CN110442658A - A kind of data correlation method and device - Google Patents
A kind of data correlation method and device Download PDFInfo
- Publication number
- CN110442658A CN110442658A CN201910749127.6A CN201910749127A CN110442658A CN 110442658 A CN110442658 A CN 110442658A CN 201910749127 A CN201910749127 A CN 201910749127A CN 110442658 A CN110442658 A CN 110442658A
- Authority
- CN
- China
- Prior art keywords
- data
- wifi
- wifi data
- score value
- face
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000001815 facial effect Effects 0.000 claims abstract description 44
- 230000000875 corresponding effect Effects 0.000 claims abstract description 20
- 230000002596 correlated effect Effects 0.000 claims abstract description 17
- 238000005070 sampling Methods 0.000 claims description 26
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Image Analysis (AREA)
Abstract
This application provides a kind of data correlation method and devices, wherein method includes: to obtain wifi data and human face data, and wifi data include: the MAC Address for connecting the equipment of wifi network, and human face data includes: facial image.Using the wifi data of MAC Address identical in wifi data as one group, obtain wifi data group, calculate MAC Address that each wifi data group the includes matching score value between face each in facial image respectively, match the size of score value, indicate respectively that the degree of the same person is positively correlated with face belonging to equipment belonging to MAC Address and facial image, for each MAC Address, by the MAC Address respectively in the matching score value between each face, according to the corresponding face of matching score value of the preset quantity of matching score value descending order, as with the associated face of the MAC Address.The application has certain accuracy to the association results that MAC Address and face are associated.
Description
Technical field
This application involves field of information processing more particularly to a kind of data correlation method and devices.
Background technique
Currently, passing through the available some valuable information of big data analysis, wherein it is big that a variety of data, which are associated,
Therefore one of data analysis by being associated available valuable information to a variety of data, and then can be use
The decision at family etc. provides foundation.
Summary of the invention
This application provides a kind of data correlation method and devices, it is therefore intended that determines between wifi data and human face data
Incidence relation.
To achieve the goals above, this application provides following technical schemes:
This application provides a kind of data correlation methods, comprising:
Obtain wifi data and human face data;The wifi data include: the MAC Address for connecting the equipment of wifi network;
The human face data includes: facial image;
Using the wifi data of identical MAC Address in the wifi data as one group, wifi data group is obtained;
Calculate the MAC Address matching between each face in the facial image point respectively that each wifi data group includes
Value;The size of the matching score value, indicates respectively with face belonging to equipment belonging to the MAC Address and the facial image
The degree of the same person is positively correlated;
For each MAC Address, by the MAC Address respectively in the matching score value between each face, according to matching score value from
The corresponding face of matching score value for arriving the preset quantity of small sequence greatly, as with the associated face of the MAC Address.
Optionally, any bar wifi data in the WiFi data group further include: the generation moment of this wifi data;Appoint
One human face data of meaning further include: the shooting time of the facial image.
Optionally, the MAC Address and any one in the human face data that any one wifi data group includes are determined
Matching score value between face, comprising:
From the human face data of the face, determines to meet between every wifi data in the wifi data group respectively and preset
The human face data of condition, as the human face data group with the wifi data correlation;It is described pre- for any one wifi data
If condition includes: to belong to predetermined time range constantly;The predetermined time range is to be with the generation moment of this wifi data
Range at the time of the front and back preset duration of center time point is constituted;
The matching score value between every wifi data and associated human face data group is calculated separately, every wifi data are obtained
Match score value;
According to the matching score value of every wifi data, determine between the MAC Address and the face that the wifi data group includes
Match score value.
Optionally, any one wifi data further include: the position of the equipment;Any one human face data
Further include: position of the facial image when being taken.
Optionally, the matching score value between any one wifi data and associated human face data group is calculated, comprising:
Calculate matching score value of this wifi data respectively with every human face data in associated human face data group;Its
In, the matching score value and gap of any one human face data in this wifi data and associated human face data group are at negative
It closes;Difference of the gap between first distance and second distance;The first distance be this wifi data in position with
Distance between router belonging to this wifi data;The second distance is position and this wifi in this human face data
Distance between router belonging to data;
This wifi data are added with the matching score value of every human face data in associated human face data group respectively
The value arrived, as the matching score value between this wifi data and associated human face data group.
Optionally, it in the matching score value calculated separately between every wifi data and associated human face data group, obtains
After the matching score value of every wifi data, and in the matching score value according to every wifi data, the wifi data are determined
Before matching score value between MAC Address and the face that group includes, further includes:
From the wifi data group, determine that the wifi data of overlapping time section in the wifi data for belonging to different routers are
One group of wifi data to be processed;
It will be ranked up at the time of the wifi data to be processed according to default sequencing, it is to be processed after being sorted
Wifi data;
The weight of the matching score value of every wifi data in wifi data to be processed after determining the sequence;It is any one
The weight of the matching score value of wifi data is negatively correlated with target range;The target range is this wifi data and adjacent
A wifi data respectively belonging to router between distance;
The matching score value of every group of wifi data to be processed is weighted respectively and, any one group of wifi data to be processed
The numerical value that matching score value is weighted and obtains is the matching score value of group wifi data to be processed;
By in the wifi data group, the matching score value and each group of the wifi data in addition to each group wifi data to be processed
The corresponding matching score value of wifi data to be processed is added;
The matching score value according to every wifi data determines MAC Address and the face that the wifi data group includes
Between matching score value, comprising:
Matching score value between MAC Address and the face that the value that will add up includes as the wifi data group.
Optionally, the preset condition, further includes: position belongs to predeterminated position range;The predeterminated position range are as follows: with
Centered on router belonging to this wifi data, border circular areas is formed by by radius of pre-determined distance;The pre-determined distance
The straight length covered for the signal of the router.
Optionally, after the acquisition wifi data and human face data, and it is described will be identical in the wifi data
MAC Address wifi data as one group, before obtaining wifi data group, further includes:
It is unified to arrive under preset reference frame by the position in the wifi data and the human face data, it is united
Wifi data after one and human face data after reunification;
Removal and the incoherent data of data correlation and second-rate data, obtain from the wifi data after reunification
Wifi data after to removal;It is described to be not belonging to preset duration range with the incoherent data of data correlation for duration
Wifi data;The second-rate data are the wifi data that signal strength is less than preset strength threshold value;
By when a length of first preset duration between the initial time and the initial time of the wifi data after the removal
Integral multiple at the time of and the removal after wifi data finish time, as wifi sampling instant;
Wifi data in wifi data after the removal in addition to the wifi data of the wifi sampling instant are carried out
It deletes, obtains pretreated wifi data;
Human face data in the human face data after reunification in addition to the human face data of predesignated personnel is deleted
It removes, the human face data after being removed;
By the initial time of the human face data after the removal, with the initial time when it is a length of it is described second it is default when
The finish time of human face data at the time of long integral multiple and after the removal is face sampling instant;
Human face data in human face data after the removal in addition to the human face data of the face sampling instant is carried out
It deletes, obtains pretreated human face data.
