CN105279496A - Human face recognition method and apparatus - Google Patents
Human face recognition method and apparatus Download PDFInfo
- Publication number
- CN105279496A CN105279496A CN201510703783.4A CN201510703783A CN105279496A CN 105279496 A CN105279496 A CN 105279496A CN 201510703783 A CN201510703783 A CN 201510703783A CN 105279496 A CN105279496 A CN 105279496A
- Authority
- CN
- China
- Prior art keywords
- information
- face characteristic
- characteristic information
- face
- user facility
- 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
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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention provides a human face recognition method and apparatus. The method comprises: obtaining first user device information and first human face feature information of a human face image acquired by a front-end device; maintaining an association relationship between second user device information and second human face feature information in advance; searching the second user device information for a plurality pieces of information that matches the first user device information within a first human face feature information acquisition period, and finding a plurality pieces of the second human face feature information associated with the first user device information; and screening out the second human face feature information corresponding to the first human face feature information from the plurality pieces of the second human face feature information associated with the first user device information, and determining a human face recognition result. By means of the technical scheme of the present invention, a management server does not only determine the human face recognition result based on human face feature information of a human face image acquired by the front-end device, and does not absolutely depend on the human face feature information of the human face image, so that human face recognition accuracy can be improved and human face recognition reliability can be improved.
Description
Technical field
The present invention relates to video technique field, particularly relate to a kind of method and apparatus of recognition of face.
Background technology
In recent years, along with the develop rapidly of computing machine, network and image procossing, transmission technology, the universalness trend of video monitoring is more and more obvious, video monitoring progressively marches toward high Qinghua, intelligent, video monitoring system can be applied to various fields, as intelligent transportation, and wisdom garden, safe city etc.
Along with the fast development of computer image recognition technology, in video monitoring system, face recognition technology has had more application demand, as the face snap of gateway, face work attendance, personnel deploy to ensure effective monitoring and control of illegal activities, and face provides abundant face characteristic information, can play an important role in personnel's identification process.
In current face recognition technology, recognition of face can be carried out based on the video image comprising face.Concrete, face characteristic information is extracted from the video image comprising face, and the face characteristic information stored in the face characteristic information of extraction and face database is carried out comparison one by one, obtain the sequencing of similarity of face, and the face selecting similarity the highest from face database exports as recognition result.
Aforesaid way places one's entire reliance upon the face characteristic information of video image, and video image is very high to environmental requirement, affect by light, angle, poor anti jamming capability, recognition of face leak knowledge rate and False Rate high, whether be real human face, as the facial image on clothes, face inverted image etc. on glass door if even cannot distinguish.
Summary of the invention
The invention provides a kind of method of recognition of face, be applied in management server, said method comprising the steps of: the first face characteristic information of the first user facility information that acquisition front-end equipment collects and facial image; Safeguard the incidence relation between the second user equipment information and the second face characteristic information in advance; From the second user equipment information, search the information of mating with the multiple first user facility informations gathered in the first face characteristic information period, find the second face characteristic information that multiple first user facility information associates; From the second face characteristic information of described multiple first user facility information associations, filter out the second face characteristic information corresponding to described first face characteristic information, determine face recognition result.
Describedly from the second user equipment information, search the information of mating with the multiple first user facility informations gathered in the first face characteristic information period, before the second face characteristic information finding multiple first user facility information to associate, described method also comprises: judge that in the second all face characteristic information, the similarity whether existed between the first face characteristic information meets the second face characteristic information of preset requirement; If so, the similarity between the first face characteristic information is met the second face characteristic information of preset requirement as face recognition result; If not, the information of searching from the second user equipment information and mating with the multiple first user facility informations gathered in the first face characteristic information period is performed, the process of the second face characteristic information finding multiple first user facility information to associate.
Describedly from the second face characteristic information of described multiple first user facility informations association, filter out the second face characteristic information corresponding to described first face characteristic information, determine the process of face recognition result, specifically comprise: each second face characteristic information that described first face characteristic information associates with described multiple first user facility informations is compared, obtain described multiple each second face characteristic information of first user facility information association and the similarity of described first face characteristic information; Select the second face characteristic information that similarity is the highest, as face recognition result.
Described second user equipment information comprises one of following or combination in any: the number of the sequence number of body-worn information, mobile terminal, the MAC Address of mobile terminal, mobile terminal, SIM card sequence number; The content that described first user facility information comprises is identical with the content that described second user equipment information comprises.
The invention provides a kind of device of recognition of face, be applied in management server, described device specifically comprises: obtain module, for obtaining the first user facility information that front-end equipment collects, and obtains the first face characteristic information corresponding to facial image that described front-end equipment collects; Maintenance module, for safeguarding the incidence relation between the second user equipment information and the second face characteristic information in advance; Enquiry module, for the information that the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period mate, finds the second face characteristic information that multiple first user facility information associates; Determination module, for filtering out the second face characteristic information corresponding to described first face characteristic information in the second face characteristic information from described multiple first user facility information associations, determines face recognition result.
Described enquiry module, information also for mating at the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period, before the second face characteristic information finding multiple first user facility information to associate, judge that in the second all face characteristic information, the similarity whether existed between described first face characteristic information meets the second face characteristic information of preset requirement; If not, then the information of searching from the second user equipment information and mating with the multiple first user facility informations gathered in the first face characteristic information period is performed, the process of the second face characteristic information finding multiple first user facility information to associate; Described determination module, also for when Query Result is for being, then meets the second face characteristic information of preset requirement as face recognition result using the similarity between described first face characteristic information.
