Summary of the invention
The cash method and system of member system provided by the invention based on recognition of face, main purpose are to overcome existingThe problem of having membership information determination process cumbersome, leading to cash register low efficiency.
In order to solve the above technical problems, the present invention adopts the following technical scheme:
A kind of cash method of the member system based on recognition of face, includes the following steps；
Photographic device by the way that cashier is arranged in obtains the first facial image of consumer, by first facial imageIt is uploaded to Cloud Server, and carries out the identification of face identity to first facial image, obtains the second facial image and correspondenceMember ID；
The third facial image that consumer is obtained again through the photographic device, by second facial image, thirdFacial image and member ID are transmitted to the cash register terminal of cash register, in cash register terminal, according to second facial image and the third partyFace image, judges whether current consumer bound member；
If current consumer bound member, membership information is obtained from Cloud Server according to member ID；
If current consumer is unbound to cross member, the identity information and third facial image of current consumer are obtained, andMember registration binding is carried out according to the identity information and third facial image, obtains membership information；
Cash register is completed according to the consumption information of the membership information and consumer.
As an embodiment, first facial image for obtaining consumer, comprises at least one of the following；
Photographic device by the way that cashier is arranged in obtains the video information of consumer, and passes through the face box pair of settingVideo information carries out analysis identification, obtains the first facial image；
Photographic device by the way that cashier is arranged in obtains the image information of consumer, and knows to described image informationNot, the first facial image is obtained.
As an embodiment, the third facial image that consumer is obtained again through the photographic device,Include the following steps；
The image information that consumer is obtained again through the photographic device that cashier is arranged in, by preset interpupillary distance to eachThe interpupillary distance of image information identification is screened, and obtains third facial image according to the selection result, and by the third facial imageIt shows in cash register terminal.
As an embodiment, the identification that face identity is carried out to first facial image, obtains secondFacial image and corresponding member ID, include the following steps；
The mapping relations of pre-set each face characteristic image and member ID in Cloud Server, by first faceImage is matched with each face characteristic image；It chooses and the highest face characteristic image of the degree of approximation in matching result is determined as theTwo facial images, and by membership information corresponding with second facial image present in the mapping relations, be determined as withThe member ID that second facial image matches.
As an embodiment, the membership information be name, cell-phone number, gender, the birthday, face characteristic image withAnd one of member ID or a variety of.
Correspondingly, the present invention also provides a kind of cash register system of member system based on recognition of face, including identification module, sentenceDisconnected module obtains module, binding module and cash register module；
The identification module obtains the first facial image of consumer for the photographic device by the way that cashier is arranged in,First facial image is uploaded to Cloud Server, and carries out the identification of face identity to first facial image, is obtainedSecond facial image and corresponding member ID；
The judgment module will be described for obtaining the third facial image of consumer again through the photographic deviceSecond facial image, third facial image and member ID are transmitted to the cash register terminal of cash register, in cash register terminal, according to described secondFacial image and third facial image, judge whether current consumer bound member；
The acquisition module obtained meeting from Cloud Server according to member ID if binding member for current consumerMember's information；
The binding module, if being used for, current consumer is unbound to cross member, obtains the identity information of current consumerWith third facial image, and member registration binding is carried out according to the identity information and third facial image, obtains membership information；
The cash register module, for completing cash register according to the consumption information of the membership information and consumer.
As an embodiment, the identification module is at least one in the first recognition unit and the second recognition unitKind；
First recognition unit obtains the video information of consumer for the photographic device by the way that cashier is arranged in,And the face box by being arranged carries out analysis identification to video information, obtains the first facial image；
Second recognition unit obtains the image information of consumer for the photographic device by the way that cashier is arranged in,And described image information is identified, obtain the first facial image.
As an embodiment, the judgment module includes screening unit；
The screening unit, for obtaining the image information of consumer again through the photographic device that cashier is arranged in,It is screened by the interpupillary distance that preset interpupillary distance identifies each image information, third facial image is obtained according to the selection result, andThe third facial image is shown in cash register terminal.
As an embodiment, the identification module is also used to；
The mapping relations of pre-set each face characteristic image and member ID in Cloud Server, by first faceImage is matched with each face characteristic image；It chooses and the highest face characteristic image of the degree of approximation in matching result is determined as theTwo facial images, and by membership information corresponding with second facial image present in the mapping relations, be determined as withThe member ID that second facial image matches.
As an embodiment, the membership information be name, cell-phone number, gender, the birthday, face characteristic image withAnd one of member ID or a variety of.
Compared with prior art, the technical program has the advantage that
The cash method and system of member system provided by the invention based on recognition of face, by the way that taking the photograph for cashier is arranged inAs the facial image of device acquisition consumer, and member's identification is carried out according to facial image automatically, so that in confirmation membership informationShi Wuxu consumer actively shows member card, to improve the confirmation efficiency of membership information, improves cash register efficiency；And utilize weightThe third facial image that newly obtains judges whether consumer bound member, so that identification is more accurate.
