B (en) * | 2015-09-30 | 2018-02-16 | 杭州海康威视数字技术股份有限公司 | Work attendance method, apparatus and system based on recognition of face |
| CN106780809B (en) * | 2016-12-06 | 2019-02-15 | 浙江水马环保科技有限公司 | Punch card method based on water purifier |
| CN107122720A (en) * | 2017-04-13 | 2017-09-01 | 瑞声科技（南京）有限公司 | Contactless biometric discrimination method and device |
| CN107292299B (en) * | 2017-08-14 | 2018-10-30 | 河南工程学院 | Side face recognition methods based on kernel specification correlation analysis |
| CN107506715A (en) * | 2017-08-17 | 2017-12-22 | 河南工程学院 | Side face identification method based on kernel specification correlation analysis |
| CN107633564A (en) * | 2017-08-31 | 2018-01-26 | 深圳市盛路物联通讯技术有限公司 | Monitoring method and Internet of Things server based on image |
| CN108010046A (en) * | 2017-12-14 | 2018-05-08 | 广西科技大学 | Based on the bionical profile testing method for improving classical receptive field |
| CN108010151A (en) * | 2017-12-26 | 2018-05-08 | 重庆大争科技有限公司 | Human face identification work-attendance checking method |
| CN109063573A (en) * | 2018-07-05 | 2018-12-21 | 怀化学院 | A kind of face recognition image display system |
| CN109376644A (en) * | 2018-10-17 | 2019-02-22 | 深圳市智滴科技有限公司 | A kind of monitoring method and system based on recognition of face |
| CN109409344A (en) * | 2018-12-23 | 2019-03-01 | 广东腾晟信息科技有限公司 | Human face data accurately compares and judgment method |
| CN109740501A (en) * | 2018-12-28 | 2019-05-10 | 广东亿迅科技有限公司 | A kind of Work attendance method and device of recognition of face |
| CN110119674B (en) * | 2019-03-27 | 2023-05-12 | 深圳数联天下智能科技有限公司 | Method, device, computing equipment and computer storage medium for detecting cheating |
| CN110119692B (en) * | 2019-04-19 | 2021-01-19 | 华南理工大学 | Planar portrait detection method based on double cameras |
| CN110163164B (en) * | 2019-05-24 | 2021-04-02 | Oppo广东移动通信有限公司 | Fingerprint detection method and device |
| CN111178259A (en) * | 2019-12-30 | 2020-05-19 | 八维通科技有限公司 | Recognition method and system supporting multi-algorithm fusion |
| CN112001219B (en) * | 2020-06-19 | 2024-02-09 | 国家电网有限公司技术学院分公司 | Multi-angle multi-face recognition attendance checking method and system |
| CN112215974A (en) * | 2020-09-27 | 2021-01-12 | 贵州永光盛安防科技有限公司 | Attendance system based on portrait recognition |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6944319B1 (en) * | 1999-09-13 | 2005-09-13 | Microsoft Corporation | Pose-invariant face recognition system and process |
| CN102622588B (en) * | 2012-03-08 | 2013-10-09 | 无锡中科奥森科技有限公司 | Dual-certification face anti-counterfeit method and device |
| CN102708383B (en) * | 2012-05-21 | 2014-11-26 | 广州像素数据技术开发有限公司 | System and method for detecting living face with multi-mode contrast function |
| CN103530659B (en) * | 2013-10-18 | 2017-08-22 | 哈尔滨工业大学深圳研究生院 | With reference to the face identification method and attendance checking system of original and symmetrical face facial image |
-
2014
- 2014-06-10 CN CN201410255700.5A patent/CN103984941B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103984941A (en) | 2014-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103984941B (en) | Face recognition checking-in method and device thereof | |
| Cao et al. | Gender recognition from body | |
| US20080260212A1 (en) | System for indicating deceit and verity | |
| CN108009482A (en) | One kind improves recognition of face efficiency method | |
| CN105989264A (en) | Bioassay method and bioassay system for biological characteristics | |
| CN106778636A (en) | Auth method and device | |
| Prabhavathi et al. | A smart technique for attendance system to recognize faces through parallelism | |
| CN109063643A (en) | A kind of facial expression pain degree recognition methods under the hidden conditional for facial information part | |
| Shrivastava et al. | Conceptual model for proficient automated attendance system based on face recognition and gender classification using Haar-Cascade, LBPH algorithm along with LDA model | |
| Mardia et al. | On statistical problems with face identification from photographs | |
| Soltane et al. | A smart system to detect cheating in the online exam | |
| Goud et al. | Smart attendance notification system using SMTP with face recognition | |
| Boncolmo et al. | Gender Identification Using Keras Model Through Detection of Face | |
| RU2005100267A (en) | METHOD AND SYSTEM OF AUTOMATIC VERIFICATION OF THE PRESENCE OF A LIVING FACE OF A HUMAN IN BIOMETRIC SECURITY SYSTEMS | |
| Thornton et al. | An evaluation of iris pattern representations | |
| CN112329698A (en) | Face recognition method and system based on intelligent blackboard | |
| Rahman et al. | Human face recognition: An eigenfaces approach | |
| Castillo et al. | Class attendance generation through multiple facial detection and recognition using artificial neural network | |
| Zaffar et al. | A Novel CNN-RNN Model for E-Cheating Detection Based on Video Surveillance | |
| Ding et al. | Discussions on some problems in face recognition | |
| Hava et al. | Free & generic facial attendance system using android | |
| Rhuma et al. | Posture Recognition Based Fall Detection System | |
| Balážia | Human gait recognition: Based on body component trajectories | |
| Chindaro et al. | Assessing visuo-spatial neglect through feature selection from shape drawing performance and sequence analysis | |
| Hameed et al. | Novel simulation framework of three-dimensional skull bio-metric measurement |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |