CN110263765A - Image processing method, device and electronic equipment - Google Patents
Image processing method, device and electronic equipment Download PDFInfo
- Publication number
- CN110263765A CN110263765A CN201910642997.3A CN201910642997A CN110263765A CN 110263765 A CN110263765 A CN 110263765A CN 201910642997 A CN201910642997 A CN 201910642997A CN 110263765 A CN110263765 A CN 110263765A
- Authority
- CN
- China
- Prior art keywords
- personage
- group
- combination
- distance
- image
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Multimedia (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Human Computer Interaction (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Image Analysis (AREA)
Abstract
The application provides a kind of image processing method, device and electronic equipment, is related to computer data processing technology field.This method comprises: obtaining the first image and the second image；According to position of the portrait of the first personage group in the first image, the first distance between any two personage in the first personage group is determined；Any two personage that first distance is less than or equal to pre-determined distance is determined as the first personage combination；Based on the portrait in the first personage combination and the portrait in the second personage group, judge same with the presence or absence of figure picture in combine with the first personage in the second personage group, and the second distance between personage is less than or equal to the second personage combination of pre-determined distance；When being combined in the second personage group there are the second personage, the problem of based on the portrait of each personage in the second personage combination, determining the character relation in the second personage combination between each personage, the time cost height for obtaining character relation data, low efficiency can be improved.
Description
Technical field
The present invention relates to computer data processing technology fields, in particular to a kind of image processing method, dress
It sets and electronic equipment.
Background technique
With the continuous development of big data, in the commercial operation area of market class, to pedestrian's sample investigation to obtain correlation
The case where data, is more and more.When being sampled investigation, it usually needs third party consulting mechanism arranges other operators
Work inquires pedestrian by artificial mode in market, and the acquisition of data could be completed after obtaining pedestrian and agreeing to, and investigates and take out
The data volume that sample needs to acquire is big, so that the difficulty of data acquisition is big.For example, being needed when investigating the character relation between pedestrian
The time of the pedestrian of matched sampling is occupied to obtain corresponding data, in addition, since the data volume for needing to obtain is big,
It is generally necessary to which staff goes to be investigated using the long period, the time cost for causing sampling to obtain character relation data is high,
Low efficiency.
Summary of the invention
The application provides a kind of image processing method, device and electronic equipment, can improve and obtain character relation number
According to the problem of time cost is high, low efficiency.
To achieve the goals above, technical solution provided by the embodiment of the present application is as follows:
In a first aspect, the embodiment of the present application provides a kind of image processing method, which comprises
The first image and the second image are obtained, the first image includes the portrait of the first personage group, second figure
Portrait as including the second personage group；According to position of the portrait of first personage group in the first image, really
The first distance between any two personage in fixed first personage group；The first distance is less than or equal to default
Any two personage of distance is determined as the first personage combination；Based on first personage combination in portrait and second people
Portrait in object group, judge it is same with the presence or absence of figure picture in being combined with first personage in second personage group, and
The second personage that second distance between personage is less than or equal to the pre-determined distance combines；It is deposited in second personage group
In second personage combination, based on the portrait of each personage in second personage combination, second person group is determined
Character relation in conjunction between each personage.
In the above scheme, it goes investigation to sample without staff to scene, obtains character relation so as to save
Human resources are conducive to improve the efficiency for obtaining character relation, so as to improve time cost height, the effect for obtaining character relation data
The low problem of rate.
With reference to first aspect, in some alternative embodiments, based on first personage combination in portrait and institute
The portrait in the second personage group is stated, is judged in second personage group with the presence or absence of personage in being combined with first personage
It is identical, and the second distance between personage is less than or equal to the second personage combination of the pre-determined distance, comprising:
By completing the deep learning model after training, to the portrait in first personage combination, second personage
Portrait in group identified, obtain each of the first personage combination object with it is every in second personage group
The similarity of a personage；Judge in second personage group with the presence or absence of each of combining object pair with first personage
The second personage answered, the corresponding target person of any personage in first personage combination be in second personage group with
The similarity highest of any personage, and similarity and the personage for being greater than or equal to preset threshold；In the second personage group
Exist when each of combine the corresponding target person of object with first personage in body, judges that first personage combines
The corresponding target person of each personage between second distance whether be less than or equal to the pre-determined distance；Described the first
When second distance between each of the object combination corresponding target person of object is less than or equal to the pre-determined distance, institute is determined
State in the second personage group that there are second personage combinations.
In the above scheme, by utilization deep learning model, the similarity in various combination between each personage is identified, then
In conjunction with the distance between each personage in same combination, helps to improve and judge in the second crowd with the presence or absence of the effect of the second personage combination
Rate and accuracy.
With reference to first aspect, in some alternative embodiments, based on each personage's in second personage combination
Portrait determines the character relation in the second personage combination between each personage, comprising:
By completing the deep learning model after training, the portrait of each personage in second personage combination is known
Not, each of obtaining combining the corresponding character attribute of object, the character attribute with second personage includes age and property
Not；The character relation in the second personage combination between each personage is determined based on the corresponding character attribute of each personage.
In the above scheme, the distance between personage, personage's respective age, gender can be used as determining character relation
Efficiency factor.It is smaller in the first image, the second image based on the distance between each personage in personage's combination, in conjunction with personage
Age, gender in combination between each personage are just conducive to estimate the character relation between personage.
With reference to first aspect, in some alternative embodiments, the first image and the second image are obtained, comprising:
In the first preset period of time, the first image, Yi Jicong are obtained from the monitor video that the first camera acquires
Second image is obtained in the monitor video of second camera acquisition, wherein first camera, the second camera
It is less than two cameras of set distance for distance in specified region.
In the above scheme, the image acquired by different cameras in different time is conducive to footpath between fields in the image of acquisition
The distance between stranger is more than pre-determined distance, maintains closer distance in a short time between stranger to help to improve and makes
The inaccurate problem of testing result.
With reference to first aspect, in some alternative embodiments, whether there is in judging second personage group
Figure picture is same in combining with first personage, and the second distance between personage is less than or equal to the second of the pre-determined distance
Before personage's combination, the method also includes:
According to position of the portrait of second personage group in second image, second personage group is determined
In any two personage between second distance.
In the above scheme, the second distance being calculated facilitates deciding in the second personage group with the presence or absence of the second people
Object combination.
With reference to first aspect, in some alternative embodiments, the method also includes:
Third image is obtained, the third image includes the portrait of third personage group；In second personage group
When being combined there are second personage, based on the portrait in second personage combination and the people in third personage group
Picture, judge it is same with the presence or absence of figure picture in being combined with second personage in third personage group, and the between personage
The third personage that three distances are less than or equal to the pre-determined distance combines；There are the third party in third personage group
When object combines, based on the portrait of each personage in third personage combination, determine in third personage combination each personage it
Between character relation.
In the above scheme, identical personage's combination is combined with the second personage by judging to whether there is in third image,
The accuracy of character relation identification is helped to improve, is improved and is maintained closer distance between stranger in a short time and detection is tied
The inaccurate problem of fruit.
Second aspect, the embodiment of the present application also provide a kind of image data processing system, and described device includes:
Image acquisition unit, for obtaining the first image and the second image, the first image includes the first personage group
Portrait, second image includes the portrait of the second personage group；
Distance determining unit, for the position according to the portrait of first personage group in the first image, really
The first distance between any two personage in fixed first personage group；
Determination unit is combined, any two personage for the first distance to be less than or equal to pre-determined distance is determined as
First personage combination；
Judging unit, for based on first personage combination in portrait and the portrait in second personage group,
Judge it is same with the presence or absence of figure picture in being combined with first personage in second personage group, and second between personage away from
It is combined from the second personage for being less than or equal to the pre-determined distance；
Relation determination unit, when for being combined in second personage group there are second personage, based on described
The portrait of each personage in second personage combination determines the character relation in the second personage combination between each personage.
In conjunction with second aspect, in some alternative embodiments, the judging unit is also used to:
By completing the deep learning model after training, to the portrait in first personage combination, second personage
Portrait in group identified, obtain each of the first personage combination object with it is every in second personage group
The similarity of a personage；
Judge in second personage group with the presence or absence of each of combining the corresponding mesh of object with first personage
Mark personage, the corresponding target person of any personage in first personage combination be in second personage group with described
The similarity highest of one personage, and similarity and the personage for being greater than or equal to preset threshold；
Exist in second personage group and each of combines the corresponding target person of object with first personage
When, judge whether the second distance between each of the first personage combination corresponding target person of object is less than or equal to
The pre-determined distance；
Second distance between each of the first personage combination corresponding target person of object is less than or equal to
When the pre-determined distance, determine that there are second personage combinations in second personage group.
The third aspect, the embodiment of the present application also provide a kind of electronic equipment, and the electronic equipment includes the place to intercouple
Device and memory are managed, is stored with computer program in the memory, when the processor runs the computer program, is made
It obtains the electronic equipment and executes above-mentioned image processing method.
Fourth aspect, the embodiment of the present application also provide a kind of computer readable storage medium, in the readable storage medium storing program for executing
It is stored with computer program, when the computer program is run on computers, so that the computer executes above-mentioned figure
As data processing method.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in the embodiment attached
Figure is briefly described.It should be appreciated that the following drawings illustrates only some embodiments of the application, therefore it is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is the block diagram of electronic equipment provided by the embodiments of the present application.
Fig. 2 is the flow diagram of image processing method provided by the embodiments of the present application.
Fig. 3 is the schematic diagram of scene where monitoring device provided by the embodiments of the present application.
Fig. 4 is the functional block diagram of image data processing system provided by the embodiments of the present application.
Icon: 10- electronic equipment；11- processing module；12- memory module；The first camera of 21-；22- second camera；
100- image data processing system；110- image acquisition unit；120- distance determining unit；130- combines determination unit；140-
Judging unit；150- relation determination unit.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application is described.It needs
It is noted that term " first ", " second " etc. are only used for distinguishing description, it is not understood to indicate or imply relatively important
Property.
Fig. 1 is please referred to, the application provides a kind of electronic equipment 10, can be used for executing following image real time transfer sides
Method can acquire character relation data from image.Understandably, character relation can include but is not limited to kinship, friend
Relationship, lover's relationship etc..Wherein, kinship includes but is not limited to set membership, mother-child relationship (MCR), father and daughter's relationship, Mu Nvguan
System, sisterhood, elder sister and younger brother's relationship, brotherhood, brother and sister's relationship etc..The character relation collected can be used for for businessman (or
Other Data Analysts) it is analyzed, for example, businessman can adjust the mode of trade marketing based on character relation, thus
Be conducive to improve marketing volume.
Wherein, electronic equipment 10 may include processing module 11, memory module 12 and image data processing system 100,
Between the processing module 11, memory module 12 and each element of image data processing system 100 directly or indirectly electrically
Connection, to realize the transmission or interaction of data.For example, these elements can pass through one or more communication bus or letter between each other
Number line, which is realized, to be electrically connected.
