CN104143083A - Face recognition system based on process management - Google Patents
Face recognition system based on process management Download PDFInfo
- Publication number
- CN104143083A CN104143083A CN201410330602.3A CN201410330602A CN104143083A CN 104143083 A CN104143083 A CN 104143083A CN 201410330602 A CN201410330602 A CN 201410330602A CN 104143083 A CN104143083 A CN 104143083A
- Authority
- CN
- China
- Prior art keywords
- face
- contrast
- user
- group
- similarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000008569 process Effects 0.000 title claims abstract description 38
- 230000001815 facial effect Effects 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 17
- 230000007613 environmental effect Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 235000012364 Peperomia pellucida Nutrition 0.000 description 1
- 240000007711 Peperomia pellucida Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000003651 drinking water Substances 0.000 description 1
- 235000020188 drinking water Nutrition 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The invention provides a face recognition system based on process management. Multiple-time face comparison is performed in the whole reading process and comparison results are recorded so as to strictly ensure that users are always in person from beginning to end. The face recognition system comprises an information setting module, a feature acquisition module, a feature comparison rule setting module and a face feature comparison module, wherein the feature comparison rule setting module is used for setting at least one feature comparison rule, providing different comparison rules for different users in different application environments and widening a system application range. During comparison, certain times of comparison can be set as one set, certain times of comparison failures can be permitted for each set so as to avoid comparison failures caused by accidental situation, and the usage of the users is prevented from being influenced. In addition, human intervention is not needed in the whole process, and automatic system management is completely achieved.
Description
Technical field
The invention belongs to technical field of biometric identification, particularly a kind of face recognition technology.
Background technology
Recognition of face is a kind of biological identification technology that the face feature information based on people is carried out identification.The image that contains face with video camera or camera collection or video flowing, and then the face detecting is carried out to a series of correlation techniques of face.
Along with the continuous progress of society and each side are for an urgent demand of identity contrast automatically fast and effectively, recognition of face is owing to having directly, close friend, feature easily, user, without any mental handicape, is easy to be accepted by user, thereby has obtained research widely and application.
And the application of face recognition technology is at present mainly from improving face identification precision, in large-scale face information database, accurately locate a certain individual, the shortcoming of this mode is exactly calculation of complex, need to process mass data, higher to memory device and computing equipment requirement, increase cost.
In addition, in the time carrying out face contrast, generally take contrast way once, the for example application in face system attendance checking system, just only in the time of the upper and lower class of employee, record work attendance information, to going out of employee midway, and employee's the running time record of having no idea.
Based on prior art, also there is certain deviation in comparing result, if completely according to the comparing result of recognition of face, system is processed, existing face identification system is to accomplish, therefore also needs a certain amount of artificial participation.
Summary of the invention
In order to overcome above shortcoming, the invention provides a kind of face identification system based on process management, detected person is repeatedly carried out within a period of time to face contrast, in the time that contrast success ratio reaches some setting values, think that detected person is me.
The technical solution adopted for the present invention to solve the technical problems is: a kind of face identification system based on process management, comprising: user log-in block, facial information acquisition module and Characteristic Contrast rule arrange module,
User log-in block: confession user logins described face identification system and user's essential information is provided;
Facial information acquisition module: for gathering user's facial information;
Characteristic Contrast rule arranges module, at least one contrast rule based on facial information is set, in the time contrasting, and can carry out different contrast rules for different users.
System also needs, for user, training storehouse is set, store exactly a certain user's various faces feature, system is passed through external unit, for example use picture pick-up device, gather user's facial information, by Gabor feature and carry out uniform sampling, and then by subspace dimension-reduction algorithm etc., the Factorial Face Code that generates people calculated.Wherein, the living things feature recognition algorithm based on subspace dimensionality reduction is a kind of algorithm known, no longer discusses here; Gabor feature can effectively represent the local feature of face picture from different directions with yardstick, its algorithm is also disclosed, also no longer discusses here.
System can define multiple contrast rule, and the contrast rule of its correspondence is set for each user, after logging in system by user, first system detects the pre-stored Factorial Face Code in training storehouse of this user, and the current Factorial Face Code of Real-time Collection user, and the two is contrasted, here, system is according to the contrast rule of setting for this user, carries out contrast.