Described determination module, specifically for filtering out the second face characteristic information corresponding to described first face characteristic information in the second face characteristic information from described multiple first user facility information associations, determine in the process of face recognition result, each second face characteristic information that described first face characteristic information associates with described multiple first user facility informations is compared, obtains described multiple each second face characteristic information of first user facility information association and the similarity of described first face characteristic information; Select the second face characteristic information that similarity is the highest, as face recognition result.
Described second user equipment information comprises one of following or combination in any: the number of the sequence number of body-worn information, mobile terminal, the MAC Address of mobile terminal, mobile terminal, SIM card sequence number; The content that described first user facility information comprises is identical with the content that described second user equipment information comprises.
Based on technique scheme, in the embodiment of the present invention, the user equipment information that can collect based on front-end equipment and the face characteristic information of facial image, determine face recognition result, instead of the face characteristic information of the facial image only collected based on front-end equipment, determine face recognition result, not exclusively depend on the face characteristic information of facial image, thus by various dimensions information determination face recognition result, promote face recognition accuracy rate, reduce face and leak discrimination and False Rate, promote the reliability of recognition of face.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the method for recognition of face in one embodiment of the present invention;
Fig. 2 is the process flow diagram of the method for recognition of face in another embodiment of the present invention;
Fig. 3 is the hardware structure diagram of the management server in one embodiment of the present invention;
Fig. 4 is the structural drawing of the device of recognition of face in one embodiment of the present invention.
Embodiment
For problems of the prior art, a kind of method of recognition of face is proposed in the embodiment of the present invention, the method can be applied in the management server of video monitoring system (as VM (VideoManagement, video management) server) on, as shown in Figure 1, the method for this recognition of face can comprise the following steps:
Step 101, the first face characteristic information of the first user facility information that acquisition front-end equipment collects and facial image.Wherein, front-end equipment can collect first user facility information and facial image, and first user facility information and facial image are sent to management server.Management server, based on the first user facility information from front-end equipment, obtains the first user facility information that front-end equipment collects.Management server, based on the facial image from front-end equipment, obtains the first face characteristic information that facial image is corresponding, the first face characteristic information of the facial image that this first face characteristic information and front-end equipment collect.
Step 102, safeguards the incidence relation between the second user equipment information and the second face characteristic information in advance.
Wherein, the incidence relation between the second user equipment information and the second face characteristic information specifically can comprise: the corresponding relation between the second user equipment information and personal information, and the corresponding relation between the second face characteristic information and personal information; In the process safeguarding the corresponding relation between the second user equipment information and personal information, corresponding relation between the second face characteristic information and personal information in advance, the second user equipment information and personal information can be obtained, and the corresponding relation between the personal information storing the second user equipment information and current acquisition in the facility information storehouse of safeguarding in advance; Obtain the second face characteristic information of facial image and personal information, and the corresponding relation between the personal information storing the second face characteristic information and current acquisition in the characteristic information storehouse of safeguarding in advance.Or, obtain the second user equipment information, the second face characteristic information of facial image and personal information, and in the characteristic information storehouse of safeguarding in advance, store the second user equipment information, corresponding relation between the second face characteristic information and the personal information of current acquisition.
Step 103, the information (i.e. the second user equipment information) of mating with the multiple first user facility informations gathered in the first face characteristic information period (front-end equipment collects the time interval of facial image) is searched, the second face characteristic information finding multiple first user facility information to associate (the second face characteristic information of multiple second user equipment information associations that multiple first user facility information is corresponding) from the second user equipment information.
Step 104, filters out the second face characteristic information corresponding to the first face characteristic information, determines face recognition result from the second face characteristic information of multiple first user facility information association.Wherein, this face recognition result is face corresponding to facial image that front-end equipment collects, and the personal information of exportable correspondence.
In the embodiment of the present invention, in the information that the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period mate, before the second face characteristic information finding multiple first user facility information to associate, all the second face characteristic information (the second all face characteristic information namely stored in characteristic information storehouse can also be judged, it can comprise the second face characteristic information corresponding to first user facility information) in, the similarity whether existed between the first face characteristic information meets the second face characteristic information of preset requirement, if so, then directly the similarity between the first face characteristic information can be met the second face characteristic information of preset requirement as face recognition result, if not, then the information of searching from the second user equipment information and mating with the multiple first user facility informations gathered in the first face characteristic information period is performed, the process of the second face characteristic information finding multiple first user facility information to associate.
In the embodiment of the present invention, the information (i.e. the second user equipment information) of mating with the multiple first user facility informations gathered in the first face characteristic information period is searched from the second user equipment information, the process of the second face characteristic information finding multiple first user facility information to associate, specifically can include but not limited to as under type: utilize the multiple first user facility informations in the collection first face characteristic information period to inquire about the corresponding relation between the second user equipment information and personal information safeguarded in advance, the second user equipment information that first user facility information multiple with this mates is found from the second user equipment information, and obtain personal information corresponding to these the second user equipment informations, the personal information obtained is utilized to inquire about the corresponding relation between the second face characteristic information and personal information safeguarded in advance, obtain corresponding multiple second face characteristic information, and the second face characteristic information obtained is the second face characteristic information of multiple first user facility information association.
In the embodiment of the present invention, collect the time interval of facial image at front-end equipment, front-end equipment collects multiple first user facility information, and corresponding multiple second face characteristic information of multiple first user facility information.Based on this, the second face characteristic information corresponding to the first face characteristic information is filtered out from the second face characteristic information of multiple first user facility information association, determine the process of face recognition result, specifically can include but not limited to as under type: each second face characteristic information associated by multiple first user facility informations corresponding with the time interval collecting facial image for the first face characteristic information is compared, obtain this multiple each second face characteristic information of first user facility information association and similarity of the first face characteristic information; Select the second face characteristic information that similarity is the highest, as face recognition result.