Optionally, first preset duration is identical as second preset duration and is equal to default delay duration;It is described
Delay between at the time of default delay duration is the generation moment of the wifi data and is connected to router.
Optionally, described for each MAC Address, which is pressed in the matching score value between each face respectively
According to matching score value descending order preset quantity the corresponding face of matching score value, as with the associated people of the MAC Address
After face, further includes:
There are identical in the face being respectively associated with each MAC Address respectively obtained in different preset time periods
In the case where MAC Address, determine with the face that repeats in the associated face of identical MAC Address, different preset as described
In period with the identical associated face of MAC Address.
Present invention also provides a kind of data association devices, comprising:
Module is obtained, for obtaining wifi data and human face data;The wifi data include: to connect setting for wifi network
Standby MAC Address;The human face data includes: facial image;
Grouping module, for obtaining wifi using the wifi data of identical MAC Address in the wifi data as one group
Data group;
Computing module, for calculate MAC Address that each wifi data group includes respectively in the facial image each one
Matching score value between face;The size of the matching score value, and belonging to equipment belonging to the MAC Address and the facial image
Face indicates respectively that the degree of the same person is positively correlated;
Sorting module presses the MAC Address in the matching score value between each face respectively for for each MAC Address
According to the corresponding face of matching score value of the preset quantity of matching score value descending order, it is associated with as with the MAC Address
Face.
Optionally, the grouping module is grouped any bar wifi data in resulting wifi data group further include: this
The generation moment of wifi data;Any one human face data for obtaining module and obtaining further include: the facial image
Shooting time.
Optionally, the computing module, for determining MAC Address that any one wifi data group includes and the face
The matching score value between any one face in image, comprising:
The computing module, specifically for determining respectively and in the wifi data group from the human face data of the face
The human face data for meeting preset condition between every wifi data, as the human face data group with the wifi data correlation;For
Any one wifi data, the preset condition include: to belong to predetermined time range constantly;The predetermined time range is with this
Wifi data generate range at the time of the front and back preset duration that the moment is center time point is constituted;
The matching score value between every wifi data and associated human face data group is calculated separately, every wifi data are obtained
Match score value;
According to the matching score value of every wifi data, determine between the MAC Address and the face that the wifi data group includes
Match score value.
Optionally, the grouping module is grouped any bar wifi data in resulting wifi data group further include: described to set
Standby position;Any one human face data for obtaining module and obtaining further include: the facial image is when being taken
Position.
Optionally, for calculating the matching score value between any one wifi data and associated human face data group, comprising:
The computing module, specifically for calculate this wifi data respectively with every people in associated human face data group
The matching score value of face data;Wherein, of this wifi data and any one human face data in associated human face data group
It is negatively correlated with score value and gap;Difference of the gap between first distance and second distance;The first distance is this
At a distance between router belonging to position in wifi data and this wifi data;The second distance is this human face data
In position and this wifi data belonging between router at a distance from;
This wifi data are added with the matching score value of every human face data in associated human face data group respectively
The value arrived, as the matching score value between this wifi data and associated human face data group.
Optionally, further includes:
Weighting block, for being calculated separately between every wifi data and associated human face data group in the computing module
Score value is matched, after obtaining the matching score value of every wifi data, and in the matching score value according to every wifi data, really
Before matching score value between MAC Address and the face that the fixed wifi data group includes, from the wifi data group, determination belongs to
The wifi data of overlapping time section are one group of wifi data to be processed in the wifi data of different routers;
It will be ranked up at the time of the wifi data to be processed according to default sequencing, it is to be processed after being sorted
Wifi data;
The weight of the matching score value of every wifi data in wifi data to be processed after determining the sequence;It is any one
The weight of the matching score value of wifi data is negatively correlated with target range;The target range is this wifi data and adjacent
A wifi data respectively belonging to router between distance;
The matching score value of every group of wifi data to be processed is weighted respectively and, any one group of wifi data to be processed
The numerical value that matching score value is weighted and obtains is the matching score value of group wifi data to be processed;
By in the wifi data group, the matching score value and each group of the wifi data in addition to each group wifi data to be processed
The corresponding matching score value of wifi data to be processed is added;
The computing module determines the MAC that the wifi data group includes for the matching score value according to every wifi data
Matching score value between address and the face, comprising:
The computing module, the MAC Address for including as the wifi data group specifically for the value that will add up with should
Matching score value between face.
Optionally, the preset condition, further includes: position belongs to predeterminated position range;The predeterminated position range are as follows: with
Centered on router belonging to this wifi data, border circular areas is formed by by radius of pre-determined distance;The pre-determined distance
The straight length covered for the signal of the router.
Optionally, further includes:
Preprocessing module is used for after the acquisition module obtains wifi data and human face data, and in the grouping
Module is using the wifi data of identical MAC Address in the wifi data as one group, will be described before obtaining wifi data group
Position in wifi data and the human face data, it is unified to arrive under preset reference frame, obtain wifi data after reunification
Human face data after reunification;
Removal and the incoherent data of data correlation and second-rate data, obtain from the wifi data after reunification
Wifi data after to removal;It is described to be not belonging to preset duration range with the incoherent data of data correlation for duration
Wifi data;The second-rate data are the wifi data that signal strength is less than preset strength threshold value;
By when a length of first preset duration between the initial time and the initial time of the wifi data after the removal
Integral multiple at the time of and the removal after wifi data finish time, as wifi sampling instant;
Wifi data in wifi data after the removal in addition to the wifi data of the wifi sampling instant are carried out
It deletes, obtains pretreated wifi data;
Human face data in the human face data after reunification in addition to the human face data of predesignated personnel is deleted
It removes, the human face data after being removed;
By the initial time of the human face data after the removal, with the initial time when it is a length of it is described second it is default when
The finish time of human face data at the time of long integral multiple and after the removal is face sampling instant;
Human face data in human face data after the removal in addition to the human face data of the face sampling instant is carried out
It deletes, obtains pretreated human face data.
Optionally, first preset duration is identical as second preset duration and is equal to default delay duration;It is described
Delay between at the time of default delay duration is the generation moment of the wifi data and is connected to router.
Optionally, further includes:
Processing module is used in the sorting module for each MAC Address, by the MAC Address respectively between each face
Matching score value in, according to matching score value descending order preset quantity the corresponding face of matching score value, as with this
After the associated face of MAC Address, the face being respectively associated with each MAC Address that is respectively obtained in different preset time periods
In there are in the case where identical MAC Address, determine with the face that repeats in the associated face of identical MAC Address, as
It is described difference preset time period in the identical associated face of MAC Address.
In data correlation method described herein and device, wifi data and human face data are obtained, wherein wifi data
It include: the MAC Address for connecting the equipment of wifi network, human face data includes: facial image, by MAC identical in wifi data
The wifi data of address obtain wifi data group as one group, calculate MAC Address that each wifi data group includes respectively with people
Matching score value in face data between each face obtains matching score value of each MAC Address respectively between each face, for each
MAC Address, by the MAC Address respectively in the matching score value between each face, according to the default of matching score value descending order
The corresponding face of matching score value of quantity, as with the associated face of the MAC Address.