Contrast rule is for process management, refer to during user's login whole in, system is constantly carried out face contrast, for example, if when similarity reaches the threshold value of setting: 77%, system contrasts successfully, can regard as is registrant, and system continues to carry out contrast next time; Otherwise, contrasting unsuccessfully, system is thought non-registered people, at this moment system triggers external interface, queued for log-on user self acknowledging.The Threshold here can be set different values according to different users, and lower if domestic consumer's threshold value can arrange, if superuser, it is higher that threshold value can arrange,
But in the time carrying out face contrast, often because unexpected situation causes contrasting unsuccessfully, for example: due to light, angle, the factors such as expression can affect the result of contrast, therefore face contrasts and unsuccessfully can not represent not to be registrant sometimes, once and because detect and be not user, will trigger interface, require human intervention, the impact producing in order to eliminate this uncertain factor user as far as possible, the present invention also provides a kind of scheme: will repeatedly contrast and be decided to be one group, in group, can allow the contrast failure of certain number of times, taking group as unit, in organizing, the frequency of failure of contrast is less than the frequency of failure of regulation, this group contrasts successfully, system is thought registered user, otherwise this group contrasts unsuccessfully, system is thought and is not registered user.
Be applied to process management based on this face recognition technology, allow the method for certain error simultaneously, can ensure that using the people of system is from start to finish me always, avoid error or the unexpected interruption causing because existing simultaneously in the process of recognition of face, and the impact that user is caused.
In addition, in order to improve the efficiency of contrast, can also adopt a species stage control methods, when similarity low (or contrast number of success is few), repeatedly contrast, when similarity high (contrast number of success is many), suitably reduces contrast frequency, to improve performance, reduce mobile device power consumption; Other contrast rules can also be set.
The present invention unlike the prior art, face characteristic contrast provided by the invention is man-to-man contrast, do not need to set up face information database, for example: face comparing function can be placed on to client, in the time that user uses system, do not need identity to retrieve, be stored in local face information and contrast with the information of the face using but directly call.
Face information can also be placed on to server end, in the time that user uses system, the information such as the login people who provides according to user, call this user and be stored in the corresponding face feature on server, end user's face feature data and the face feature data that are stored in this user on server are contrasted.
Therefore, recognition of face ratio carry out in pairs than, be different from safety check, the one-to-many contrasts such as work attendance, do not need to set up a large amount of crowds' face information database, do not need identity to retrieve.Therefore fiduciary level is higher, and usefulness is better, can also support off-line.
In addition, the present invention also provides system interface, for called or call other foreign currency systems by external system.
Beneficial effect of the present invention is mainly reflected in, and face recognition application is in the benefit of process management:
1. more convenient, do not need user to do extra operation,
2. process management, strict guarantee end user is from start to finish me always.
3. pinpoint accuracy is not asked most in recognition of face contrast, therefore requires low to operand and arithmetic facility.Smart mobile phone also can meet performance requirement.
4. by common picture pick-up device, realize face recognition process, compare professional identification equipment, cost is lower, and application scenarios is wider.
5. whole process, without human intervention, realizes system automation management completely.
Brief description of the drawings
Below in conjunction with accompanying drawing, face identification system of the present invention is specifically described.
Fig. 1 is functional block diagram of the present invention;
Fig. 2 is general rule process flow diagram of the present invention;
Fig. 3 the present invention is directed to embodiment bis-, the process flow diagram dividing into groups in contrast;
Fig. 4 the present invention is directed to embodiment tri-, every group of different process flow diagram of the number of times that need to contrast.
Embodiment
As shown in Fig. 1 functional block diagram, the technical solution adopted for the present invention to solve the technical problems is: a kind of face identification system based on process management, comprising: user log-in block, facial information acquisition module and Characteristic Contrast rule arrange module,
User log-in block:;
1. user log-in block
Confession user logins described face identification system and user's essential information is provided, for example: the contents such as the contrast rule of login name, user name, authority, correspondence.
| Login name | User name | Authority | Contrast rule | …… |
| Zhang | Zhang San | Keeper | Rule 1 | ? |
| Li | Li Si | Domestic consumer | Rule 1 | ? |
| wang | King five | Domestic consumer | Rule 2 | ? |
2. training storehouse:
| User name | Factorial Face Code | … |
| Zhang San | [CvArr formatted data 1] | ? |
| Zhang San | [CvArr formatted data 2] | ? |
| Zhang San | [CvArr formatted data 3] | ? |
| Li Si | [CvArr formatted data 1] | ? |
| Li Si | [CvArr formatted data 2] | ? |
| Li Si | [CvArr formatted data 3] | ? |
| King five | [CvArr formatted data 1] | ? |
| King five | [CvArr formatted data 2] | ? |
| King five | [CvArr formatted data 3] | ? |
In the time setting user information, need to generate this user's Factorial Face Code, the Factorial Face Code is here one group, multiple Factorial Face Codes of this user, from different perspectives, under different illumination conditions, generate, in the present invention, specifically generating Factorial Face Code can realize with three functions:
1) face detect BoolDetectFace (IplImageframe, CvMat*faceImg8) function,
2) add training storehouse BoolAddTrainFace function
3) generate face characteristic storehouse String TrainFace () function.