In the said process of the embodiment of the present invention, second user equipment information specifically can include but not limited to one of following or combination in any: body-worn information (as bluetooth bracelet information, bluetooth earphone information etc.), the sequence number of mobile terminal, the MAC (MediaAccessControl of mobile terminal, media interviews control) address, mobile terminal number, SIM (SubscriberIdentityModule, client identification module) card sequence number etc.; The content that first user facility information comprises is identical with the content that the second user equipment information comprises.
Such as, when the second user equipment information is body-worn information (as bluetooth bracelet information, bluetooth earphone information etc.), then the first user facility information that front-end equipment collects is body-worn information (as bluetooth bracelet information, bluetooth earphone information etc.).When the second user equipment information is the MAC Address of mobile terminal, then when the first user facility information that front-end equipment collects is the MAC Address of mobile terminal.
Certainly, in actual applications, first user facility information and the second user equipment information are all not limited to above-mentioned user equipment information, for other user equipment information, do not repeat them here.
Based on technique scheme, in the embodiment of the present invention, the user equipment information that can collect based on front-end equipment and the face characteristic information of facial image, determine face recognition result, instead of the face characteristic information of the facial image only collected based on front-end equipment, determine face recognition result, not exclusively depend on the face characteristic information of facial image, thus by various dimensions information determination face recognition result, promote face recognition accuracy rate, reduce face and leak discrimination and False Rate, promote the reliability of recognition of face.
Below in conjunction with specific embodiment, said process is further detailed.
A kind of method of recognition of face is proposed in the embodiment of the present invention, the method can be applied in video monitoring system, this video monitoring system at least can comprise front-end equipment (as video camera) and management server (as VM server), as shown in Figure 2, the method for this recognition of face specifically can comprise the following steps:
Step 201, management server obtains the second user equipment information and personal information, and stores the corresponding relation between this second user equipment information and this personal information in the facility information storehouse of safeguarding in advance.
Wherein, the second user equipment information includes but not limited to one of following or combination in any: the body-worn information (as bluetooth bracelet information, bluetooth earphone information etc.) of personnel, the information of mobile terminal (sequence number, the MAC Address of mobile terminal, the number, SIM card sequence number etc. of mobile terminal as mobile terminal) of personnel.
Wherein, management server can store the body-worn information of personnel, corresponding relation between the information of mobile terminal of personnel and personal information in same facility information storehouse, such as, in facility information storehouse 1, store the body-worn information of personnel, corresponding relation between the information of mobile terminal of personnel and personal information.Management server also can store the corresponding relation between corresponding relation between the body-worn information of personnel and personal information, the information of mobile terminal of personnel and personal information in different facility information storehouses, such as, in facility information storehouse 1, store the corresponding relation between the body-worn information of personnel and personal information, in facility information storehouse 2, store the corresponding relation between the information of mobile terminal of personnel and personal information.
Step 202, management server obtains the second face characteristic information and the personal information of facial image, and in the characteristic information storehouse of safeguarding in advance, store the corresponding relation between this second face characteristic information and this personal information.Wherein, this facial image specifically can refer to the video image comprising face.
Wherein, can comprise this facial image in the second face characteristic information, can also comprise the out of Memory of face, as the distance between two, the distance between eyebrow, shape of face feature etc., repeat no more this.
Under embody rule scene, corresponding manner can be adopted to obtain the second user equipment information, the second face characteristic information and personal information.Such as, for certain garden, can require that personnel in garden provide self personal information (as name, I.D., sex, age etc.) and the second user equipment information, and gather the facial image (this collection is the collection of aligning personnel, and sharpness, accuracy are all very high) of these personnel.
For step 201 and step 202, management server is in characteristic information storehouse and facility information storehouse, store the corresponding relation between the second face characteristic information and personal information, the corresponding relation between the second user equipment information and personal information respectively, in actual applications, management server can also store the second user equipment information, corresponding relation between the second face characteristic information and the personal information of current acquisition in the same information bank (as characteristic information storehouse), and this process repeats no longer in detail at this.
Step 203, the first face characteristic information of the first user facility information that management server acquisition front-end equipment collects and facial image.Wherein, front-end equipment can collect first user facility information and facial image, and this first user facility information and this facial image are sent to management server.Further, management server based on the first user facility information from front-end equipment, can obtain the first user facility information that front-end equipment collects.Management server based on the facial image from front-end equipment, can also obtain the first face characteristic information that this facial image is corresponding, and this first face characteristic information is the first face characteristic information corresponding to this facial image that front-end equipment collects.
In the embodiment of the present invention, in order to distinguish conveniently, the face characteristic information stored in characteristic information storehouse is called the second face characteristic information, and the face characteristic information that the facial image collected by front-end equipment is corresponding is called the first face characteristic information.The user equipment information stored in facility information storehouse is called the second user equipment information, and the user equipment information collected by front-end equipment is called first user facility information.
Wherein, by being arranged on the imageing sensor of the front-end equipment (as video camera) of the specific region of crucial gateway, facial image can be captured, and this facial image is sent to management server by front-end equipment, obtained the first face characteristic information of the facial image that front-end equipment collects by management server.