Due to matching the size of score value, indicated respectively together with face belonging to equipment belonging to MAC Address and facial image
The degree of one people is at positive association, and therefore, what this programme was determined has certain accuracy with the associated face of MAC Address,
Therefore, this application provides a kind of data correlation schemes, also, to the association results that MAC Address and face are associated
With certain accuracy.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of application for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of application scenarios schematic diagram of data correlation method disclosed in the embodiment of the present application;
Fig. 2 is a kind of flow chart of data correlation method disclosed in the embodiment of the present application;
Fig. 3 is a kind of structural schematic diagram of data association device disclosed in the embodiment of the present application.
Specific embodiment
Fig. 1 is a kind of application scenarios schematic diagram of data correlation method provided by the embodiments of the present application, comprising: the first equipment
101, the second equipment 102 and data association device provided by the embodiments of the present application 103, wherein the first equipment 101 is for providing
Wifi data, the second equipment 102 are used to determine what the first equipment 101 provided for providing human face data, data association device 103
The human face data that wifi data and the second equipment 102 provide is associated, and matches the wifi data for belonging to the same person and people
Face data.
In the embodiment of the present application, the human face data that the wifi data and the second equipment that the first equipment provides provide refers to: referring to
Determine the wifi data and human face data that the same pre-set space generates in place.For example, specified place is the market XX, place is specified
The wifi data and human face data that the interior same pre-set space can generate for same layer in the market XX.
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
Fig. 2 is a kind of data correlation method provided by the embodiments of the present application, comprising the following steps: the present embodiment is to finger
Determine to be introduced for wifi data and human face data caused by field.
S201, wifi data and human face data are obtained.
In the embodiment of the present application, wifi data are that equipment scanning is routed to the router in specified place under each moment
The data that device generates, wherein wifi data specifically include: it is connected to the MAC Address of the equipment of wifi network, router generates
At the time of wifi data, signal strength and position.Wherein, equipment can be mobile phone, and the present embodiment is not to the specific shape of equipment
Formula is limited, and position indicates position of the equipment relative to the router scanned.For example, the equipment that MAC Address is A is scanned to quotient
Router B in two layers, also, signal strength scan according to the equipment can determine equipment that MAC Address is A and
Distance between router B, since position of the router B in market is known, hence, it can be determined that MAC Address is A's out
Position of the equipment in market is to be formed by area near round edge by radius of calculated distance by the center of circle of router B
Domain.Specified place can be set according to actual business scenario, for example, being set as specified market.
Specifically, wifi data can be obtained by wifi probe.In the present embodiment, by a MAC Address at one
When the wifi data inscribed claim a wifi data.
Human face data is the position that each personage in specified place inscribes when each, specifically, human face data includes: face
Image, facial image shooting time and facial image in position of face when being photographed in specified place.
Specifically, the acquisition modes of human face data may include: to be clapped by each camera installed out of specified place
In the video flowing taken the photograph, the shooting time of facial image and facial image is identified, and carry out three according to the image in video flowing
Dimension is rebuild, and determines that the personage that each facial image identified includes is specifying the position at each moment in place, and will identify that
Facial image, facial image shooting time and facial image personage's shooting time for including in specified place
Set of locations is at human face data.
For convenience, in the present embodiment, the face number that a position face inscribed at one is constituted
It is stated to be a human face data.
The position in position and human face data in S202, the wifi data that will acquire, it is unified to arrive preset reference coordinate
Under system, wifi data after reunification and human face data after reunification are obtained.
Specifically, the realization process of this step is the prior art, which is not described herein again.
S203, wifi data after reunification and human face data after reunification are pre-processed respectively, after obtaining pretreatment
Wifi data and pretreated human face data.
It include two aspects to the pretreatment that wifi data after reunification and human face data after reunification carry out respectively, respectively
For first aspect and second aspect.
Wherein, first aspect are as follows: removal and the incoherent data of data correlation and quality from wifi data after reunification
Poor data, the wifi data after being removed.Removal and the incoherent number of data correlation from human face data after reunification
According to human face data after being removed.Second aspect are as follows: to the wifi data after the human face data and removal after removal respectively into
Row data sampling obtains pretreated wifi data and pretreated human face data.
For wifi data: the processing to the first aspect of wifi data progress after reunification includes: from after reunification
Removal includes: to remove duration from wifi data after reunification not belong to the incoherent data of data correlation in wifi data
In the wifi data of preset duration range.It includes: from after reunification that second-rate data are removed from wifi data after reunification
Wifi data in removal signal strength be less than preset strength threshold value wifi data.
Wherein, preset duration range is made of maximum time limit value and duration lower limit value, wherein maximum time limit value is for area
Divide the duration of designated person and non-designated personnel.Also using specified place as market, customer be personnel relevant to data correlation (i.e.
Designated person) for, since the duration of customer and customer non-(staff) in market is different, hence, it can be determined that
A duration value for distinguishing staff and customer is as maximum time limit value out, for example, maximum time limit value can be small for 8
When.Specifically, the specific value of maximum time limit value needs to be set according to the actual situation, the present embodiment is not to duration upper limit value
Specific value limit.
Wherein, duration lower limit value be for distinguish personnel that equipment belonging to MAC Address in wifi data indicates whether be
Personnel in specified place.Also by taking specified place is market and customer is designated person as an example, due to pass by market and not into
The equipment for entering the personnel in market may also scan the router into market, still, pass by market and do not enter the personnel in market
It is not the relevant personnel of data correlation (non-designated personnel).Also, it relative to the personnel in market, passes by market and does not enter quotient
The duration that the WiFi data of the personnel of field is lasting is shorter, therefore, can be set one for distinguishing MAC Address institute in wifi data
Whether the personnel of the equipment instruction of category are the duration value of personnel in market as duration lower limit value.For example, duration lower limit value is 2
Minute.
Wherein, the wifi data that signal strength is less than preset strength threshold value indicate the poor wifi data of signal strength, tool
Body, the value of preset strength threshold value can be -90dB, and certainly, the specific value of preset strength threshold value is needed according to practical feelings
Condition is set, and the present embodiment does not limit the specific value of preset strength threshold value.
For human face data: in this step, the processing that first aspect is carried out to human face data after reunification includes: from people
The human face data outside the human face data for the personnel of preassigning is removed in face data, wherein designated person is actual business scenario
The required personnel for carrying out data correlation.Also by taking specified place is market as an example, customer is designated person.
Specifically, from the process packet of the human face data outside the human face data for removing designated person in human face data after reunification
It includes: the human face data outside the human face data for identifying designated person in human face data after reunification, and the face that will identify that
Data are deleted.Wherein, from the human face data outside the human face data for identifying designated person in human face data after reunification
Process may include: the face information of prior statistics designated person, and the people counted is identified from human face data after reunification
Face information, and then obtain the human face data in human face data after reunification in addition to the human face data of designated person.Certainly, this reality
It applies example and merely provides and a kind of identify the human face data in addition to the human face data of designated person from human face data after reunification
Mode in practice can also by other means, the present embodiment does not limit specific identification method.