Wherein, the function that generation face characteristic storehouse String TrainFace () function is mainly realized is exactly: repeatedly call face and detect BoolDetectFace function and add training storehouse BoolAddTrainFace function, face detects BoolDetectFace function, be used for capturing face information, and this information is converted into CvMat form, for adding training storehouse this parameter of BoolAddTrainFace function call, face information is stored in database.Repeatedly call and add training built-in function owing to generating face characteristic built-in function, so what store in database is this user's various faces information.
In addition, also need to arrange a reference table, for example: the contents such as user name, reading content, executing rule.
| User name | Reading content | Executing rule | … |
| Zhang San | Document 1 | Rule 1 | ? |
| Li Si | Game 1 | Rule 2 | ? |
| King five | Program 1 | Rule 1 | ? |
| Zhang San | Document 2 | Rule 2 | ? |
| Li Si | Game 2 | Rule 1 | ? |
| King five | Program 2 | Rule 2 | ? |
2. face characteristic collection:
By the view data of input camera collection face, according to the face picture or the video recording that collect, by face feature, the ratio value of face, by Gabor feature and carry out uniform sampling, and then by subspace dimension-reduction algorithm etc., calculate and generate binary coding, to represent the characteristic information of face.
Living things feature recognition algorithm based on subspace dimensionality reduction is a kind of algorithm known, no longer discusses here; Gabor feature can effectively represent the local feature of face picture from different directions with yardstick, its algorithm is also disclosed, also no longer discusses here.
Implementation method of the present invention is: open camera by cross-platform computer vision storehouse (OpenCV), extract the image of camera collection, then detect face by the image detection algorithm of OpenCV, after face being detected, adjust face angle (face alignment) by ASM algorithm (Active Shape Model active shape model) algorithm again, carry out unitary of illumination processing by homomorphic filtering+histogram specification algorithm, finally obtain face characteristic information, this information can be applied in two places
1) arrange in module in essential information, in the time of a newly-built user profile, can be used as this user's Factorial Face Code storage;
2) in reading process, the Factorial Face Code generating as interface, is stored in this registered user: the Factorial Face Code in training storehouse contrasts, to be used for whether detecting current face as registered user.
3. Characteristic Contrast rule arranges
Whether the registrant that the present invention sets is that I adopt similarity to confirm, this similarity can be arranged by the user with specified permission, and for example, in some strict occasion, similarity is 80% to think me; At some, reader is required not strict, but reading process is required very smooth time, similarity can be made as 60% etc.
Unlike the prior art, the present invention realizes a period of time internal procedure monitoring in the present invention, but not disposable contrast.As shown in Fig. 2 general rule setting procedure figure, face characteristic contrast of the present invention, is in whole reading process, repeatedly contrast, if judge that reader is not registrant in the process of contrast, System Halt reading content, waits for that user further operates; If reader is me, time delay a period of time is carried out face contrast again, finishes until read.After reading finishes, reading result is uploaded onto the server, reading result can be the video recording of whole process, also can be situation with each contrast of data recording (for example: each time the time of contrast, similarity, Factorial Face Code, system to information etc.), reading result can automatically be uploaded by system after reading completes, and also can be selected to upload by user.
But in the time carrying out face contrast, often because unexpected situation causes contrasting unsuccessfully, for example: because the factors such as light, angle, expression can affect the result of contrast, therefore Real-time Collection to Factorial Face Code do not reach similarity requirement with the Factorial Face Code that is stored in feature database, this can not represent not to be registrant; Again for example, reading in so relatively long time, also there is the of short duration situation of leaving camera (for example drinking water, take thing etc.) of user, cause not collecting facial information.Therefore, cause reading interruption in order to eliminate this uncertain factor as far as possible, and the impact that user is produced the present invention proposes the second embodiment: the face comparison process of certain number of times can be made as to one group, in this group comparison process, in the time that contrast number of success reaches a certain setting value, think that the contrast of this group passes through, this group comparative information is recorded, if the frequency of failure reaches limit value simultaneously, just think that this group contrasts unsuccessfully, system just can be ended reading content.As shown in the process flow diagram that Fig. 3 divides into groups in contrast, in whole reading process, be divided into some groups by certain contrast number of times, group comparing result is contrasted at every turn, if judge that reader is not registrant in the process of contrast, System Halt reading content, if reader is me, time delay a period of time restarts to carry out contrast.For example, can be set to: contrast per second once, is submitted a correlation data to every five times, and wherein twice correlation data failure, judges and contrast unsuccessfully, and after 5 seconds, content is ended; Can also be set to: contrast in every two seconds once, is submitted a correlation data to every five times, contrasts unsuccessfully for three times, judges and contrasts unsuccessfully, and after 10 seconds, content is ended.