Specific embodiment
Below in conjunction with attached drawing, the technical characteristic and advantage above-mentioned and other to the present invention are clearly and completely described,Obviously, described embodiment is only section Example of the invention, rather than whole embodiments.
Referring to Fig. 1, the cash method for the member system based on recognition of face that the embodiment of the present invention one provides, including it is followingStep；
S100, the first facial image that consumer is obtained by the photographic device that cashier is arranged in, by the first face figureAs being uploaded to Cloud Server, and the identification of face identity is carried out to the first facial image, obtain the second facial image and correspondingMember ID；
S200, the third facial image that consumer is obtained again through photographic device, by the second facial image, third faceImage and member ID are transmitted to the cash register terminal of cash register, in cash register terminal, according to the second facial image and third facial image, sentenceWhether disconnected current consumer bound member；
If S300, current consumer bound member, membership information is obtained from Cloud Server according to member ID；
If S400, the unbound member excessively of current consumer, obtain the identity information and third face figure of current consumerPicture, and member registration binding is carried out according to identity information and third facial image, obtain membership information；
S500, cash register is completed according to the consumption information of membership information and consumer.
It should be noted that the present invention only needs to be arranged in cashier to adopt in real time for existing cash register systemCollection, and the photographic device connecting with cash register terminal or Cloud Server, photographic device can be video camera, camera and have camera shootingThe devices such as the face box of function.Existing cash register system can't be caused stress, there is very strong Generalization Ability.ForFor one facial image and the second facial image, they are obtained by photographic device, in this present embodiment, for obtainingMeans and analysis means of identification and be not limited.For example, can be carried out by open source opencv for the first facial imageVideo flowing analysis and recognition of face.
Database has been preset in Cloud Server, in the database, the corresponding second face figure of each membership information in storagePicture and member ID have mapping relations between this three, that is to say, that by one of information, can find out uniqueFor other two information.For example, obtaining membership information from Cloud Server according to member ID by reflection mechanism.It is however emphasized thatBe that member ID is unique identity, and membership information be name, cell-phone number, gender, the birthday, face characteristic image andOne of member ID or a variety of.For data such as name, cell-phone number, gender and birthdays, it can be and inputted by cash register terminal.Be only registered members with complete membership information, complete membership information refer at least having facial image, member ID withAnd identity can be characterized, such as one of name, cell-phone number, gender, birthday.To be subsequent analysis member behavior and battalionPin activity provides data supporting.In other embodiments, can using other data in membership information as member ID come using.Certainly, the data in database can be downloaded in cash register terminal from Cloud Server, avoid service efficiency caused by network transmission lowThe problem of.After each member registration binds and generates membership information, the membership information in Cloud Server can be all updated, fromAnd guaranteeing the second facial image obtained every time has high accuracy.
And judging whether current consumer bound member, it is that will reacquire obtained third facial image and knowIt is carried out between the second facial image not obtained.It should be appreciated that when carrying out images match identification according to the data in database, it is rightImage pattern requirement is very high in database, and in general, the image pattern in database is more, and identification accuracy is got overIt is high.In this present embodiment, the recognizer and image pattern that are related to for database are simultaneously not limited, but the present invention is baseFoundation will lead to due to the influence of the factors such as live light, makeup in the image information that the real-time collection analysis of photographic device obtainsWhether the second facial image that the first facial image identifies is that customer himself and its whether register bound member's presenceCertain error.And the present invention is to reacquire to obtain third facial image to judge whether current consumer bound member, fromAnd the requirement to sample size, quality and recognizer in database is reduced, it can be realized quickly judging for member's bindingJourney improves recognition efficiency to reduce operating pressure.It, can also be according to name, cell-phone number, gender, life in other embodimentsWhether one of day and member ID or a variety of inquiry consumers bound member.If binding member, basis is obtainedThe third facial image and existing member data taken regenerates complete membership information, that is to say, that by facial image and meetingMember's information is bound.
The cash method and system of member system provided by the invention based on recognition of face, by the way that taking the photograph for cashier is arranged inAs the facial image of device acquisition consumer, and member's identification is carried out according to facial image automatically, so that in confirmation membership informationShi Wuxu consumer actively shows member card, to improve the confirmation efficiency of membership information, improves cash register efficiency；And utilize weightThe third facial image that newly obtains judges whether consumer bound member, so that identification is more accurate.
For the first facial image and the second facial image, they obtain means be all it is identical, difference is,The timing node of acquisition and specific analysis identification process.Certainly, what is obtained is all the face with consumer's face featureImage.
For obtain consumer facial image means there are many, in this present embodiment, one kind at least below；ItsIn one is by be arranged in cashier photographic device obtain consumer video information, and by setting face box pairVideo information carries out analysis identification, obtains the first facial image；Existing common photographic device is directlyed adopt to acquire view in real timeFrequency information, the algorithm that presupposition analysis identifies in face box, to carry out analysis identification to video information.Another is to pass throughThe photographic device that cashier is arranged in obtains the image information of consumer, and identifies to image information, obtains the first faceImage.Namely realized using the camera with facial image identification function.To reduce the operation pressure to Cloud ServerPower does not need to improve existing cash register terminal yet.