The electronic equipment 10 may be, but not limited to, smart phone, PC (Personal Computer,
PC), tablet computer, personal digital assistant (Personal Digital Assistant, PDA), mobile internet surfing equipment (Mobile
Internet Device, MID), server etc..
In the present embodiment, electronic equipment 10 directly or indirectly can obtain the view that monitoring device acquires from monitoring device
Frequency or image.For example, electronic equipment 10 can be with prison when electronic equipment 10 directly gets video or image from monitoring device
Equipment communication connection is controlled, the video itself acquired or image can be sent to electronic equipment 10 by network by monitoring device, or
Person's electronic equipment 10 can obtain monitoring device video collected or image from monitoring device by network.In another example
When electronic equipment 10 gets video or image from monitoring device indirectly, monitoring device video collected or image are stored in
In moveable computer readable storage medium, which can be but not limited to hard disk, USB flash disk etc., electricity
Sub- equipment 10 can read monitoring device video collected or image from computer-readable storage medium, so that electric
Sub- equipment 10 can get video or image from monitoring device indirectly.
In the present embodiment, monitoring device includes at least one camera for being used to acquire video or image.One electronics
Equipment 10 can be established with an at least monitoring device and be communicated to connect.Understandably, camera included by a monitoring device
Quantity can be configured according to the actual situation, can for one or more.The monitoring communicated to connect with electronic equipment 10 is set
Standby data can be configured according to the actual situation, can be one or more.
Optionally, monitoring device can also include the display screen of the video or image for playing camera acquisition, so as to
In supervisor's checking monitoring equipment picture collected.
In the present embodiment, the camera in monitoring device may be mounted in store or recreation ground, for monitoring quotient
Pedestrian's situation in city or recreation ground specified region.In addition, the image of camera acquisition is used as determining in image
The material image of character relation.
The processing module 11 can be a kind of IC chip, the processing capacity with signal.Above-mentioned processing module
11 can be general processor.For example, the processor can be central processing unit (Central Processing Unit,
CPU), graphics processor (Graphics Processing Unit, GPU), network processing unit (Network Processor,
NP) etc.；It can also be digital signal processor (Digital Signal Processing, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), field programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components may be implemented or execute disclosed each method, step and logic diagram in the embodiment of the present application.
The memory module 12 may be, but not limited to, random access memory, and read-only memory may be programmed read-only deposit
Reservoir, Erasable Programmable Read Only Memory EPROM, electrically erasable programmable read-only memory etc..In the present embodiment, the storage
Module 12 can be used for storing the first image, the second image, third image, deep learning model.Certainly, the memory module 12
It can be also used for storage program, the processing module 11 executes the program after receiving and executing instruction.
Described image data processing equipment 100 includes that at least one can be stored in the form of software or firmware (firmware)
In the memory module 12 or the software that is solidificated in 10 operating system of electronic equipment (Operating System, OS)
Functional module.The processing module 11 is for executing the executable module stored in the memory module 12, such as image data
Software function module included by processing unit 100 and computer program etc..
It is understood that structure shown in FIG. 1 is only a kind of structural schematic diagram of electronic equipment 10, the electronic equipment
10 can also include than more components shown in Fig. 1.Each component shown in Fig. 1 can be real using hardware, software, or its combination
It is existing.
Referring to figure 2., the embodiment of the present application also provides a kind of image processing method, which can
To be applied in above-mentioned electronic equipment 10, can be executed by electronic equipment 10 or each step of implementation method.
In the present embodiment, image processing method may comprise steps of:
Step S210, obtains the first image and the second image, the first image include the portrait of the first personage group, institute
State the portrait that the second image includes the second personage group；
Step S220 determines described according to position of the portrait of first personage group in the first image
The first distance between any two personage in one personage group；
Any two personage that the first distance is less than or equal to pre-determined distance is determined as the first personage by step S230
Combination；
Step S240, based on the portrait in first personage combination and the portrait in second personage group, judgement
It is same with the presence or absence of figure picture in being combined with first personage in second personage group, and the second distance between personage is small
In or equal to the pre-determined distance the second personage combination；
Step S250 is based on second personage when combining in second personage group there are second personage
The portrait of each personage in combination determines the character relation in the second personage combination between each personage.
Understandably, in this kind of Recreational places such as market, recreation ground, even at different times, relationship
The distance between toots's object is usually relatively close, and the distance between not closely related personage usually can not protect in different moments
It holds in closer distance range.For example, the distance between intimate personage almost remains at relatively close in market
Distance range in, and the distance between stranger can not usually be maintained at for a long time in closer distance range.Electronic equipment
10 can be by being tracked identification in different time, different location to identical personage combination, and combines two people in personage's combination
Distance, can determine that each personage in personage's combination with the presence or absence of intimate relationship, and estimates specific personage
Relationship.
In the above-described embodiment, method is by the portrait in the personage group in the first image, the second image, to identify
The character relation of each personage in image personage combination.Sampling is investigated to obtain character relation to crowd scene with staff
It compares, in scheme provided by the present application, goes investigation to sample without staff to scene, so as to save acquisition character relation
Human resources.In addition, when acquiring character relation using electronic equipment 10, electronic equipment 10 can be constantly to different the
One the second image of image carries out identifying processing, and to obtain character relation, processing speed is faster than by investigating crowd manually to obtain
The speed for taking character relation was conducive to improve the efficiency for obtaining character relation, so as to improve the time for obtaining character relation data
The problem of at high cost, low efficiency.
Incorporated by reference to referring to Fig. 2 and Fig. 3, each step of image processing method shown in Fig. 2 will be carried out below detailed
It is thin to illustrate:
Step S210, obtains the first image and the second image, the first image include the portrait of the first personage group, institute
State the portrait that the second image includes the second personage group.