Wherein, (can be arranged on front-end equipment by the sensing detecting module of the front-end equipment (as video camera) being arranged on the specific region of crucial gateway, as WIFI (WirelessFidelity, Wireless Fidelity) wireless module, bluetooth module etc.), first user facility information can be collected, and this first user facility information is sent to management server by front-end equipment, the first user facility information collected by management server acquisition front-end equipment, and the content that first user facility information comprises is identical with the content that the second user equipment information comprises.Such as, when the second user equipment information is body-worn information (as bluetooth bracelet information, bluetooth earphone information etc.), then the first user facility information that front-end equipment collects is body-worn information (as bluetooth bracelet information, bluetooth earphone information etc.).When the second user equipment information is the MAC Address of mobile terminal, then when the first user facility information that front-end equipment collects is the MAC Address of mobile terminal.Certainly, in actual applications, first user facility information and the second user equipment information are all not limited to above-mentioned user equipment information, for other user equipment information, do not repeat them here.
Wherein, because the user being positioned at the same area may be multiple, therefore in each collection moment, the first user facility information that front-end equipment collects can be all multiple first user facility information.
Wherein, the collection moment of facial image, when sending to management server the facial image collected, is also sent to management server by front-end equipment, by the first face characteristic information and the corresponding relation gathering the moment of record management server facial image.The collection moment of first user facility information, when sending to management server the first user facility information collected, can also be sent to management server by front-end equipment, by record management server first user facility information and the corresponding relation gathering the moment.
Wherein, first user facility information is gathered for the WIFI wireless module by front-end equipment, WIFI wireless module periodically sends Beacon (detection) message, when mobile terminal is positioned at the investigative range of WIFI wireless module, then mobile terminal can return response message to WIFI wireless module, WIFI wireless module can obtain first user facility information from the response message that mobile terminal returns, and obtains the collection moment of first user facility information.Or, mobile terminal active period sends probe requests thereby message, when mobile terminal is positioned at the investigative range of WIFI wireless module, the probe requests thereby message that then mobile terminal sends can be received by WIFI wireless module, WIFI wireless module can obtain first user facility information from from the probe requests thereby message of mobile terminal, and obtains the collection moment of first user facility information.
Wherein, when front-end equipment sends to management server the first user facility information collected, in the time period T that can arrange, front-end equipment only sends a first user facility information to management server.
Step 204, management server searches the information of mating with the multiple first user facility informations gathered in the first face characteristic information period (namely front-end equipment collects the time interval of facial image) from the second user equipment information, the second face characteristic information finding multiple first user facility information to associate (the second face characteristic information of multiple second user equipment information associations that multiple first user facility information is corresponding).
In the embodiment of the present invention, management server searches the information (i.e. the second user equipment information) of mating with the multiple first user facility informations gathered in the first face characteristic information period from the second user equipment information, the process of the second face characteristic information finding multiple first user facility information to associate, specifically can include but not limited to as under type: management server utilizes the multiple first user facility informations in the collection first face characteristic information period to inquire about the corresponding relation between the second user equipment information and personal information safeguarded in advance, management server finds the second user equipment information that first user facility information multiple with this mates from the second user equipment information, and obtain personal information corresponding to these the second user equipment informations, management server utilizes the personal information obtained to inquire about the corresponding relation between the second face characteristic information and personal information safeguarded in advance, obtain corresponding multiple second face characteristic information, and the second face characteristic information that management server obtains is the second face characteristic information of multiple first user facility information association.
Step 205, management server filters out the second face characteristic information corresponding to the first face characteristic information from the second face characteristic information that multiple first user facility information associates, and determines face recognition result.Wherein, this face recognition result is face corresponding to facial image that front-end equipment collects, and management server, after obtaining face recognition result, can also export the personal information that face recognition result is corresponding.
Wherein, the time interval of facial image is collected at front-end equipment, front-end equipment collects multiple first user facility information, the corresponding relation between the second user equipment information and personal information that management server utilizes multiple first user facility information to inquire about to safeguard in advance, obtain multiple personal information, and the corresponding relation between the second face characteristic information and personal information utilizing multiple personal information to inquire about to safeguard in advance, obtain multiple second face characteristic information, i.e. corresponding multiple second face characteristic information of multiple first user facility information.
Based on this, in the embodiment of the present invention, the second face characteristic information corresponding to the first face characteristic information is filtered out the second face characteristic information (multiple second face characteristic information) that management server associates from multiple first user facility information, determine the process of face recognition result, specifically can include but not limited to as under type: each second face characteristic information associated by multiple first user facility informations corresponding with the time interval collecting facial image for the first face characteristic information is compared by management server, obtain this multiple each second face characteristic information of first user facility information association and similarity of the first face characteristic information, management server selects the second face characteristic information that similarity is the highest, as face recognition result.
Wherein, management server is after the first face characteristic information obtaining facial image, determine the collection moment of this facial image, consider the impact of the factors such as error, the time interval (namely comprising the time interval in the collection moment of facial image) collecting facial image can also be determined, and collecting the time interval of facial image, management server can determine multiple first user facility information.
In the embodiment of the present invention, the information that management server is mated at the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period, before the second face characteristic information finding multiple first user facility information to associate, management server can also judge all the second face characteristic information (the second all face characteristic information namely stored in characteristic information storehouse, it can comprise the second face characteristic information corresponding to first user facility information) in, the similarity whether existed between the first face characteristic information meets the second face characteristic information of preset requirement.If so, then the similarity between the first face characteristic information directly can be met the second face characteristic information of preset requirement as face recognition result by management server, no longer performs above-mentioned steps 204 and step 205.If not, then management server performs and from the second user equipment information, searches the information of mating with the multiple first user facility informations gathered in the first face characteristic information period, the process of the second face characteristic information finding multiple first user facility information to associate.