And the analysis identification process of third facial image, i.e., the third face figure of consumer is obtained again through photographic devicePicture includes the following steps；
The image information that consumer is obtained again through the photographic device that cashier is arranged in, by preset interpupillary distance to eachThe interpupillary distance of image information identification is screened, and obtains third facial image according to the selection result, and third facial image is shownIn cash register terminal.Screening process is mainly to filter out consumer's face and photographic device apart from the third face figure properly obtainedPicture.Two eye pupils are calculated by judging the pixel of pupil in image information away from thus convert consumer's face and cameraDistance.The third facial image so screened is consumer in the people being taken apart from photographic device pre-determined distance threshold valueFace image.In cash register terminal, third facial image can be shown between two parties, specifically by third facial image central pointCash register terminal central point is marked to judge at a distance from position in entire Pictures location and cashier.
Further, the identification that face identity is carried out to the first facial image, obtains the second facial image and corresponding meetingMember ID, includes the following steps；
The mapping relations of pre-set each face characteristic image and member ID in Cloud Server, by the first facial imageIt is matched with each face characteristic image；It chooses and the highest face characteristic image of the degree of approximation in matching result is determined as the second peopleFace image, and by membership information corresponding with the second facial image present in mapping relations, it is determined as and the second facial imageThe member ID to match.First facial image is being matched with each face characteristic image, is obtaining corresponding matching degree, foundationMatching degree is ranked up each face characteristic image, chooses the highest face characteristic image of the degree of approximation and is determined as the second face figurePicture.To simplify the data volume of subsequent calculating.
Based on the same inventive concept, the embodiment of the present invention also provides a kind of cash register system of member system based on recognition of faceSystem, the process that the implementation of the system can refer to the above method is realized, it is no longer redundant later to repeat place.
As shown in Fig. 2, being the structure of the cash register system of the member system provided by Embodiment 2 of the present invention based on recognition of faceSchematic diagram, including identification module 100, judgment module 200, acquisition module 300, binding module 400 and cash register module 500；KnowOther module 100 is used to obtain the first facial image of consumer by the photographic device that cashier is arranged in, by the first face figureAs being uploaded to Cloud Server, and the identification of face identity is carried out to the first facial image, obtain the second facial image and correspondingMember ID；Judgment module 200 is used to obtain the third facial image of consumer again through photographic device, by the second face figurePicture, third facial image and member ID are transmitted to the cash register terminal of cash register, in cash register terminal, according to the second facial image and thirdFacial image, judges whether current consumer bound member；Member was bound for current consumer if obtaining module 300,Membership information is then obtained from Cloud Server according to member ID；If binding module 400 is used for, current consumer is unbound to cross member,The identity information and third facial image of current consumer are obtained, and member's note is carried out according to identity information and third facial imageVolume binding, obtains membership information；Cash register module 500 is used to complete cash register according to the consumption information of membership information and consumer.
The cash register system of member system provided by the invention based on recognition of face can be improved the confirmation effect of membership informationRate improves cash register efficiency；And judge whether consumer bound member using the third facial image of reacquisition, so thatIt is more accurate to identify.
Further, identification module 100 is at least one of the first recognition unit and the second recognition unit；First identificationUnit is used to obtain the video information of consumer by the photographic device that cashier is arranged in, and passes through the face box pair of settingVideo information carries out analysis identification, obtains the first facial image；Second recognition unit is used for the camera shooting by the way that cashier is arranged inDevice obtains the image information of consumer, and identifies to image information, obtains the first facial image.
In order to improve operational efficiency, judgment module 200 includes screening unit 210；Screening unit 210, for again throughThe photographic device that cashier is arranged in obtains the image information of consumer, the pupil identified by preset interpupillary distance to each image informationAway from being screened, third facial image is obtained according to the selection result, and third facial image is shown in cash register terminal.
In order to improve recognition accuracy, identification module 100 is also used to；Pre-set each face characteristic in Cloud ServerThe mapping relations of image and member ID match the first facial image with each face characteristic image；It chooses matching resultThe middle highest face characteristic image of the degree of approximation is determined as the second facial image, and by present in mapping relations with the second face figureAs corresponding membership information, it is determined as the member ID to match with the second facial image.
Further, membership information is one in name, cell-phone number, gender, birthday, face characteristic image and member IDKind is a variety of.
Although the invention has been described by way of example and in terms of the preferred embodiments, but it is not for limiting the present invention, any this fieldTechnical staff without departing from the spirit and scope of the present invention, may be by the methods and technical content of the disclosure above to this hairBright technical solution makes possible variation and modification, therefore, anything that does not depart from the technical scheme of the invention, and according to the present inventionTechnical spirit any simple modifications, equivalents, and modifications to the above embodiments, belong to technical solution of the present inventionProtection scope.