In the present embodiment, electronic equipment 10 can get the first image, the second image from monitoring device.Wherein,
The image that one image, the second image can collect for same camera interval preset period of time in monitoring device；Alternatively, the
One image, the second image can be the image that different cameras interval preset period of time collects in the same monitoring device；Or
Person, the image that the first image, the second image can collect for the different cameras in different monitoring equipment.Wherein, it presets
Period can be configured according to the actual situation, such as can be the durations such as 3 minutes, 5 minutes, be not especially limited here.
As an alternative embodiment, step S210 may include: in the first preset period of time, from the first camera
The first image is obtained in the monitor video of acquisition, and obtains described second from the monitor video that second camera acquires
Image, wherein first camera, the second camera are two camera shootings that distance is less than set distance in specified region
Head.Wherein, the first preset period of time can be configured according to the actual situation, such as can be the durations such as 5 minutes, 6 minutes, here
It is not especially limited.Set distance can be configured according to the actual situation, for example, can for 10 meters, 20 meters it is equidistant.
For example, the first camera 21, second camera 22 are separately positioned in a market, certainly, in market in Fig. 3
Greater number of camera can also be installed.It include retail shop and passageway in market in Fig. 3.Pedestrian can walk on passageway,
First camera 21, second camera 22 are separately mounted to position different in the market, for monitoring the people in different location
Group's (in/out mouth that camera may be mounted in market, upper and lower staircase, enter brand shops).In Fig. 3, formed by dotted line
Fan-shaped region can be regarded as camera and can clearly collect the effective coverage range of face.The effective coverage range can root
It is determined according to the performance parameter of camera, which includes but is not limited to the angular field of view of camera, collects clearly
Effective distance range of face etc. is here not especially limited the performance parameter of camera.
Step S220 determines described according to position of the portrait of first personage group in the first image
The first distance between any two personage in one personage group.
As an alternative embodiment, camera can be used for obtaining the distance between target point and camera.Electricity
Sub- equipment 10 can use camera location algorithm and determine the distance between two personages in monitored picture.Wherein, location algorithm
It can be configured according to the actual situation, for example, electronic equipment 10 is pre-established with two target points in the image of camera acquisition
Mapping relations in the distance between pixel distance, two target points and camera, actual scene at a distance from two target points.Electronics is set
Standby 10 after the pixel distance for determining two target points in camera shooting picture, based between this two target point and camera
Distance, mapping relations and pixel distance, can determine the distance between two target points.
Wherein, two target points can be respectively as the point in any two personage in the first personage group, for example, two targets
Point can be respectively the both feet and the central point of ground face contact of 's object, alternatively, two target points can be respectively two personage heads
Central point.At this point, the distance between two target points can be used as first between corresponding two personages in the first personage group
Distance.
As an alternative embodiment, camera can be used for obtaining the distance between target point and camera, benefit
Pixel distance in angular field of view and shooting picture known to camera between two o'clock can determine two target point pair in image
When should be in actual scene, the angle that two target points and camera line are formed, two mesh in the actual scene obtained based on measurement
The distance of two target points can be calculated using trigonometric function in the distance and angle of punctuate to camera.Calculating two
When the distance between personage, by using two personages as two target points, the distance of calculated two target point is just above-mentioned
One distance.
Any two personage that the first distance is less than or equal to pre-determined distance is determined as the first personage by step S230
Combination.
In the present embodiment, pre-determined distance can be configured according to the actual situation, for example, pre-determined distance can be 0.5
Rice, 1 meter equidistant.
Understandably, if pre-determined distance is 0.5 meter, it is assumed that there are the personages such as A, B, C, D in the first personage group, and determine
A and B out, the distance between A and B be less than 0.5 meter, then can be combined using A with B as the first personage, meanwhile, B and C can
To be combined as another group of the first personage.
Step S240, based on the portrait in first personage combination and the portrait in second personage group, judgement
It is same with the presence or absence of figure picture in being combined with first personage in second personage group, and the second distance between personage is small
In or equal to the pre-determined distance the second personage combination.
In the present embodiment, the distance between any two personage in the second personage group can obtain for measurement in advance
, it is also possible to what the measurement when carrying out judgement processing obtained.
As an alternative embodiment, method can also include: according to second personage before step S240
Position of the portrait of group in second image determines between any two personage in second personage group
Two distances.
Understandably, the mode for measuring second distance is similar with the mode of above-mentioned measurement first distance, no longer superfluous here
It states.After measuring second distance, facilitates electronic equipment 10 and judge to whether there is in the second personage group based on second distance
Figure picture is same in first personage combination, and the second distance between personage is less than or equal to the second person group of the pre-determined distance
It closes.
As an alternative embodiment, step S240 may include:
By completing the deep learning model after training, to the portrait in first personage combination, second personage
Portrait in group identified, obtain each of the first personage combination object with it is every in second personage group
The similarity of a personage；
Judge in second personage group with the presence or absence of each of combining object corresponding the with first personage
Two personages, the corresponding target person of any personage in first personage combination be in second personage group with described
The similarity highest of one personage, and similarity and the personage for being greater than or equal to preset threshold；
Exist in second personage group and each of combines the corresponding target person of object with first personage
When, judge whether the second distance between each of the first personage combination corresponding target person of object is less than or equal to
The pre-determined distance；
Second distance between each of the first personage combination corresponding target person of object is less than or equal to
When the pre-determined distance, determine that there are second personage combinations in second personage group.