After the human face data after the wifi data and removal after being removed, after the wifi data and removal after removal
Human face data carry out the processing of second aspect respectively, i.e., carry out data sampling respectively, obtain pretreated human face data and
Pretreated wifi data.
In the present embodiment, in order to reduce computing resource and improve computational efficiency, the is carried out to the human face data after removal
The processing of two aspects, specially samples the human face data after removal.Specific sampling process includes: by the people after removal
The initial time of face data, with initial time when a length of first preset duration integral multiple at the time of and finish time make
For face sampling instant.Human face data in human face data after removal in addition to the human face data of face sampling instant is deleted,
Obtain pretreated human face data.Wherein, the specific value of the first preset duration can be set according to practical business scene
It sets, the present embodiment does not limit the value of the first preset duration.
The processing for carrying out second aspect to the wifi data after removal is to carry out data sampling to the wifi data after removal.
Specifically, to after removal wifi data carry out data sampling process include: by the initial time of the wifi data after removal,
Between initial time when a length of second preset duration integral multiple at the time of and finish time as wifi sampling instant,
Wifi data in wifi data after removal in addition to the wifi data of wifi sampling instant are deleted, are obtained pretreated
Wifi data.Wherein, the specific value of the second preset duration needs to be set according to specific business scenario, and the present embodiment is not to the
The specific value of two preset durations limits.
In practice, the generation moment of wifi data and scanning to the moment (scanning generate wifi data router when
Carve) between there are certain delays, in the present embodiment, based on delay setting one delay duration, for example, delay duration be 5s,
Certainly, in practice, default delay duration can also be other values, and the present embodiment is not to the specific value of default delay duration
It limits.In the present embodiment, when the first preset duration is identical as the second preset duration, also, it is equal to default delay duration
In the case of, subsequent step is carried out based on pretreated wifi data and pretreated human face data, it is finally obtained
The accuracy of the association results of wifi data and human face data can be improved.
It should be noted that this step is not necessarily meant to the step of executing in practice.
S204, using the wifi data of MAC Address same in wifi data as one group, obtain wifi data group.
This step can be executed to pretreated wifi data, the wifi data being also possible to after reunification execute this step
Suddenly.
S205, the MAC Address matching between face each in human face data point respectively that each wifi data group includes is calculated
Value, obtains matching score value of each MAC Address respectively between each face.
In this step, the MAC Address that determining and every group of wifi data include is respectively between face each in human face data
The process of matching score value be it is identical, for convenience, by taking any one wifi data group and any one face as an example, be situated between
It continues and determines the wifi data group MAC Address that includes and the process for matching score value between the face.Specifically include step A1~step
Rapid A5:
The human face data group that every wifi data in A1, determination and the wifi data group match.
In this step, the process of determining human face data group associated with every wifi data is identical, in order to retouch
It states conveniently, with any one wifi data instance, introduces determining and this wifi data correlation from the human face data of the face
Human face data group.Specifically, determining that the human face data for meeting preset condition is this wifi from the human face data of the face
Associated human face data group, wherein preset condition includes: to belong to predetermined time range at the generation moment.Wherein, predetermined time range
Including upper limit value and lower limit value, wherein upper limit value is when being preset before using the generation moment of this wifi data as center time point
At the time of long, lower limit value using the generation moment of this wifi data as center time point after preset duration at the time of.
For any one wifi data, determine to meet preset condition in the human face data of the face by this step
Human face data group is the human face data group of this wifi data correlation.In the present embodiment, need to judge this wifi data packet
Whether the face that equipment belonging to the MAC Address contained and the human face data group determined include indicates respectively the same person.In reality
In border, facial image that equipment belonging to the MAC Address that this wifi data include sometimes and associated human face data group include
The face of instruction is clearly not the instruction same person, if determining that matching score value is meaningless according to the following steps.For example, quotient
The wifi data that the router that field is three buildings generates and the human face data for meeting preset condition are lineup's face of market Stall
Data.
Therefore, in this step, in order to improve the computational efficiency of the present embodiment, preset condition further include: position belongs to pre-
If position range, specifically, centered on predeterminated position range includes: the router belonging to this wifi data, with it is default away from
Border circular areas is formed by from for radius, wherein the pre-determined distance is the length that the signal of the router can cover
Degree.
If there is no the human face data groups for meeting preset condition in the human face data of the face, by this wifi data
Matching score value between the face is set as 0, and the movement without A2 according to the following steps executes, it is of course also possible to continue to execute step
The movement of rapid A2.
Matching score value between A2, every wifi data of calculating and associated human face data group, obtains every wifi data
Match score value.
In this step, to the calculating process phase for matching score value of any one wifi data and associated human face data group
Together, for convenience, by taking any one wifi data and associated human face data group as an example, this wifi data and pass are introduced
The calculating process of matching score value between the human face data group of connection, specifically includes: calculating separately this wifi data and associated people
The matching score value between each human face data in face data group, and the calculated each matching score value of institute is added, if being added institute
Obtained value is not more than preset matching score threshold, then will add up obtained value as this wifi data and associated face
Matching score value between data group divides preset matching if being added obtained value greater than in the case where preset matching point threshold
It is worth threshold value as the matching score value between this wifi data and associated human face data group.For convenience, by this wifi
Matching score value between data and associated human face data group, referred to as the matching score value of this wifi data.
Wherein, this wifi data and the matching score value between any one human face data in the human face data group of connection are calculated
Process include: firstly, according in this wifi data signal strength and signal strength and distance between preset relation, meter
The distance for calculating the router belonging to this article of wifi data of equipment belonging to the MAC Address that this article of wifi data include is the
One distance.Secondly, being the at a distance between determining router belonging to this article of human face data face that includes and this article of wifi data
Two distances.Finally, being determined between this wifi data and this human face data according to the gap between first distance and second distance
Match score value, wherein the matching between the gap between first distance and second distance and this wifi data and this human face data
Score value is negatively correlated, i.e. the gap of first distance and second distance is smaller, then between this wifi data and this human face data
It is higher to match score value, on the contrary, the matching score value between this wifi data and this human face data is lower.
Pass through the matching score value of every wifi data in the available wifi data group of this step.
A3, according to the matching score value of every wifi data, determine between the MAC Address and the face that the wifi data group includes
Matching score value.
In this step, it can be calculated by two ways, respectively first way and the second way.
Wherein, first way includes: to be added the matching score value of every wifi data in the wifi data group, acquired
Value include as the wifi data group MAC Address and the face between matching score value.
In practice, due to there are the wifi data for belonging to different routers in the wifi data group, and belong to difference
There are the wifi data of period overlapping in the wifi data of router, i.e., there are noise datas in the wifi data group, cause
It cannot reach good with the accuracy for matching score value between the face according to the wifi data group that first way is determined
Effect.Belong to the wifi data of router A for example, existing in the wifi data group, the period be 9 points 00 minute~9 points 30 minutes, together
When, there are the wifi data for belonging to router B in the wifi data group, the period be 9 points 20 minutes~9 points 40 minutes, wherein 9 points
20 points~9 points 30 minutes be overlapping time section.