Based on aforesaid way, in the embodiment of the present invention, management server first judges that, in the second all face characteristic information, the similarity whether existed between the first face characteristic information meets the second face characteristic information of preset requirement.When the first face characteristic information of video image is more clear, then all the second face characteristic information similarities that can exist between the first face characteristic information meet the second face characteristic information of preset requirement, directly the similarity between the first face characteristic information can be met the second face characteristic information of preset requirement as face recognition result like this, thus accurately draw face recognition result.Such as, when obtain clearly the first face characteristic information 1 time, even if the second face characteristic information that first user facility information is corresponding is the second face characteristic information 1, second face characteristic information 2, second face characteristic information 3, then management server is also first judge in all the second face characteristic information (as the second face characteristic information 1-second face characteristic information 1000), the similarity whether existed between the first face characteristic information 1 meets the second face characteristic information of preset requirement, suppose that the similarity between the second face characteristic information 100 and the first face characteristic information meets preset requirement, then using the second face characteristic information 100 as face recognition result, this face recognition result can be a face recognition result accurately, now, management server does not need from the second face characteristic information 1, second face characteristic information 2, the face recognition result of the first face characteristic information 1 correspondence is selected in second face characteristic information 3, namely face recognition result accurately can be obtained.
Wherein, management server is after the first face characteristic information obtaining facial image, can also analyze this facial image, to detect whether this facial image meets recognition of face requirement, whether the light as facial image meets the requirements, whether angle meets the requirements, whether two eye distances from meeting the requirements, management server can use Adaboost detection of classifier facial image whether to meet recognition of face requirement, and concrete testing process repeats no longer in detail.If facial image does not meet recognition of face requirement, then the first face characteristic information and each second face characteristic information are compared by management server, obtain the similarity of each second face characteristic information and the first face characteristic information, and the second face characteristic information selecting similarity the highest, as face recognition result.Or management server, directly by each second face characteristic information, as face recognition result, now has multiple face recognition result.When the first face characteristic information and each second face characteristic information are compared by management server, the unique point quantity of comparison can be reduced, such as, during normal face characteristic information comparison, need 100 unique points, when facial image does not meet recognition of face requirement, when face characteristic information comparison, only use 20 unique points.
In a kind of embody rule, if facial image meets recognition of face requirement, then all second face characteristic information (as the second face characteristic information 1-second face characteristic information 1000) in the first face characteristic information and characteristic information storehouse are compared by management server, obtain the similarity of each second face characteristic information in characteristic information storehouse and the first face characteristic information, if there is similarity to exceed setting threshold value (illustrating that the second face characteristic information corresponding to this similarity is the second face characteristic information that similarity between the first face characteristic information meets preset requirement), then select the second face characteristic information that similarity is the highest, as face recognition result, no longer perform above-mentioned steps 204 and step 205, if do not have similarity to exceed setting threshold value, then each second face characteristic information (as the second face characteristic information 1, second face characteristic information 2, second face characteristic information 3) corresponding with first user facility information for the first face characteristic information is compared, obtain the similarity of each second face characteristic information corresponding to first user facility information and the first face characteristic information, and the second face characteristic information selecting similarity the highest, as face recognition result.
Wherein, two face characteristic information are being compared, when obtaining the similarity of two face characteristic information, similarity algorithm can be adopted to process, such as, similarity algorithm can perception hash algorithm, SIM (StructuralSIMilarity, structural similarity) algorithm etc., repeats no longer in detail at this.
Based on technique scheme, in the embodiment of the present invention, the user equipment information that can collect based on front-end equipment and the face characteristic information of facial image, determine face recognition result, instead of the face characteristic information of the facial image only collected based on front-end equipment, determine face recognition result, not exclusively depend on the face characteristic information of facial image, thus by various dimensions information determination face recognition result, promote face recognition accuracy rate, reduce face and leak discrimination and False Rate, promote the reliability of recognition of face.
Based on the inventive concept same with said method, additionally provide a kind of device of recognition of face in the embodiment of the present invention, the application of installation of this recognition of face on the management server.Wherein, the device of this recognition of face can pass through software simulating, also can be realized by the mode of hardware or software and hardware combining.For software simulating, as the device on a logical meaning, be the processor of the management server by its place, computer program instructions corresponding in nonvolatile memory read in internal memory to run and formed.Say from hardware view, as shown in Figure 3, for a kind of hardware structure diagram of the management server at the device place of the recognition of face of the present invention's proposition, except the processor shown in Fig. 3, network interface, internal memory and nonvolatile memory, management server can also comprise other hardware, as the forwarding chip etc. of responsible process message; From hardware configuration, this management server may be also distributed apparatus, may comprise multiple interface card, to carry out the expansion of Message processing at hardware view.
As shown in Figure 4, for the structural drawing of the device of the recognition of face of the present invention's proposition, described device specifically comprises: obtain module 11, for obtaining the first user facility information that front-end equipment collects, and obtains the first face characteristic information corresponding to facial image that described front-end equipment collects; Maintenance module 12, for safeguarding the incidence relation between the second user equipment information and the second face characteristic information in advance; Enquiry module 13, for the information that the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period mate, finds the second face characteristic information that multiple first user facility information associates; Determination module 14, for filtering out the second face characteristic information corresponding to described first face characteristic information in the second face characteristic information from described multiple first user facility information associations, determines face recognition result.