In the present embodiment, electronic equipment 10 can be previously stored with the deep learning model after completing training, the depth
Learning model includes but is not limited to convolutional neural networks (Convolutional Neural Networks, CNN) model, circulation
Neural network (Recurrent Neural Network, RNN) model.When being trained to model, electronic equipment 10 can be with
It will be inputted in deep learning model from a large amount of character images that network or other equipment are got, in training, every figure map
A personage is generally included as in.In addition, the gender that can be marked in the character image and age.It completes to train
Afterwards, when identifying portrait using model, by two image input deep learning models comprising portrait, a figure can be obtained
The similarity of any one personage and any one personage in another image as in.In addition, deep learning model can be with defeated
Each personage corresponding age and gender in image out.The preset threshold of similarity can be configured according to the actual situation, such as
Preset threshold can be the threshold values such as 95%, 99%.
Judge in the second crowd in order to make it easy to understand, below illustrating citing with the presence or absence of the realization of the second personage combination
Journey:
For example, it is assumed that in the first personage group including A, B, C, D etc. personages, and determine A and B, the distance between A and B
It is to be less than pre-determined distance；Include the personages such as A ', B ', C ', E ', F ' in second personage group, and determine A ' and B ' between
Distance is less than pre-determined distance.If electronic equipment 10 is using above-mentioned deep learning model in the first personage group, the second personage group
In in identify: in A and the second personage group in the similarity of each personage, the similarity of A and A ' is maximum, and is greater than or equal to
Preset threshold, then it is assumed that A and A ' is same people (understandably, if the similarity of A and A ' is less than preset threshold, then it is assumed that A and A '
It is not same people；If the similarity of A and A ' is greater than or equal to preset threshold, but it is A and the second crowd that the similarity of A and A ', which is not,
In other people-similarities it is maximum, then it is assumed that A and A ' is not same people).If electronic equipment 10 also identifies B and B ' it is same
People, C and C ' is after same people.Electronic equipment 10 may determine that the first personage in the first personage group combination be A and B, B with
C；There are the first personages to combine identical second personage combination in second personage group, which is combined into A ' and B '.
In the above example, if the distance between A ' and B ' are greater than pre-determined distance, other conditions are constant, then, electronics
Equipment 10, which judges out to be not present in the second personage group, combines identical second personage combination with the first personage.Similarly, if
Electronic equipment 10 identifies A and A ' be not the same person, electronic equipment 10 also judge out in the second personage group there is no with
First personage combines identical second personage combination.
Step S250 is based on second personage when combining in second personage group there are second personage
The portrait of each personage in combination determines the character relation in the second personage combination between each personage.
As an alternative embodiment, step S250 may include: the deep learning model after being trained by completion,
The portrait of each personage in second personage combination is identified, each of obtains combining object with second personage
Corresponding character attribute, the character attribute include age and gender；Based on described in the corresponding character attribute determination of each personage
Character relation in second personage combination between each personage.
In the present embodiment, deep learning model can be also used for identifying the age of personage in image and gender.
Because will do it model training before deep learning model identifies character attribute, so that when model identifies character attribute, identification
The accuracy of obtained gender and age is high, to be conducive to improve the accuracy of determining character relation.In addition, deep learning
Model is fast to the speed of data identifying processing, is conducive to character relation and rapidly obtains.
When character relation in determining the second personage combination using character attribute between personage, implementation can be with
Are as follows: based on the gender between personage and specified the range of age where age gap, determine the character relation between personage.Wherein,
Specified the range of age can be configured according to the actual situation.For example, the corresponding specified the range of age of age gap (being expressed as Y) can
To include: 0 < Y≤10；10<Y≤20；Y > 20 etc..
For example, in the second personage combination, if the gender for identifying two people is the same sex, it is assumed that identify the age of two people
Difference is greater than 20 years old, and two people are all male, then can judge the relationship of two people tentatively for set membership, if both are women,
It then can tentatively judge the relationship of two people for mother and daughter relationship.When two people are with gender, if two people's age gaps are less than or equal to 20 years old,
Then think that two people are friend's relationship.It, can be with if the age gap of two people is less than or equal to 10 years old when the gender of two people is anisotropic
Primarily determine that the relationship between two people is lover's relationship；If the age gap of two people is greater than 10 years old, and is less than or equal to 20 years old, then may be used
To be initially believed that the relationship of two people for friend's relationship；If the age gap of two people is greater than 20 years old, and the big female of male is small, then can tentatively recognize
Relationship for two people is father and daughter's relationship；If the age gap of two people is greater than 20 years old, and the small female of male is big, then can be initially believed that two people's
Relationship is mother-child relationship (MCR).
As an alternative embodiment, the face characteristic that character attribute can also include personage can in step S250
Come with the face characteristic in conjunction with each personage in the second personage combination, age and gender come determine in the second personage combination each personage it
Between character relation.It wherein, may include eye feature, nose feature, mouth feature, face contour feature in face characteristic
Deng.