In order to further increase the accuracy of the matching score value between the MAC Address and the face that the wifi data group includes,
This step provides the second way.Specifically, calculating the MAC Address and be somebody's turn to do that the wifi data group includes by the second way
The process of matching score value between face includes step B1~step B3:
B1, using in the wifi data for belonging to different routers in the wifi data group period be overlapped wifi data as
One group of wifi data to be processed.
Also by 9 points 20 minutes~9 points 30 minutes for overlapping time section for, in this step, then will be deposited in the wifi data group
The moment belongs to 9 points of 20 minutes~9 points 30 minutes wifi data and the wifi data group in the wifi data for belonging to router A
It is middle exist belong to the moment in the wifi data of router B to belong to 9 points of 20 minutes~9 points 30 minutes wifi data to be processed as one group
Wifi data.
The weight of the matching score value of each item wifi data to be processed in B2, wifi data to be processed for every group.
Specifically, with any one group of wifi data instance, by the wifi data in group wifi data group to be processed when
Quarter is ranked up according to default sequencing, the wifi data to be processed after being sorted.Wifi to be processed after determining sequence
The weight of the matching score value of every wifi data in data, wherein the weight of the matching score value of any one wifi data with
Target range is negatively correlated, wherein target range is road belonging to this wifi data and adjacent wifi data difference
By the distance between device.
Wherein, adjacent can be previous item or latter item, still, for calculating the matching of one group of wifi data to be processed
During score value, wifi data to be processed for every, an adjacent wifi data are all " previous wifi data ", or
Person, an adjacent wifi data are all " latter wifi data ", specially " previous wifi data " still " latter item
This embodiment is not limited for wifi data ", as long as same group of wifi data to be processed are unified.
Specifically, the weight can be arranged if the distance between affiliated router is greater than pre-determined distance threshold value respectively
For a numerical value greater than 1, also, the bigger weight of distance is bigger.If the distance between router belonging to respectively is less than default
The weight is then set smaller than 1 numerical value by distance threshold, also, smaller apart from smaller weight.Any group is waited locating
Wifi data are managed from first wifi data (feelings being compared with previous wifi data being ranked up in rear on the time
Condition), weight can be set as 1.Wifi data to be processed for the group are from the last item wifi being ranked up in rear on the time
Data (the case where being compared with latter wifi data), weight can be set as 1.
B3, by the wifi data group, the matching score value of the wifi data in addition to each group wifi data to be processed, and each
The value that the corresponding matching score value of group wifi data to be processed is added, and will add up is as the wifi data group packet
Matching score value between the MAC Address contained and the face.
S206, the MAC Address separately included for each wifi data group, by the MAC Address respectively between each face
Match score value in, according to matching score value descending order preset quantity the corresponding face of matching score value, as with
The associated face of the MAC Address.
It is introduced by taking any one wifi data group as an example.Specifically, the MAC Address for including by the wifi data group point
Score value is matched according to sequence from high to low not between each face, the corresponding people of matching score value of preset quantity before determining
Face, as the associated face of MAC Address for including with the wifi data group.
In this step, the value of preset quantity can determine that the present embodiment is not to default according to actual business scenario
The value of quantity limits.In the present embodiment, the value of preset quantity can be 5.
For example, the MAC Address that the wifi data group includes is A, include with the associated face array of the wifi data group
Face includes: face A, face B, face C and face D, wherein the matching score value between A and face A is 80 points, between A and face B
Matching score value be 90 points, the score value that match between A and face C be the matching score value between 90 points and A and face D is 100 points, default
The value of quantity be 3, then in this step, using face D, face C and face B as with the associated face of MAC Address A.
There are identical MAC Address in S207, the face being respectively associated with each MAC Address in different preset time periods
In the case where, determine with the face that repeats in the associated face of identical MAC Address, as in the difference preset time period and
The identical associated face of MAC Address.
In this step, different preset time periods can be different days, for example, on June 20th, 2019 and June 21 in 2019
Day is two different preset time periods.
In the present embodiment, it can be determined and wifi from the wifi data and human face data in each preset time
The face that each MAC Address that data include is respectively associated.In this step, in different preset time periods with each MAC
There are in the case where identical MAC Address in the face that location is respectively associated, MAC Address identical in different preset time periods is distinguished
The face repeated in associated face, as associated face of MAC Address identical with this.
For example, every on June 20th, 2019, on June 21st, 2019, on June 22nd, 2019 and on June 23rd, 2019
It, all obtains the associated face of MAC Address for including with wifi data group.In this step, include from wifi data group
In the associated face of MAC Address, the face repeated is determined.Using the face repeated determined as with wifi number
The associated face of MAC Address for including according to group, the association results obtained at this time accuracy with higher.
For example, it is small it is red on June 20th, 2019 together with friend A, friend B in the market X, be likely to be obtained and small red equipment
The associated face of MAC Address has small red, friend A and friend B.It is small it is red on June 21st, 2019 together with friend C, friend D in X
Market, being likely to be obtained has small red, friend C and friend D with the associated face of MAC Address of small red equipment.It is assumed that small red 2019
On June 20, in and on the June 21st, 2019 of equipment entrained by the market X are the same equipment, then available June 20 in 2019
Day and on June 21st, 2019, with the associated face of MAC Address of small red equipment in the face that repeats be it is small red, then can be with
It determines to be small red with the associated face of MAC Address of small red equipment, this result is consistent with actual conditions, so, it will be with
The face repeated in the associated face of MAC Address has higher as this association results of the associated face of the MAC Address
Accuracy.
The present embodiment has the advantages that
Beneficial effect one,
In the present embodiment, wifi data and human face data are obtained, wherein wifi data include: MAC Address and MAC
Position of the equipment that location includes in specified place, human face data includes: the people in facial image, shooting time and facial image
Position of the face when being taken in specified place, it is unified to arrive preset reference by the position in wifi data and human face data
Under coordinate system, wifi data after reunification and human face data after reunification are obtained.By identical MAC in wifi data after reunification
The wifi data of address obtain multiple wifi data groups as one group, calculate the MAC Address difference that each wifi data group includes
With the matching score value between face each in human face data, matching score value of each MAC Address respectively between each face is obtained, for every
A MAC Address, by the MAC Address respectively in the matching score value between each face, according to the pre- of matching score value descending order
If the corresponding face of matching score value of quantity, as with the associated face of the MAC Address.
The equipment as belonging to the size and MAC Address of matching score value and face indicate respectively the degree of the same person at just
Association, accordingly, it is determined that is gone out has certain accuracy with the associated face of MAC Address.
Beneficial effect two,
In the present embodiment, to wifi data after reunification and after reunification, human face data is pre-processed respectively, specifically
Including both sides processing, for wifi data, the processing of first aspect includes: removal and the incoherent wifi number of data correlation
According to second-rate wifi data, wifi data after being removed.For human face data, the first convenient processing includes:
Removal and the incoherent human face data of data correlation, the human face data after being removed.Respectively to wifi data and human face data
The processing of first aspect is carried out, so that the human face data after the wifi data and removal after removal counts for data correlation
It is improved according to quality, so that the finally obtained accuracy with the result of each associated face of MAC Address in wifi data has
There is certain guarantee.