Described enquiry module 13, information also for mating at the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period, before the second face characteristic information finding multiple first user facility information to associate, judge that in the second all face characteristic information, the similarity whether existed between described first face characteristic information meets the second face characteristic information of preset requirement; If not, then the information of searching from the second user equipment information and mating with the multiple first user facility informations gathered in the first face characteristic information period is performed, the process of the second face characteristic information finding multiple first user facility information to associate; Described determination module 14, also for when Query Result is for being, meets the second face characteristic information of preset requirement as face recognition result using the similarity between described first face characteristic information.
Described determination module 14, specifically for filtering out the second face characteristic information corresponding to described first face characteristic information in the second face characteristic information from described multiple first user facility information associations, determine in the process of face recognition result, each second face characteristic information that described first face characteristic information associates with described multiple first user facility informations is compared, obtains described multiple each second face characteristic information of first user facility information association and the similarity of described first face characteristic information;
Select the second face characteristic information that similarity is the highest, as face recognition result.
In the embodiment of the present invention, described second user equipment information comprises one of following or combination in any: the number of the sequence number of body-worn information, mobile terminal, the MAC address of mobile terminal, mobile terminal, client identification module SIM card sequence number; The content that described first user facility information comprises is identical with the content that described second user equipment information comprises.
Wherein, the modules of apparatus of the present invention can be integrated in one, and also can be separated deployment.Above-mentioned module can merge into a module, also can split into multiple submodule further.
Through the above description of the embodiments, those skilled in the art can be well understood to the mode that the present invention can add required general hardware platform by software and realize, and can certainly pass through hardware, but in a lot of situation, the former is better embodiment.Based on such understanding, technical scheme of the present invention can embody with the form of software product the part that prior art contributes in essence in other words, this computer software product is stored in a storage medium, comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform method described in each embodiment of the present invention.It will be appreciated by those skilled in the art that accompanying drawing is the schematic diagram of a preferred embodiment, the module in accompanying drawing or flow process might not be that enforcement the present invention is necessary.
It will be appreciated by those skilled in the art that the module in the device in embodiment can carry out being distributed in the device of embodiment according to embodiment description, also can carry out respective change and be arranged in the one or more devices being different from the present embodiment.The module of above-described embodiment can merge into a module, also can split into multiple submodule further.The invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.
Be only several specific embodiment of the present invention above, but the present invention is not limited thereto, the changes that any person skilled in the art can think of all should fall into protection scope of the present invention.
Claims (8)
1. a method for recognition of face, is applied in management server, it is characterized in that, said method comprising the steps of:
First face characteristic information of the first user facility information that acquisition front-end equipment collects and facial image;
Safeguard the incidence relation between the second user equipment information and the second face characteristic information in advance;
From the second user equipment information, search the information of mating with the multiple first user facility informations gathered in the first face characteristic information period, find the second face characteristic information that multiple first user facility information associates;
From the second face characteristic information of described multiple first user facility information associations, filter out the second face characteristic information corresponding to described first face characteristic information, determine face recognition result.
2. method according to claim 1, it is characterized in that, describedly from the second user equipment information, search the information of mating with the multiple first user facility informations gathered in the first face characteristic information period, before the second face characteristic information finding multiple first user facility information to associate, described method also comprises:
Judge that in the second all face characteristic information, the similarity whether existed between the first face characteristic information meets the second face characteristic information of preset requirement; If so, the similarity between the first face characteristic information is met the second face characteristic information of preset requirement as face recognition result; If not, the information of searching from the second user equipment information and mating with the multiple first user facility informations gathered in the first face characteristic information period is performed, the process of the second face characteristic information finding multiple first user facility information to associate.
3. method according to claim 1, it is characterized in that, describedly from the second face characteristic information of described multiple first user facility informations association, filter out the second face characteristic information corresponding to described first face characteristic information, determine the process of face recognition result, specifically comprise:
Each second face characteristic information that described first face characteristic information associates with described multiple first user facility informations is compared, obtains described multiple each second face characteristic information of first user facility information association and the similarity of described first face characteristic information;
Select the second face characteristic information that similarity is the highest, as face recognition result.
4. the method according to any one of claim 1-3, it is characterized in that, described second user equipment information comprises one of following or combination in any: the number of the sequence number of body-worn information, mobile terminal, the MAC address of mobile terminal, mobile terminal, client identification module SIM card sequence number; The content that described first user facility information comprises is identical with the content that described second user equipment information comprises.
5. a device for recognition of face, is applied in management server, it is characterized in that, described device specifically comprises:
Obtaining module, for obtaining the first user facility information that front-end equipment collects, and obtaining the first face characteristic information corresponding to facial image that described front-end equipment collects;
Maintenance module, for safeguarding the incidence relation between the second user equipment information and the second face characteristic information in advance;
Enquiry module, for the information that the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period mate, finds the second face characteristic information that multiple first user facility information associates;
Determination module, for filtering out the second face characteristic information corresponding to described first face characteristic information in the second face characteristic information from described multiple first user facility information associations, determines face recognition result.
6. device according to claim 5, is characterized in that,
Described enquiry module, information also for mating at the multiple first user facility informations searched from the second user equipment information with gather in the first face characteristic information period, before the second face characteristic information finding multiple first user facility information to associate, judge that in the second all face characteristic information, the similarity whether existed between described first face characteristic information meets the second face characteristic information of preset requirement; If not, then the information of searching from the second user equipment information and mating with the multiple first user facility informations gathered in the first face characteristic information period is performed, the process of the second face characteristic information finding multiple first user facility information to associate;
Described determination module, also for when Query Result is for being, then meets the second face characteristic information of preset requirement as face recognition result using the similarity between described first face characteristic information.