In addition, the invention allows for the third embodiment, is to improve on the basis of the second embodiment, and it is no longer definite that concrete improvement is exactly every group of number of times that need to carry out face contrast, for example, in the time carrying out face contrast, when similarity is low, repeatedly contrast, when similarity is high, suitably reduce contrast frequency, to improve performance, reduce mobile device power consumption.Therefore,, if Fig. 4 is in the time dividing into groups, shown in every group of different process flow diagram of the number of times that need to contrast, can adopt a kind of stepped appearance algorithm:
1) in the time contrasting similarity lower than threshold value 70% first, at this moment one group of needs is set contrasts 5 times, in the time that in follow-up four whens contrast, have the similarity of 3 times all higher than threshold value, be judged to be me, " there is environmental impact factor, please re-register face information " and continue to carry out face and contrast in system meeting; In the time that the contrast of follow-up four times has twice contrast similarity lower than threshold value, system judges that reader is not registrant, System Halt reading content.
2) contrasting similarity higher than threshold value first, but lower than 75% time, at this moment one group of needs is being set and contrasts 5 times, in the time having the similarity of 3 times all higher than threshold value in the contrast similarity of follow-up four times, be judged to be I, system continues to carry out face and contrasts; In the time that the contrast of follow-up four times has twice contrast similarity lower than threshold value, system judges that reader is not registrant, System Halt reading content.
3) to contrast first similarity high during for 75%-80%, at this moment one group of needs is set contrast 3 times, in the time having the similarity of 1 time all higher than threshold value in the contrast similarity of follow-up twice, is judged to be I, system continues to carry out face and contrasts; In the time that the contrast of follow-up 2 times has all lower than threshold value, system judges that reader is not registrant, System Halt reading content.
4) in the time contrasting first that similarity is high is 80%, at this moment one group of needs is set and contrasts 1 time, therefore, no longer need follow-up contrast to be directly judged to be me.
After this group has contrasted, again perform step 1), carry out the contrast of next group.
Except the above embodiments, the present invention can also have other contrast rule, for example can be by set of time contrast rule, for example, in the time that similarity is higher, can extend reduced time next time, in the time that similarity is lower, need to shortens the contrast rule based on process management such as reduced time and all be suitable for.
4. face characteristic contrast module
The present invention unlike the prior art, face characteristic contrast of the present invention does not need to set up face information database, but face recognition code is stored in database, corresponding his the face recognition code of each user, when user uses system, first needs to sign in in native system, when contrast at every turn, contrast with the face recognition code that is stored in this registered user in computing machine as long as gather the current face recognition code of reader, therefore fiduciary level is higher, and usefulness is better.
As another kind of real-time mode of the present invention, the present invention can also support off-line, the content that namely user will read downloads to local client, the face recognition code of registration when simultaneously local client also stores this user and registers, in the time that this user reads, can log on network, but directly read the local content of having downloaded at local client, simultaneity factor is also to detect this user's of local storage face recognition code and the whether consistent registered user of determining whether of the face recognition code reading in reading process.
In the present invention, face characteristic contrast module is in user's reading process, the rule corresponding to content of reading for this user and this user arranging is set carries out the program of this rule correspondence in contrast rule arranges module according to information in module-reference table.For example, for user Zhang San, in the time of reading documents 1 and document 2, called respectively rule 1 and rule 2, regular 1 correspondence be exactly the program of embodiment mono-, rule 2 is exactly the program of corresponding embodiment bis-, therefore,
1) when Zhang San signs in to system, and when opening document 1, system is carried out embodiment bis-: detect in real time this user's Factorial Face Code, the Factorial Face Code being simultaneously stored in system with this user contrasts, if similarity reaches threshold value, think me, after time delay a period of time, the face feature that again gathers this user continues contrast, until read, or similarity do not reach threshold value, end reading content, wait for that user logins again.
2) when Zhang San signs in to system, and when opening document 2, system is carried out embodiment bis-, and (contrast per second once, submit a correlation data to every five times, submit correlation data failure for twice to, judge and contrast unsuccessfully): system detects this user's Factorial Face Code in real time, while contrasts with the Factorial Face Code that this user is stored in system, and carry out verification taking every 5 times as one group, if contrasting failed number of times in these 5 times is less than 2 times, think me, the face feature that again gathers this user continues contrast, until read, or in the time that its errors number of comparing result of one group of data is more than or equal to 2 times, think and be not that registration in person, end reading content, wait for that user logins again.