Meanwhile the processing of the second aspect in preprocessing process includes: to the people after the wifi data and removal after removal
Face data carry out data sampling respectively, since there are a large amount of redundancies in the human face data after the wifi data and removal after removal
Data, after the processing of second aspect through this embodiment, after the wifi data and removal after removal can be greatly lowered
Redundant data in human face data obtains pretreated wifi data and pretreated human face data, so that subsequent be based on
The process of final association results is calculated in pretreated wifi data and pretreated human face data, can save compared with
More computing resources, and improve computational efficiency.
Fig. 3 be a kind of data association device provided by the embodiments of the present application, comprising: obtain module 301, grouping module 302,
Computing module 303 and sorting module 304.
Wherein, module 301 is obtained, for obtaining wifi data and human face data, wifi data include: connection wifi network
Equipment MAC Address, human face data includes: facial image.Grouping module 302, for by MAC identical in wifi data
The wifi data of location obtain wifi data group as one group.Computing module 303 includes for calculating each wifi data group
The MAC Address matching score value between face each in facial image respectively, matches the size of score value, with equipment belonging to MAC Address
Indicate respectively that the degree of the same person is positively correlated with face belonging to facial image.Sorting module 304, for for each
MAC Address, by the MAC Address respectively in the matching score value between each face, according to the default of matching score value descending order
The corresponding face of matching score value of quantity, as with the associated face of the MAC Address.
Optionally, grouping module 302 is grouped any bar wifi data in resulting wifi data group further include: this wifi
The generation moment of data obtains any one human face data that module 301 obtains further include: the shooting time of facial image.
Optionally, computing module 303, the MAC Address and facial image for including for determining any one wifi data group
In any one face between matching score value, comprising:
Computing module 303, specifically for from the human face data of the face, determine respectively with it is every in the wifi data group
The human face data for meeting preset condition between wifi data, as the human face data group with wifi data correlation, for any one
Wifi data, preset condition includes: to belong to predetermined time range constantly;Predetermined time range is with the production of this wifi data
The raw moment is range at the time of the front and back preset duration of center time point is constituted.Calculate separately every wifi data and associated face
Matching score value between data group obtains the matching score value of every wifi data.According to the matching score value of every wifi data, determine
Matching score value between MAC Address and the face that the wifi data group includes.
Optionally, the grouping module 302 is grouped any bar wifi data in resulting wifi data group further include: equipment
Position, obtain module obtain any one human face data further include: position of the facial image when being taken.
Optionally, computing module 303, for calculating the matching between any one wifi data and associated human face data group
Score value, comprising:
Computing module 303, specifically for calculate this wifi data respectively with every people in associated human face data group
The matching score value of face data;Wherein, of this wifi data and any one human face data in associated human face data group
It is negatively correlated with score value and gap, difference of the gap between first distance and second distance, first distance is this wifi data
In position and this wifi data belonging between router at a distance from, second distance be position in this human face data with should
Distance between router belonging to wifi data.By this wifi data respectively with every people in associated human face data group
The value that the matching score value of face data is added, as the matching score value between this wifi data and associated human face data group.
Optionally, the device further include: weighting block 305, for calculating separately every wifi data in computing module 303
Matching score value between associated human face data group, after obtaining the matching score value of every wifi data, and according to every
The matching score value of wifi data, before determining the matching score value between the MAC Address and the face that the wifi data group includes, from this
In wifi data group, determine that the wifi data of overlapping time section in the wifi data for belonging to different routers are one group to be processed
Wifi data.
It will be ranked up at the time of wifi data to be processed according to default sequencing, the wifi to be processed after being sorted
Data.The weight of the matching score value of every wifi data in wifi data to be processed after determining sequence, any one wifi
The weight of the matching score value of data is negatively correlated with target range, and target range is this wifi data and an adjacent wifi
Distance between router belonging to data difference.The matching score value of every group of wifi data to be processed is weighted and is appointed respectively
The numerical value that the matching score value of one group of wifi data to be processed of anticipating is weighted and obtains is the matching of group wifi data to be processed
Score value.By in the wifi data group, the matching score value and each group of the wifi data in addition to each group wifi data to be processed wait locating
The corresponding matching score value of reason wifi data is added.
Computing module 303 determines the MAC that the wifi data group includes for the matching score value according to every wifi data
Matching score value between address and the face, comprising:
Computing module 303, the MAC Address for including as the wifi data group specifically for the value that will add up and the people
Matching score value between face.
Optionally, preset condition, further includes: position belongs to predeterminated position range, predeterminated position range are as follows: with this wifi
Centered on router belonging to data, border circular areas is formed by by radius of pre-determined distance;Pre-determined distance is the router
Signal covering straight length.
Optionally, the device further include: preprocessing module 306, for obtaining the acquisition wifi data of module 301 and face
After data, and wifi is obtained using the wifi data of MAC Address identical in wifi data as one group in grouping module 302
It is unified to arrive under preset reference frame by the position in wifi data and human face data before data group, it obtains after reunification
Wifi data and human face data after reunification.Removal and the incoherent data of data correlation and matter from wifi data after reunification
Poor data are measured, the wifi data after being removed are that duration is not belonging to preset with the incoherent data of data correlation
The wifi data of duration range, second-rate data are the wifi data that signal strength is less than preset strength threshold value.
By between the initial time and initial time of the wifi data after removal when a length of first preset duration integral multiple
At the time of, and the finish time of the wifi data after removal, as wifi sampling instant.It will be removed in wifi data after removal
Wifi data outside the wifi data of wifi sampling instant are deleted, and pretreated wifi data are obtained.It will after reunification
Human face data in human face data in addition to the human face data of predesignated personnel is deleted, the face number after being removed
According to.By the initial time of the human face data after removal, with initial time when a length of second preset duration integral multiple at the time of,
And the finish time of the human face data after removal is face sampling instant.When by removing face sampling in the human face data after removal
Human face data outside the human face data at quarter is deleted, and pretreated human face data is obtained.
Optionally, the first preset duration is identical as the second preset duration and is equal to default delay duration, presets delay duration
Delay between for the generation moment of wifi data and at the time of be connected to router.
Optionally, device further include: processing module 307 is used in sorting module 304 for each MAC Address, by this
MAC Address in the matching score value between each face, divides respectively according to the matching of the preset quantity of matching score value descending order
Be worth corresponding face, as with after the associated face of the MAC Address, respectively obtained in different preset time periods with it is each
There are in the case where identical MAC Address in the face that MAC Address is respectively associated, determine and the associated face of identical MAC Address
In the face that repeats, as in different preset time periods with the associated face of identical MAC Address.