7. device according to claim 5, is characterized in that,
Described determination module, specifically for filtering out the second face characteristic information corresponding to described first face characteristic information in the second face characteristic information from described multiple first user facility information associations, determine in the process of face recognition result, each second face characteristic information that described first face characteristic information associates with described multiple first user facility informations is compared, obtains described multiple each second face characteristic information of first user facility information association and the similarity of described first face characteristic information;
Select the second face characteristic information that similarity is the highest, as face recognition result.
8. according to the arbitrary described device of claim 5-7, it is characterized in that, described second user equipment information comprises one of following or combination in any: the number of the sequence number of body-worn information, mobile terminal, the MAC address of mobile terminal, mobile terminal, client identification module SIM card sequence number; The content that described first user facility information comprises is identical with the content that described second user equipment information comprises.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510703783.4A CN105279496B (en) | 2015-10-26 | 2015-10-26 | A kind of method and apparatus of recognition of face |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510703783.4A CN105279496B (en) | 2015-10-26 | 2015-10-26 | A kind of method and apparatus of recognition of face |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105279496A true CN105279496A (en) | 2016-01-27 |
| CN105279496B CN105279496B (en) | 2019-10-18 |
Family
ID=55148480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510703783.4A Active CN105279496B (en) | 2015-10-26 | 2015-10-26 | A kind of method and apparatus of recognition of face |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105279496B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105872979A (en) * | 2016-05-31 | 2016-08-17 | 王方松 | Method and device for obtaining crowd information in set place |
| CN106408013A (en) * | 2016-09-12 | 2017-02-15 | 湖南数据魔方信息科技有限公司 | Method for applying video-based human body characteristic identification technology in wireless positioning |
| CN106548148A (en) * | 2016-10-26 | 2017-03-29 | 北京意泰物联软件科技有限公司 | The recognition methodss of unknown face and system in video |
| CN108228742A (en) * | 2017-12-15 | 2018-06-29 | 深圳市商汤科技有限公司 | Face duplicate checking method and apparatus, electronic equipment, medium, program |
| CN108271069A (en) * | 2017-12-11 | 2018-07-10 | 北京奇艺世纪科技有限公司 | The segment filter method and device of a kind of video frequency program |
| CN108540760A (en) * | 2017-03-01 | 2018-09-14 | 中国电信股份有限公司 | Video monitoring recognition methods, device and system |
| CN109766736A (en) * | 2017-11-09 | 2019-05-17 | 中国电信股份有限公司 | Face identification method, device and system |
| CN109963113A (en) * | 2017-12-14 | 2019-07-02 | 浙江宇视科技有限公司 | A kind of monitoring method and device of interesting target |
| CN110825765A (en) * | 2019-10-23 | 2020-02-21 | 中国建设银行股份有限公司 | Face recognition method and device |
| CN113158823A (en) * | 2021-03-30 | 2021-07-23 | 浙江大华技术股份有限公司 | Face recognition method and device |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073859A (en) * | 2011-02-14 | 2011-05-25 | 汉王科技股份有限公司 | Child picking and sending system on basis of face identification and control method thereof |
| CN102184391A (en) * | 2011-05-19 | 2011-09-14 | 汉王科技股份有限公司 | Distributed type face recognition method and system as well as face recognition terminal |
| CN102360421A (en) * | 2011-10-19 | 2012-02-22 | 苏州大学 | Face identification method and system based on video streaming |
| US8244002B2 (en) * | 2009-03-18 | 2012-08-14 | Industrial Technology Research Institute | System and method for performing rapid facial recognition |
| CN103824058A (en) * | 2014-02-26 | 2014-05-28 | 杨勇 | Face recognition system and method based on locally distributed linear embedding algorithm |
| CN103995818A (en) * | 2013-02-15 | 2014-08-20 | 三星电子株式会社 | Electronic device and object recognition method in electronic device |
| CN104008317A (en) * | 2013-02-22 | 2014-08-27 | 富士施乐株式会社 | Authentication apparatus and authentication method |
| CN104021398A (en) * | 2014-06-16 | 2014-09-03 | 北京云视智通科技有限公司 | Wearable intelligent device and method for assisting identity recognition |
| CN104318217A (en) * | 2014-10-28 | 2015-01-28 | 吴建忠 | Face recognition information analysis method and system based on distributed cloud computing |
| CN104765739A (en) * | 2014-01-06 | 2015-07-08 | 南京宜开数据分析技术有限公司 | Large-scale face database searching method based on shape space |
| CN105205457A (en) * | 2015-09-10 | 2015-12-30 | 上海卓易科技股份有限公司 | Information acquisition system and method based on face recognition |
| CN106611151A (en) * | 2015-10-23 | 2017-05-03 | 杭州海康威视数字技术股份有限公司 | A human face identification method and apparatus |
-
2015
- 2015-10-26 CN CN201510703783.4A patent/CN105279496B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8244002B2 (en) * | 2009-03-18 | 2012-08-14 | Industrial Technology Research Institute | System and method for performing rapid facial recognition |
| CN102073859A (en) * | 2011-02-14 | 2011-05-25 | 汉王科技股份有限公司 | Child picking and sending system on basis of face identification and control method thereof |
| CN102184391A (en) * | 2011-05-19 | 2011-09-14 | 汉王科技股份有限公司 | Distributed type face recognition method and system as well as face recognition terminal |