When progress character relation determines, electronic equipment 10 can be to the face characteristic of two personages in the second personage combination
It is compared, to obtain similarity every in face characteristic.Electronic equipment 10 can the similarity based on various features come really
The character relation of two people in fixed second personage combination.For example, electronic equipment 10 can be in any one of face characteristic feature
When similarity is more than preset threshold (preset threshold can be configured according to the actual situation, for example be 90%, 95% etc.), or
Person is more than preset threshold in the comprehensive similarity of face characteristic, it is determined that the relationship of two personages is kinship.Then it is combining
The gender of two personages and age determine specific kinship.Wherein, comprehensive similarity can for each phase character average value or
For weighted average, wherein the weight of each phase character can be configured according to the actual situation,
For example, electronic equipment 10 detects that the comprehensive similarity of the face characteristic of two people is greater than in the second personage combination
First preset threshold, the age gap of two people is greater than default age threshold (such as default age threshold be 20 years old), and the small female of male
Greatly, it may be considered that the relationship of two people is mother-child relationship (MCR).Alternatively, electronic equipment 10 detects that the eye feature similarity of two people is big
In the second preset threshold, the age gap of two people is greater than default age threshold (such as default age threshold be 20 years old), and male is small
Female is big, it may be considered that the relationship of two people is mother-child relationship (MCR).Wherein, the first preset threshold and the second preset threshold can be according to realities
Border situation is configured, and may be the same or different, and is not especially limited here.
Understandably, the reason of between lineal relative because of gene genetic, (face can correspond to the face of two personages
The face characteristic stated) in usually have partly or entirely it is similar.The similarity in conjunction with face come judge the second personage combination in people
When relationship between object, be conducive to the accuracy judged kinship raising.
As an alternative embodiment, method can also include:
Third image is obtained, the third image includes the portrait of third personage group；
When being combined in second personage group there are second personage, based on the people in second personage combination
Portrait in picture and third personage group, judging whether there is in third personage group combines with second personage
Middle figure picture is same, and the third distance between personage is less than or equal to the third personage combination of the pre-determined distance；
When being combined in third personage group there are the third personage, based on each in third personage combination
The portrait of personage determines the character relation in the third personage combination between each personage.
In the present embodiment, electronic equipment 10 can get the third figure of third camera acquisition from third camera
Picture.Wherein, the installation site of third camera at a distance from the first camera 21, second camera 22 within the set range, should
Setting range can be configured according to the actual situation, for example can be any one distance range before 10 meters to 20 meters,
Any one distance range between 15 meters to 30 meters.
It is worth noting that electronic equipment 10 is combined by the second personage, third image recognition obtains third personage and combines
In character relation between each personage mode, be based on the first image and the second image with above-mentioned and identify to obtain the second person group
The mode of character relation in conjunction between each personage is similar, and specific implementation can refer to the detailed of step S240, S250
Description, which is not described herein again.
Based on this, method, which can improve, leads to character relation judgement not because the short time maintains closer distance between stranger
Accurate problem.
Certainly, electronic equipment 10 can also combine the image combination third personage that other cameras in specified region acquire
In personage carry out identifying processing, to improve the accuracy of determining character relation.
Referring to figure 4., the embodiment of the present application also provides a kind of image data processing system 100.Image real time transfer dress
Setting 100 can be applied in above-mentioned electronic equipment 10, can be used for executing or realizing above-mentioned image processing method.
In the present embodiment, image data processing system 100 may include image acquisition unit 110, distance determining unit
120, determination unit 130, judging unit 140 and relation determination unit 150 are combined.Each unit in image data processing system 100
Function it is as follows:
Image acquisition unit 110, for obtaining the first image and the second image, the first image includes the first personage group
The portrait of body, second image include the portrait of the second personage group；
Distance determining unit 120, for the position according to the portrait of first personage group in the first image,
Determine the first distance between any two personage in first personage group；
Determination unit 130 is combined, any two personage for the first distance to be less than or equal to pre-determined distance is true
It is set to the first personage combination；
Judging unit 140, for based on the portrait in first personage combination and the people in second personage group
Picture, judge it is same with the presence or absence of figure picture in being combined with first personage in second personage group, and the between personage
The second personage that two distances are less than or equal to the pre-determined distance combines；
Relation determination unit 150 is based on institute when for combining in second personage group there are second personage
The portrait for stating each personage in the second personage combination determines the character relation in the second personage combination between each personage.
Optionally, the judging unit 140 is also used to: by completing the deep learning model after training, to described first
Portrait in personage's combination, the portrait in second personage group identify, obtain every in the first personage combination
The similarity of each of a personage and second personage group object；Judge in second personage group with the presence or absence of with
Each of the first personage combination corresponding target person of object, any personage in the first personage combination are corresponding
Target person is similarity highest with any personage in second personage group, and similarity and is greater than or equal to pre-
If the personage of threshold value；Exist in second personage group and each of combines the corresponding target of object with first personage
When personage, judge the second distance between each of the first personage combination corresponding target person of object whether be less than or
Equal to the pre-determined distance；Second distance between each of the first personage combination corresponding target person of object is small
When the pre-determined distance, determine that there are second personage combinations in second personage group.
Optionally, relation determination unit 150 is also used to: by completing the deep learning model after training, to described second
The portrait of each personage in personage's combination identifies, each of obtains combining the corresponding personage of object with second personage
Attribute, the character attribute include age and gender；Second person group is determined based on the corresponding character attribute of each personage
Character relation in conjunction between each personage.
Optionally, image acquisition unit 110 is also used to: in the first preset period of time, from the prison of the first camera 21 acquisition
The first image is obtained in control video, and obtains second image from the monitor video that second camera 22 acquires,
Wherein, first camera 21, the second camera 22 are two cameras that distance is less than set distance in specified region.
Optionally, judge to whether there is in second personage group in judging unit 140 and be combined with first personage
Middle figure picture is same, and the second distance between personage is less than or equal to before the second personage combination of the pre-determined distance, distance
Determination unit 120 can be also used for: according to position of the portrait of second personage group in second image, determine institute
State the second distance between any two personage in the second personage group.