If function described in the embodiment of the present application method is realized in the form of SFU software functional unit and as independent production
Product when selling or using, can store in a storage medium readable by a compute device.Based on this understanding, the application is real
The part for applying a part that contributes to existing technology or the technical solution can be embodied in the form of software products,
The software product is stored in a storage medium, including some instructions are used so that a calculating equipment (can be personal meter
Calculation machine, server, mobile computing device or network equipment etc.) execute each embodiment the method for the application whole or portion
Step by step.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), with
Machine accesses various Jie that can store program code such as memory (RAM, Random Access Memory), magnetic or disk
Matter.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (14)
1. a kind of data correlation method characterized by comprising
Obtain wifi data and human face data;The wifi data include: the MAC Address for connecting the equipment of wifi network;It is described
Human face data includes: facial image;
Using the wifi data of identical MAC Address in the wifi data as one group, wifi data group is obtained;
Calculate MAC Address that each wifi data group the includes matching score value between each face in the facial image respectively;Institute
The size for stating matching score value, indicates respectively same with face belonging to equipment belonging to the MAC Address and the facial image
Personal degree is positively correlated;
For each MAC Address, by the MAC Address respectively in the matching score value between each face, according to matching score value from greatly to
The corresponding face of matching score value of the preset quantity of small sequence, as with the associated face of the MAC Address.
2. the method according to claim 1, wherein any bar wifi data are also wrapped in the WiFi data group
It includes: the generation moment of this wifi data;Any one human face data further include: the shooting time of the facial image.
3. according to the method described in claim 2, it is characterized in that, determining the MAC Address that any one wifi data group includes
With the matching score value between any one face in the facial image, comprising:
From the human face data of the face, determination meets preset condition between every wifi data in the wifi data group respectively
Human face data, as the human face data group with the wifi data correlation;For any one wifi data, the default item
Part includes: to belong to predetermined time range constantly;The predetermined time range is using the generation moment of this wifi data as the time
Range at the time of the front and back preset duration at midpoint is constituted;
The matching score value between every wifi data and associated human face data group is calculated separately, the matching of every wifi data is obtained
Score value;
According to the matching score value of every wifi data, the matching between the MAC Address and the face that the wifi data group includes is determined
Score value.
4. according to the method described in claim 3, it is characterized in that, any one wifi data further include: the equipment
Position;Any one human face data further include: position of the facial image when being taken.
5. according to the method described in claim 4, it is characterized in that, calculating any one wifi data and associated human face data
Matching score value between group, comprising:
Calculate matching score value of this wifi data respectively with every human face data in associated human face data group;Wherein, should
The matching score value of any one human face data in wifi data and associated human face data group is negatively correlated with gap;It is described
Difference of the gap between first distance and second distance;The first distance is position and this wifi in this wifi data
Distance between router belonging to data;The second distance is belonging to position and this wifi data in this human face data
Router between distance;
This wifi data are added with the matching score value of every human face data in associated human face data group respectively
Value, as the matching score value between this wifi data and associated human face data group.
6. according to the method described in claim 3, it is characterized in that, calculating separately every wifi data and associated people described
Matching score value between face data group, after obtaining the matching score value of every wifi data, and described according to every wifi data
Matching score value, before determining the matching score value between the MAC Address and the face that the wifi data group includes, further includes:
From the wifi data group, determine that the wifi data of overlapping time section in the wifi data for belonging to different routers are one group
Wifi data to be processed;
It will be ranked up at the time of the wifi data to be processed according to default sequencing, the wifi to be processed after being sorted
Data;
The weight of the matching score value of every wifi data in wifi data to be processed after determining the sequence;Any one
The weight of the matching score value of wifi data is negatively correlated with target range;The target range is this wifi data and adjacent
Distance between router belonging to one wifi data difference;
The matching score value of every group of wifi data to be processed is weighted and the matching of any one group of wifi data to be processed respectively
The numerical value that score value is weighted and obtains is the matching score value of group wifi data to be processed;
By in the wifi data group, the matching score value and each group of the wifi data in addition to each group wifi data to be processed wait locating
The corresponding matching score value of reason wifi data is added;
The matching score value according to every wifi data determines between the MAC Address and the face that the wifi data group includes
Match score value, comprising:
Matching score value between MAC Address and the face that the value that will add up includes as the wifi data group.
7. according to the method described in claim 3, it is characterized in that, the preset condition, further includes: position belongs to predeterminated position
Range;The predeterminated position range are as follows: centered on the router belonging to this wifi data, using pre-determined distance as radius institute shape
At border circular areas;The pre-determined distance is the straight length that the signal of the router covers.
8. the method according to claim 1, wherein after the acquisition wifi data and human face data, and
Described using the wifi data of identical MAC Address in the wifi data as one group, before obtaining wifi data group, also wrap
It includes:
It is unified to arrive under preset reference frame by the position in the wifi data and the human face data, it obtains after reunification
Wifi data and human face data after reunification;
Removal and the incoherent data of data correlation and second-rate data, are gone from the wifi data after reunification
Wifi data after removing;The wifi number for being not belonging to preset duration range for duration with the incoherent data of data correlation
According to;The second-rate data are the wifi data that signal strength is less than preset strength threshold value;
By between the initial time and the initial time of the wifi data after the removal when a length of first preset duration it is whole
The finish time of wifi data at the time of several times and after the removal, as wifi sampling instant;
Wifi data in wifi data after the removal in addition to the wifi data of the wifi sampling instant are deleted,
Obtain pretreated wifi data;
Human face data in the human face data after reunification in addition to the human face data of predesignated personnel is deleted, is obtained
Human face data after to removal;
By the initial time of the human face data after the removal, with when a length of second preset duration of the initial time
The finish time of human face data at the time of integral multiple and after the removal is face sampling instant;
Human face data in human face data after the removal in addition to the human face data of the face sampling instant is deleted,
Obtain pretreated human face data.
9. according to the method described in claim 8, it is characterized in that, first preset duration and the second preset duration phase
With and equal to default delay duration;The default delay duration is the generation moment of the wifi data and is connected to router
Delay between moment.
10. the method according to claim 1, wherein described for each MAC Address, by the MAC Address point
In matching score value not between each face, according to the corresponding people of matching score value of the preset quantity of matching score value descending order
Face, as with after the associated face of the MAC Address, further includes:
In the face being respectively associated with each MAC Address respectively obtained in different preset time periods there are identical MAC
In the case where location, determine with the face that repeats in the associated face of identical MAC Address, as the difference preset time periods
It is interior with the identical associated face of MAC Address.
11. a kind of data association device characterized by comprising
Module is obtained, for obtaining wifi data and human face data;The wifi data include: the equipment for connecting wifi network
MAC Address;The human face data includes: facial image;
Grouping module, for obtaining wifi data using the wifi data of identical MAC Address in the wifi data as one group
Group;
Computing module, for calculating MAC Address that each wifi data group includes respectively between each face in the facial image
Matching score value;The size of the matching score value, with face belonging to equipment belonging to the MAC Address and the facial image
Indicate respectively that the degree of the same person is positively correlated;
Sorting module, for for each MAC Address, by the MAC Address respectively in the matching score value between each face, according to
The corresponding face of matching score value of preset quantity with score value descending order, as with the associated people of the MAC Address
Face.