After reading completes, system will be returned to the process of whole reading tracking, the result of returning can be whole section and read video recording, also can be the result through systematic analysis processing, for example, in embodiment mono-, can return to the number of times that altogether detects, detect number of success, the frequency of failure, reading divide several times complete, the similarity of judgement at every turn and real-time recognition of face code information etc. content; For example, in embodiment bis-, can return to the number of times, packet count, the every group of detection number of success that altogether detect, the frequency of failure, reads point similarity that completes several times, judges at every turn and real-time recognition of face code information etc. content.For the whole procedural information mode of reading is stored, transmission, Data Comparison etc.; can protect privacy of user; improve arithmetic speed, the large group that is simultaneously more convenient for is followed the tracks of and excavated to reading process the processing of the supervise feedback of forcing reading task to employee.
In the present invention, face contrast can be used function: floatVerifyFace (StringfaceModel, CvMat*face, long isNewModel) realize, here there are two parameters, one is the pre-stored Factorial Face Code of this user, also have one to be to call face to detect BoolDetectFace (IplImageframe, CvMat*faceImg8) the real-time face information of this reader that function obtains, function f loatVerifyFace contrasts these two parameters, and the result of returning is the similarity of these two values.
5. system interface:
First native system needs to be installed on the hardware device with electronic reading function, for example: mobile phone, palm PC, panel computer, notebook computer, on the equipment such as desktop computer, then to needing the content of carrying out user's contrast in use procedure to monitor, for example, can monitor: electronic document, video file, webpage, flash, mail, OA system, CR system, game etc.When above-mentioned file is monitored, (this application program is customer-furnished need to first to carry out an application program, user can be opened the content of needs monitoring by this program, for example this application program can be the OA system that user provides, also can be only a process, whether this process is exactly to need the content of monitoring to be opened etc. for catching), and opening after this program, the interface function providing of native system is provided, the content of needs monitoring is carried out to Real-Time Monitoring.The specific description of interface function is shown in the 6th part.
6. function declaration:
1) BoolDetectFace (IplImageframe, CvMat*faceImg8): face detection function, the view data of input camera collection, output face facial information.
Wherein, [in] frame: the view data of camera collection
Type declaration: IplImage, IplImage is the structure type of OpenCV definition;
[out] faceImg8: the face facial information that interface generates
Type declaration: CvMat*, CvArr is the structure type of OpenCV definition;
Rreturn value: face (true/false) whether detected;
2) BoolAddTrainFace (CvMat*face): add training storehouse, face is detected to the face facial information that BoolDetectFace function obtains and add training storehouse.When initialization system, store the face information collecting in database (static data), whether for reading later time, detecting reader is me.
Wherein, the face facial information that [in] face:DetectFace interface returns
Type declaration: CvMat*, CvArr is the structure type of OpenCV definition;
Rreturn value: whether add successfully (true/false);
3) String TrainFace (): generate face characteristic storehouse, return to the face characteristic storehouse generating after training.This function is mainly carried out and is repeatedly called BoolDetectFace and BoolAddTrainFace, realizes the various faces condition code of storing this user in user basic information.
4)floatVerifyFace(StringfaceModel,CvMat*face,long?isNewModel)：
Face contrast: program face characteristic storehouse is generated to the BoolDetectFace current reader who obtains and the data of training library storage and contrast, whether the people who detects current reading is user.
Wherein, the face characteristic storehouse that [in] faceModel:TrainFace interface returns
Type declaration: character string type is that described user is stored in the multiple Factorial Face Codes in described training storehouse;
The face facial information that [in] face:DetectFace interface returns
Type declaration: CvMat*, CvArr is the structure type of OpenCV definition, is the described Factorial Face Code of the described login user that obtains of described condition code generation module;
[in] isNewModel: whether be new face characteristic storehouse
Type declaration: 1 (new feature storehouse), 0 (former feature database);
Rreturn value: return to comparing result, >0 (similarity) ,-1 (non-I);
Face contrast function: floatVerifyFace by parameter f aceModel and face by inner computing with relatively obtain the two similarity value, as the rreturn value of described function f loatVerifyFace, the value of the described similarity of returning when described face contrast function is one and is greater than 0 and while being less than the similarity threshold of definition or-1 time, described user is not registered user; Otherwise login user is registered user.
The above is only preferred embodiment of the present invention, not the present invention is done to any pro forma restriction, although the present invention discloses as above with preferred embodiment, but not in order to limit the present invention, any technician who is familiar with this patent is not departing within the scope of technical solution of the present invention, when can utilizing the technology contents of above-mentioned prompting to make a little change or being modified to the equivalent embodiment of equivalent variations, in every case be the content that does not depart from technical solution of the present invention, any simple modification of above embodiment being done according to technical spirit of the present invention, equivalent variations and modification, all still belong in the present invention program's scope.