Optionally, image acquisition unit 110 can also be used in: obtain third image, the third image includes third personage
The portrait of group.Judging unit 140 can also be used in: when being combined in second personage group there are second personage, base
In second personage combination in portrait and the portrait in third personage group, judge be in third personage group
It is no same in the presence of figure picture in being combined with second personage, and the third distance between personage is less than or equal to the pre-determined distance
Third personage combination.Relation determination unit 150 can also be used in: there are the third person groups in third personage group
When conjunction, based on the portrait of each personage in third personage combination, determine in the third personage combination between each personage
Character relation.
It should be noted that it is apparent to those skilled in the art that, for convenience and simplicity of description, on
The specific work process of the image data processing system 100 of description is stated, each step corresponding process in preceding method can be referred to,
It no longer excessively repeats herein.
The embodiment of the present application also provides a kind of computer readable storage medium.Computer journey is stored in readable storage medium storing program for executing
Sequence, when computer program is run on computers, so that computer is executed as at above-mentioned image data as described in the examples
Manage construction method.
Through the above description of the embodiments, those skilled in the art can be understood that the application can lead to
Hardware realization is crossed, the mode of necessary general hardware platform can also be added to realize by software, based on this understanding, this Shen
Technical solution please can be embodied in the form of software products, which can store in a non-volatile memories
In medium (can be CD-ROM, USB flash disk, mobile hard disk etc.), including some instructions are used so that a computer equipment (can be
Personal computer, server or network equipment etc.) execute method described in each implement scene of the application.
In conclusion the application provides a kind of image processing method, device and electronic equipment.This method comprises: obtaining
The first image and the second image are taken, the first image includes the portrait of the first personage group, and the second image includes the second personage group
Portrait；According to position of the portrait of the first personage group in the first image, any two in the first personage group are determined
First distance between personage；Any two personage that first distance is less than or equal to pre-determined distance is determined as the first person group
It closes；Based on the portrait in the first personage combination and the portrait in the second personage group, judge to whether there is in the second personage group
Figure picture is same in combining with the first personage, and the second distance between personage is less than or equal to the second person group of pre-determined distance
It closes；When combining in the second personage group there are the second personage, based on the portrait of each personage in the second personage combination, the is determined
Character relation in two personages combination between each personage goes investigation to sample without staff to scene, so as to improve people is obtained
The problem of time cost of object relation data is high, low efficiency.
It can replace, can be realized wholly or partly by software, hardware, firmware or any combination thereof.When
When using software realization, can entirely or partly it realize in the form of a computer program product.The computer program product
Including one or more computer instructions.It is all or part of when loading on computers and executing the computer program instructions
Ground is generated according to process or function described in the embodiment of the present application.The computer can be general purpose computer, special purpose computer,
Computer network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or
Person is transmitted from a computer readable storage medium to another computer readable storage medium, for example, the computer instruction
Wired (such as coaxial cable, optical fiber, digital subscriber can be passed through from a web-site, computer, server or data center
Line (DSL)) or wireless (such as infrared, wireless, microwave etc.) mode to another web-site, computer, server or data
It is transmitted at center.The computer readable storage medium can be any usable medium that computer can access and either wrap
The data storage devices such as server, the data center integrated containing one or more usable mediums.The usable medium can be magnetic
Property medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk
Solid State Disk (SSD)) etc..
The foregoing is merely preferred embodiment of the present application, are not intended to limit this application, for the skill of this field
For art personnel, various changes and changes are possible in this application.Within the spirit and principles of this application, made any to repair
Change, equivalent replacement, improvement etc., should be included within the scope of protection of this application.
Claims (10)
1. a kind of image processing method, which is characterized in that the described method includes:
The first image and the second image are obtained, the first image includes the portrait of the first personage group, the second image packet
Include the portrait of the second personage group；
According to position of the portrait of first personage group in the first image, determine in first personage group
First distance between any two personage；
Any two personage that the first distance is less than or equal to pre-determined distance is determined as the first personage combination；
Based on the portrait in first personage combination and the portrait in second personage group, the second personage group is judged
It is same with the presence or absence of figure picture in being combined with first personage in body, and the second distance between personage is less than or equal to described pre-
If the second personage of distance combines；
When being combined in second personage group there are second personage, based on each personage in second personage combination
Portrait, determine the character relation in second personage combination between each personage.
2. the method according to claim 1, wherein based on the portrait and described the in first personage combination
Portrait in two personage groups judges in second personage group with the presence or absence of figure picture in combining with first personage
Together, the second personage that the second distance and between personage is less than or equal to the pre-determined distance combines, comprising:
By completing the deep learning model after training, to the portrait in first personage combination, second personage group
In portrait identified, obtain each of first personage combination each of object and second personage group
The similarity of object；
Judge in second personage group with the presence or absence of each of combining corresponding second people of object with first personage
Object, the corresponding target person of any personage in first personage combination be in second personage group with any people
The similarity highest of object, and similarity and the personage for being greater than or equal to preset threshold；
Exist when each of combining the corresponding target person of object with first personage in second personage group, sentences
It is described whether the second distance between each of the first personage combination corresponding target person of object that breaks is less than or equal to
Pre-determined distance；
Second distance between each of the first personage combination corresponding target person of object is less than or equal to described
When pre-determined distance, determine that there are second personage combinations in second personage group.
3. the method according to claim 1, wherein the people based on each personage in second personage combination
Picture determines the character relation in the second personage combination between each personage, comprising:
By completing the deep learning model after training, the portrait of each personage in second personage combination is identified,
Each of obtaining combining the corresponding character attribute of object, the character attribute with second personage includes age and gender；
The character relation in the second personage combination between each personage is determined based on the corresponding character attribute of each personage.