12. device according to claim 11, which is characterized in that the grouping module is grouped in resulting wifi data group
Any bar wifi data further include: the generation moment of this wifi data;Any one people for obtaining module and obtaining
Face data further include: the shooting time of the facial image;The computing module, for determining any one wifi data group packet
Matching score value between any one face in MAC Address and the facial image contained, comprising:
The computing module, specifically for from the human face data of the face, determine respectively with every in the wifi data group
The human face data for meeting preset condition between wifi data, as the human face data group with the wifi data correlation;For any
One wifi data, the preset condition include: to belong to predetermined time range constantly;The predetermined time range is with this
Wifi data generate range at the time of the front and back preset duration that the moment is center time point is constituted;
The matching score value between every wifi data and associated human face data group is calculated separately, the matching of every wifi data is obtained
Score value;
According to the matching score value of every wifi data, the matching between the MAC Address and the face that the wifi data group includes is determined
Score value.
13. device according to claim 12, which is characterized in that the grouping module is grouped in resulting wifi data group
Any bar wifi data further include: the position of the equipment;Any one human face data for obtaining module acquisition is also
It include: position of the facial image when being taken;
The computing module, for calculating the matching score value between any one wifi data and associated human face data group, comprising:
The computing module, specifically for calculate this wifi data respectively with every face number in associated human face data group
According to matching score value;Wherein, the matching of this wifi data and any one human face data in associated human face data group point
Value is negatively correlated with gap;Difference of the gap between first distance and second distance;The first distance is this wifi
At a distance between router belonging to position in data and this wifi data;The second distance is in this human face data
At a distance between router belonging to position and this wifi data;
This wifi data are added with the matching score value of every human face data in associated human face data group respectively
Value, as the matching score value between this wifi data and associated human face data group.
14. device according to claim 13, which is characterized in that further include:
Weighting block, for calculating separately the matching between every wifi data and associated human face data group in the computing module
Score value, after obtaining the matching score value of every wifi data, and in the matching score value according to every wifi data, determining should
Before matching score value between MAC Address and the face that wifi data group includes, from the wifi data group, determination belongs to difference
The wifi data of overlapping time section are one group of wifi data to be processed in the wifi data of router;
It will be ranked up at the time of the wifi data to be processed according to default sequencing, the wifi to be processed after being sorted
Data;
The weight of the matching score value of every wifi data in wifi data to be processed after determining the sequence;Any one
The weight of the matching score value of wifi data is negatively correlated with target range;The target range is this wifi data and adjacent
Distance between router belonging to one wifi data difference;
The matching score value of every group of wifi data to be processed is weighted and the matching of any one group of wifi data to be processed respectively
The numerical value that score value is weighted and obtains is the matching score value of group wifi data to be processed;
By in the wifi data group, the matching score value and each group of the wifi data in addition to each group wifi data to be processed wait locating
The corresponding matching score value of reason wifi data is added;
The computing module determines the MAC Address that the wifi data group includes for the matching score value according to every wifi data
Matching score value between the face, comprising:
The computing module, the MAC Address and the face for including as the wifi data group specifically for the value that will add up
Between matching score value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910749127.6A CN110442658B (en) | 2019-08-14 | 2019-08-14 | Data association method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910749127.6A CN110442658B (en) | 2019-08-14 | 2019-08-14 | Data association method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110442658A true CN110442658A (en) | 2019-11-12 |
CN110442658B CN110442658B (en) | 2022-03-04 |
Family
ID=68435413
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910749127.6A Active CN110442658B (en) | 2019-08-14 | 2019-08-14 | Data association method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110442658B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130290347A1 (en) * | 2012-04-26 | 2013-10-31 | Appsense Limited | Systems and methods for providing data-driven document suggestions |
CN104992075A (en) * | 2015-07-30 | 2015-10-21 | 浙江宇视科技有限公司 | Multi-source information correlation method based on big data |
CN106303398A (en) * | 2015-05-12 | 2017-01-04 | 杭州海康威视数字技术股份有限公司 | monitoring method, server, system and image collecting device |
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 |
-
2019
- 2019-08-14 CN CN201910749127.6A patent/CN110442658B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130290347A1 (en) * | 2012-04-26 | 2013-10-31 | Appsense Limited | Systems and methods for providing data-driven document suggestions |
CN106303398A (en) * | 2015-05-12 | 2017-01-04 | 杭州海康威视数字技术股份有限公司 | monitoring method, server, system and image collecting device |
CN104992075A (en) * | 2015-07-30 | 2015-10-21 | 浙江宇视科技有限公司 | Multi-source information correlation method based on big data |
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 |
Also Published As
Publication number | Publication date |
---|---|
CN110442658B (en) | 2022-03-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105976400B (en) | Method for tracking target and device based on neural network model | |
CN109376596A (en) | Face matching process, device, equipment and storage medium | |
WO2019200902A1 (en) | Image recognition method and device | |
CN108411565A (en) | Washing machine control method and device | |
CN110363076A (en) | Personal information correlating method, device and terminal device | |
CN109034078A (en) | Training method, age recognition methods and the relevant device of age identification model | |
CN109461003A (en) | Plurality of human faces scene brush face payment risk preventing control method and equipment based on multi-angle of view | |
CN108875522A (en) | Face cluster methods, devices and systems and storage medium | |
CN108875517A (en) | Method for processing video frequency, device and system and storage medium | |
CN108875525A (en) | Behavior prediction method, apparatus, system and storage medium | |
CN109886951A (en) | Method for processing video frequency, device and electronic equipment | |
CN108256404A (en) | Pedestrian detection method and device | |
CN108875481A (en) | Method, apparatus, system and storage medium for pedestrian detection | |
CN112364054B (en) | Wind control decision method, device, electronic device and storage medium | |
CN110443224A (en) | Page turning detection method and device, electronic equipment and storage medium | |
CN110298318A (en) | Number of people human body associated detecting method, device and electronic equipment | |
CN109615616A (en) | A kind of crack identification method and system based on ABC-PCNN | |
CN111081257A (en) | Voice acquisition method, device, equipment and storage medium | |
CN109740567A (en) | Key point location model training method, localization method, device and equipment | |
CN110119307A (en) | Processing method, device, storage medium and the electronic device of data processing request | |
US9135562B2 (en) | Method for gender verification of individuals based on multimodal data analysis utilizing an individual's expression prompted by a greeting | |
CN110852372B (en) | Data association method, device and equipment and readable storage medium | |
CN107016324A (en) | A kind of fingerprint image processing method and fingerprint detection equipment | |
CN112199976A (en) | Certificate picture generation method and device | |
CN110442658A (en) | A kind of data correlation method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A data association method and device Granted publication date: 20220304 Pledgee: Haidian Beijing science and technology enterprise financing Company limited by guarantee Pledgor: AIBEE (BEIJING) INTELLIGENT TECHNOLOGY Co.,Ltd. Registration number: Y2024110000191 |
|
PE01 | Entry into force of the registration of the contract for pledge of patent right |