Claims (9)
1. the face identification system based on process management, is characterized in that, comprising: user log-in block, facial information acquisition module and Characteristic Contrast rule arrange module, wherein,
Described user log-in block: confession user logins described face identification system and user's essential information is provided;
Described facial information acquisition module: for gathering user's facial information;
Described Characteristic Contrast rule arranges module, at least one contrast rule based on facial information is set, in the time contrasting, and can carry out different contrast rules for different users.
2. face identification system according to claim 1, it is characterized in that, also comprise condition code generation module and training storehouse, described condition code generation module, for by from described facial information acquisition module to user's facial information, generate binary code through computing, as the Factorial Face Code of user described in this;
Described training storehouse, for storing described user's described Factorial Face Code corresponding to various faces image;
In the time that the described contrast of definition is regular, condition code generation module described in real-time calling, generates described user's current Factorial Face Code, and carries out face contrast with the described Factorial Face Code that described user stores in described training storehouse.
3. face identification system according to claim 2, it is characterized in that, described face contrast is used face contrast function: floatVerifyFace, and described face contrast function comprises input parameter: faceModel and face, the rreturn value of described face contrast function is similarity value
Wherein, described parameter f aceModel is character string type, is that described user is stored in the multiple Factorial Face Codes in described training storehouse; Parameter f ace is the structure type of instrument OpenCV definition, is the described Factorial Face Code of the described login user that obtains of described condition code generation module,
The threshold value of the similarity of described face identification system definition is more than 66%, preferably 77%,
Described face contrast function: floatVerifyFace by parameter f aceModel and face by inner computing with relatively obtain the two similarity value, as the rreturn value of described function f loatVerifyFace, when the value of the described similarity of returning when described face contrast function is less than the described threshold value of described similarity, described face contrasts unsuccessfully, and described login user is not described user; Otherwise described face contrasts successfully, described login user is described user.
4. face identification system according to claim 3, it is characterized in that, in the time completing described face contrast at every turn, described face identification system record: described user's name, reduced time, described similarity value, in the time that described face contrasts unsuccessfully, described face identification system is ended program, waits for next step operation of described user; In the time that described face contrasts successfully, described face identification system contrasts next time, until described face identification system is complete.
5. face identification system according to claim 4, it is characterized in that, described contrast rule also comprises group and the maximum contrast frequency of failure, described group comprises repeatedly contrast, the described maximum frequency of failure is less than described contrast number of times, if in the time carrying out the contrast of described face, the frequency of failure in described group is less than the described maximum frequency of failure, and described face contrasts successfully; Otherwise the described face of described group contrasts unsuccessfully.
6. face identification system according to claim 5, is characterized in that, 5-10 contrast of every group of definition, and preferably 5 times, the described maximum frequency of failure is 2-3 time.
7. face identification system according to claim 5, is characterized in that, also comprises highest similarity value, and described highest similarity value is greater than described threshold value, and in the time that described similarity value is high, the contrast number of times of every group of definition is few; When described similarity value is low, every group definition contrast often, that is:
1) in the time having contrasted first for described every group, if lower than described threshold value, arranging in described group, described similarity needs to contrast 5 times, when there being twice contrast failure, the described face of described group contrasts unsuccessfully, System Halt reading content; Otherwise the described face of described group contrasts successfully, system prompt: the contrast of next group is carried out in " have environmental impact factor, please re-register face information " continuation;
2) described every group contrast completes first time, if described similarity is equal to or higher than described threshold value, but during lower than described highest similarity, described every group of needs are set to be contrasted 5 times, when there being twice contrast failure, the described face of described group contrasts unsuccessfully, System Halt reading content; Otherwise the described face of described group contrasts successfully, continue to carry out the contrast of next group;
3) described every group contrast completes first time, if when described similarity is not less than described highest similarity, every group of needs is set and contrasts 1 time, system judges that described group contrasts successfully at once, continues to carry out the contrast of next group.
8. face identification system according to claim 7, it is characterized in that: also comprise one or more medium similarity value, the value of described medium similarity is greater than described threshold value and is less than described highest similarity simultaneously, if described similarity is equal to or higher than threshold value, but during lower than the described medium similarity of system definition, described every group of needs are set and contrast 3 times, when there being twice contrast failure, the described face of described group contrasts unsuccessfully, System Halt reading content; The described face of described group contrasts successfully, continues to carry out the contrast of next group.