| CN102360421A (en) * | 2011-10-19 | 2012-02-22 | 苏州大学 | Face identification method and system based on video streaming |
| CN102360421B (en) * | 2011-10-19 | 2014-05-28 | 苏州大学 | Face identification method and system based on video streaming |
| CN103995818A (en) * | 2013-02-15 | 2014-08-20 | 三星电子株式会社 | Electronic device and object recognition method in electronic device |
| CN104008317A (en) * | 2013-02-22 | 2014-08-27 | 富士施乐株式会社 | Authentication apparatus and authentication method |
| CN104765739A (en) * | 2014-01-06 | 2015-07-08 | 南京宜开数据分析技术有限公司 | Large-scale face database searching method based on shape space |
| CN103824058A (en) * | 2014-02-26 | 2014-05-28 | 杨勇 | Face recognition system and method based on locally distributed linear embedding algorithm |
| CN104021398A (en) * | 2014-06-16 | 2014-09-03 | 北京云视智通科技有限公司 | Wearable intelligent device and method for assisting identity recognition |
| CN104318217A (en) * | 2014-10-28 | 2015-01-28 | 吴建忠 | Face recognition information analysis method and system based on distributed cloud computing |
| CN105205457A (en) * | 2015-09-10 | 2015-12-30 | 上海卓易科技股份有限公司 | Information acquisition system and method based on face recognition |
| CN106611151A (en) * | 2015-10-23 | 2017-05-03 | 杭州海康威视数字技术股份有限公司 | A human face identification method and apparatus |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105872979A (en) * | 2016-05-31 | 2016-08-17 | 王方松 | Method and device for obtaining crowd information in set place |
| CN105872979B (en) * | 2016-05-31 | 2019-11-26 | 王方松 | A kind of method and device obtaining crowd's information in setting place |
| CN106408013A (en) * | 2016-09-12 | 2017-02-15 | 湖南数据魔方信息科技有限公司 | Method for applying video-based human body characteristic identification technology in wireless positioning |
| CN106548148A (en) * | 2016-10-26 | 2017-03-29 | 北京意泰物联软件科技有限公司 | The recognition methodss of unknown face and system in video |
| CN108540760A (en) * | 2017-03-01 | 2018-09-14 | 中国电信股份有限公司 | Video monitoring recognition methods, device and system |
| CN109766736A (en) * | 2017-11-09 | 2019-05-17 | 中国电信股份有限公司 | Face identification method, device and system |
| CN108271069A (en) * | 2017-12-11 | 2018-07-10 | 北京奇艺世纪科技有限公司 | The segment filter method and device of a kind of video frequency program |
| CN109963113B (en) * | 2017-12-14 | 2021-06-22 | 浙江宇视科技有限公司 | Method and device for monitoring interested target |
| CN109963113A (en) * | 2017-12-14 | 2019-07-02 | 浙江宇视科技有限公司 | A kind of monitoring method and device of interesting target |
| CN108228742A (en) * | 2017-12-15 | 2018-06-29 | 深圳市商汤科技有限公司 | Face duplicate checking method and apparatus, electronic equipment, medium, program |
| CN110825765A (en) * | 2019-10-23 | 2020-02-21 | 中国建设银行股份有限公司 | Face recognition method and device |
| CN110825765B (en) * | 2019-10-23 | 2022-10-04 | 中国建设银行股份有限公司 | Face recognition method and device |
| CN113158823A (en) * | 2021-03-30 | 2021-07-23 | 浙江大华技术股份有限公司 | Face recognition method and device |
| CN113158823B (en) * | 2021-03-30 | 2023-05-12 | 浙江大华技术股份有限公司 | Face recognition method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105279496B (en) | 2019-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105279496A (en) | Human face recognition method and apparatus | |
| CN107506684B (en) | Gait recognition method and device | |
| US8116534B2 (en) | Face recognition apparatus and face recognition method | |
| CN107392222B (en) | Face clustering method and device and storage medium | |
| TWI743835B (en) | Information processing method, device and storage medium thereof | |
| CN105631430A (en) | Matching method and apparatus for face image | |
| CN107292240A (en) | It is a kind of that people's method and system are looked for based on face and human bioequivalence | |
| CN109740004B (en) | Filing method and device | |
| US8266174B2 (en) | Behavior history retrieval apparatus and behavior history retrieval method | |
| JP6254836B2 (en) | Image search apparatus, control method and program for image search apparatus | |
| US20210357678A1 (en) | Information processing method and apparatus, and storage medium | |
| CN102119343B (en) | Target detection device and target detection method | |
| US9947105B2 (en) | Information processing apparatus, recording medium, and information processing method | |
| US20130121535A1 (en) | Detection device and method for transition area in space | |
| CN109783685A (en) | A kind of querying method and device | |
| CN107886079A (en) | Object identifying method, apparatus and system | |
| CN109815775A (en) | A kind of face identification method and system based on face character | |
| JP2018011263A (en) | Monitor system, monitor camera, and management device | |
| CN108563651B (en) | Multi-video target searching method, device and equipment | |
| CN103530648A (en) | Face recognition method based on multi-frame images | |
| CN109800664B (en) | Method and device for determining passersby track | |
| CN109784220B (en) | Method and device for determining passerby track | |
| CN110909612A (en) | Gait recognition method and system based on deep neural network and machine vision | |
| CN103745223A (en) | Face detection method and apparatus | |
| CN110825893A (en) | Target searching method, device, system and storage medium |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200602 Address after: 250001 whole floor, building 3, Aosheng building, 1166 Xinluo street, Jinan area, Jinan pilot Free Trade Zone, Shandong Province Patentee after: Jinan boguan Intelligent Technology Co., Ltd Address before: Hangzhou City, Zhejiang province 310051 Binjiang District West Street Jiangling Road No. 88 building 10 South Block 1-11 Patentee before: ZHEJIANG UNIVIEW TECHNOLOGIES Co.,Ltd. |
|
| TR01 | Transfer of patent right |