9. face identification system according to claim 1, it is characterized in that, also comprise that user and rule arrange storehouse, for associated described user and described contrast rule, in the time that described user signs in to described face identification system, carry out described user, with rule, described contrast rule corresponding to user described in storehouse is set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410330602.3A CN104143083B (en) | 2014-07-11 | 2014-07-11 | A kind of face identification system of Kernel-based methods management |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410330602.3A CN104143083B (en) | 2014-07-11 | 2014-07-11 | A kind of face identification system of Kernel-based methods management |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104143083A true CN104143083A (en) | 2014-11-12 |
| CN104143083B CN104143083B (en) | 2018-03-02 |
Family
ID=51852253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410330602.3A Expired - Fee Related CN104143083B (en) | 2014-07-11 | 2014-07-11 | A kind of face identification system of Kernel-based methods management |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104143083B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966078A (en) * | 2015-07-22 | 2015-10-07 | 胡东雁 | Student identity recognition system and method for on-line training curriculum |
| CN105389551A (en) * | 2015-10-28 | 2016-03-09 | 广东欧珀移动通信有限公司 | Eyeball recognition device detection method and detection device |
| CN105844203A (en) * | 2015-01-12 | 2016-08-10 | 阿里巴巴集团控股有限公司 | Human face vivo detection method and device |
| CN106022313A (en) * | 2016-06-16 | 2016-10-12 | 湖南文理学院 | Scene-automatically adaptable face recognition method |
| CN106257493A (en) * | 2016-08-30 | 2016-12-28 | 重庆市城投金卡信息产业股份有限公司 | Traffic promotional card falsely uses recognition methods and identification system |
| CN106372856A (en) * | 2016-08-31 | 2017-02-01 | 北京汇通天下物联科技有限公司 | Driver work attendance method and driver work attendance system |
| CN106447949A (en) * | 2015-08-12 | 2017-02-22 | 阿里巴巴集团控股有限公司 | Information processing method and device |
| CN106850667A (en) * | 2017-03-03 | 2017-06-13 | 杭州智贝信息科技有限公司 | It is a kind of to continue certification security protection system and its method |
| CN106992968A (en) * | 2017-03-03 | 2017-07-28 | 杭州智贝信息科技有限公司 | A kind of client-based face continues authentication method |
| CN107679514A (en) * | 2017-10-20 | 2018-02-09 | 维沃移动通信有限公司 | A kind of face identification method and electronic equipment |
| CN107832598A (en) * | 2017-10-17 | 2018-03-23 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
| CN108038363A (en) * | 2017-12-05 | 2018-05-15 | 吕庆祥 | Improve the method and device of Terminal security |
| CN108549946A (en) * | 2018-04-10 | 2018-09-18 | 阳光暖果（北京）科技发展有限公司 | A kind of plant maintenance and maintenance artificial intelligence supervisory systems and method |
| CN109615750A (en) * | 2018-12-29 | 2019-04-12 | 深圳市多度科技有限公司 | The recognition of face control method and device of door access machine, access control equipment, storage medium |
| CN109815837A (en) * | 2018-12-29 | 2019-05-28 | 维沃移动通信有限公司 | A kind of face information typing control method and mobile terminal |
| CN110751758A (en) * | 2019-09-29 | 2020-02-04 | 湖北美和易思教育科技有限公司 | Intelligent lock system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101162499A (en) * | 2006-10-13 | 2008-04-16 | 上海银晨智能识别科技有限公司 | Method for using human face formwork combination to contrast |
| CN102004908A (en) * | 2010-11-30 | 2011-04-06 | 汉王科技股份有限公司 | Self-adapting face identification method and device |
| US20130121584A1 (en) * | 2009-09-18 | 2013-05-16 | Lubomir D. Bourdev | System and Method for Using Contextual Features to Improve Face Recognition in Digital Images |
| CN103810663A (en) * | 2013-11-18 | 2014-05-21 | 北京航天金盾科技有限公司 | Demographic data cleaning method based on face recognition |
-
2014
- 2014-07-11 CN CN201410330602.3A patent/CN104143083B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101162499A (en) * | 2006-10-13 | 2008-04-16 | 上海银晨智能识别科技有限公司 | Method for using human face formwork combination to contrast |
| US20130121584A1 (en) * | 2009-09-18 | 2013-05-16 | Lubomir D. Bourdev | System and Method for Using Contextual Features to Improve Face Recognition in Digital Images |
| CN102004908A (en) * | 2010-11-30 | 2011-04-06 | 汉王科技股份有限公司 | Self-adapting face identification method and device |
| CN103810663A (en) * | 2013-11-18 | 2014-05-21 | 北京航天金盾科技有限公司 | Demographic data cleaning method based on face recognition |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105844203A (en) * | 2015-01-12 | 2016-08-10 | 阿里巴巴集团控股有限公司 | Human face vivo detection method and device |
| CN105844203B (en) * | 2015-01-12 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of human face in-vivo detection method and device |
| CN104966078A (en) * | 2015-07-22 | 2015-10-07 | 胡东雁 | Student identity recognition system and method for on-line training curriculum |
| CN106447949A (en) * | 2015-08-12 | 2017-02-22 | 阿里巴巴集团控股有限公司 | Information processing method and device |
| CN105389551B (en) * | 2015-10-28 | 2018-07-27 | 广东欧珀移动通信有限公司 | The detection method and detection device of eyeball identification device |
| CN105389551A (en) * | 2015-10-28 | 2016-03-09 | 广东欧珀移动通信有限公司 | Eyeball recognition device detection method and detection device |
| CN106022313A (en) * | 2016-06-16 | 2016-10-12 | 湖南文理学院 | Scene-automatically adaptable face recognition method |
| CN106257493A (en) * | 2016-08-30 | 2016-12-28 | 重庆市城投金卡信息产业股份有限公司 | Traffic promotional card falsely uses recognition methods and identification system |
| CN106257493B (en) * | 2016-08-30 | 2024-03-19 | 重庆市城投金卡信息产业(集团)股份有限公司 | Identification method and identification system for traffic preference card |
| CN106372856A (en) * | 2016-08-31 | 2017-02-01 | 北京汇通天下物联科技有限公司 | Driver work attendance method and driver work attendance system |
| CN106850667A (en) * | 2017-03-03 | 2017-06-13 | 杭州智贝信息科技有限公司 | It is a kind of to continue certification security protection system and its method |
| CN106992968A (en) * | 2017-03-03 | 2017-07-28 | 杭州智贝信息科技有限公司 | A kind of client-based face continues authentication method |
| CN107832598A (en) * | 2017-10-17 | 2018-03-23 | 广东欧珀移动通信有限公司 | Solve lock control method and Related product |
| CN107679514A (en) * | 2017-10-20 | 2018-02-09 | 维沃移动通信有限公司 | A kind of face identification method and electronic equipment |
| CN108038363A (en) * | 2017-12-05 | 2018-05-15 | 吕庆祥 | Improve the method and device of Terminal security |
| CN108549946A (en) * | 2018-04-10 | 2018-09-18 | 阳光暖果（北京）科技发展有限公司 | A kind of plant maintenance and maintenance artificial intelligence supervisory systems and method |
| CN109615750A (en) * | 2018-12-29 | 2019-04-12 | 深圳市多度科技有限公司 | The recognition of face control method and device of door access machine, access control equipment, storage medium |
| CN109815837A (en) * | 2018-12-29 | 2019-05-28 | 维沃移动通信有限公司 | A kind of face information typing control method and mobile terminal |
| CN109615750B (en) * | 2018-12-29 | 2021-12-28 | 深圳市多度科技有限公司 | Face recognition control method and device for access control machine, access control equipment and storage medium |
| CN110751758A (en) * | 2019-09-29 | 2020-02-04 | 湖北美和易思教育科技有限公司 | Intelligent lock system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104143083B (en) | 2018-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104143083A (en) | Face recognition system based on process management | |
| WO2019120115A1 (en) | Facial recognition method, apparatus, and computer apparatus | |
| US20210287472A1 (en) | Attendance management system and method, and electronic device | |
| CN110751022A (en) | Urban pet activity track monitoring method based on image recognition and related equipment | |
| CN110781805B (en) | Target object detection method, device, computing equipment and medium | |
| CN106709417A (en) | Multimodal biological recognition system and use method thereof | |
| CN207663490U (en) | A kind of mixing recognition access control system management system based on neural calculation rod | |
| CN106878344A (en) | A kind of biological characteristic authentication, register method and device | |
| CN102890777A (en) | Computer system capable of identifying facial expressions | |
| CN109919296A (en) | A kind of deep neural network training method, device and computer equipment | |
| JP2020113233A (en) | Method and apparatus for detecting abnormal behavior of groupware user | |
| CN112634017A (en) | Remote card opening activation method and device, electronic equipment and computer storage medium | |
| CN114329455B (en) | User abnormal behavior detection method and device based on heterogeneous graph embedding | |
| CN112949346A (en) | Feature library updating method and device, inference server and storage medium | |
| Yang et al. | Energy-efficient w-layer for behavior-based implicit authentication on mobile devices | |
| CN205644823U (en) | Social security self -service terminal device | |
| Adal et al. | Android based advanced attendance vigilance system using wireless network with fusion of bio-metric fingerprint authentication | |
| CN117095436A (en) | Intelligent management system and method for enterprise employee information | |
| CN116152573A (en) | Image recognition method, device, electronic equipment and computer readable storage medium | |
| Yoganathan et al. | Location Based Smart Attendance System Using GPS | |
| Batool et al. | Biometric authentication in cloud computing | |
| Yadav et al. | In-browser attendance system using face recognition and serverless edge computing | |
| CN114821844B (en) | Attendance checking method and device based on face recognition, electronic equipment and storage medium | |
| CN115248910A (en) | Identity authentication method and device applied to mobile terminal | |
| dos Santos et al. | Trends in user identity and continuous authentication |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180302 |