4. the method according to claim 1, wherein obtaining the first image and the second image, comprising:
In the first preset period of time, the first image is obtained from the monitor video that the first camera acquires, and from second
Second image is obtained in the monitor video of camera acquisition, wherein first camera, the second camera are to refer to
Determine two cameras that distance in region is less than set distance.
5. the method according to claim 1, wherein whether there is and institute in judging second personage group
It is same to state figure picture in the first personage combination, and the second distance between personage is less than or equal to the second personage of the pre-determined distance
Before combination, the method also includes:
According to position of the portrait of second personage group in second image, determine in second personage group
Second distance between any two personage.
6. the method according to claim 1, wherein the method also includes:
Third image is obtained, the third image includes the portrait of third personage group；
In second personage group there are second personage combine when, based on second personage combination in portrait and
Portrait in third personage group judges in third personage group with the presence or absence of people in combining with second personage
Object is identical, and the third distance between personage is less than or equal to the third personage combination of the pre-determined distance；
When being combined in third personage group there are the third personage, based on each personage in third personage combination
Portrait, determine the character relation in third personage combination between each personage.
7. a kind of image data processing system, which is characterized in that described device includes:
Image acquisition unit includes the people of the first personage group for the first image of acquisition and the second image, the first image
Picture, second image include the portrait of the second personage group；
Distance determining unit determines institute for the position according to the portrait of first personage group in the first image
State the first distance between any two personage in the first personage group；
Determination unit is combined, any two personage for the first distance to be less than or equal to pre-determined distance is determined as first
Personage's combination；
Judging unit, for based on the portrait in first personage combination and the portrait in second personage group, judgement
It is same with the presence or absence of figure picture in being combined with first personage in second personage group, and the second distance between personage is small
In or equal to the pre-determined distance the second personage combination；
Relation determination unit is based on described second when for combining in second personage group there are second personage
The portrait of each personage in personage's combination determines the character relation in the second personage combination between each personage.
8. device according to claim 7, which is characterized in that the judging unit is also used to:
By completing the deep learning model after training, to the portrait in first personage combination, second personage group
In portrait identified, obtain each of first personage combination each of object and second personage group
The similarity of object；
Judge in second personage group with the presence or absence of each of combining the corresponding target person of object with first personage
Object, the corresponding target person of any personage in first personage combination be in second personage group with any people
The similarity highest of object, and similarity and the personage for being greater than or equal to preset threshold；
Exist when each of combining the corresponding target person of object with first personage in second personage group, sentences
It is described whether the second distance between each of the first personage combination corresponding target person of object that breaks is less than or equal to
Pre-determined distance；
Second distance between each of the first personage combination corresponding target person of object is less than or equal to described
When pre-determined distance, determine that there are second personage combinations in second personage group.
9. a kind of electronic equipment, which is characterized in that the electronic equipment includes the processor and memory to intercouple, described to deposit
Computer program is stored in reservoir, when the processor runs the computer program, so that the electronic equipment executes
Image processing method as described in any one of claim 1-6.
10. a kind of computer readable storage medium, which is characterized in that it is stored with computer program in the readable storage medium storing program for executing,
When the computer program is run on computers, so that the computer is executed such as any one of claim 1-6 institute
The image processing method stated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910642997.3A CN110263765A (en) | 2019-07-16 | 2019-07-16 | Image processing method, device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910642997.3A CN110263765A (en) | 2019-07-16 | 2019-07-16 | Image processing method, device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110263765A true CN110263765A (en) | 2019-09-20 |
Family
ID=67926625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910642997.3A Pending CN110263765A (en) | 2019-07-16 | 2019-07-16 | Image processing method, device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110263765A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724454A (en) * | 2021-08-25 | 2021-11-30 | 上海擎朗智能科技有限公司 | Interaction method of mobile equipment, device and storage medium |
| WO2023147764A1 (en) * | 2022-02-04 | 2023-08-10 | Huawei Technologies Co., Ltd. | Systems, methods, and media for main group identification in images via social relation recognition |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103697A (en) * | 2009-12-22 | 2011-06-22 | 索尼公司 | Information processing device, method, and program |
| US8321358B2 (en) * | 2009-10-26 | 2012-11-27 | Hon Hai Precision Industry Co., Ltd. | Interpersonal relationships analysis system and method which computes distances between people in an image |
| CN105183758A (en) * | 2015-07-22 | 2015-12-23 | 深圳市万姓宗祠网络科技股份有限公司 | Content recognition method for continuously recorded video or image |
| US20170124387A1 (en) * | 2015-10-30 | 2017-05-04 | Canon Kabushiki Kaisha | Control apparatus and control method for determining relation of persons included in an image, and storage medium storing a program therefor |
| CN106709765A (en) * | 2017-01-11 | 2017-05-24 | 北京图知天下科技有限责任公司 | Advertisement delivery management method |
| CN106803056A (en) * | 2015-11-26 | 2017-06-06 | 华为技术有限公司 | The method of estimation and device of a kind of limbs relation |
| CN106951858A (en) * | 2017-03-17 | 2017-07-14 | 中国人民解放军国防科学技术大学 | A kind of recognition methods of personage's affiliation and device based on depth convolutional network |
| CN106980830A (en) * | 2017-03-17 | 2017-07-25 | 中国人民解放军国防科学技术大学 | One kind is based on depth convolutional network from affiliation recognition methods and device |
| CN106980831A (en) * | 2017-03-17 | 2017-07-25 | 中国人民解放军国防科学技术大学 | Based on self-encoding encoder from affiliation recognition methods |
| US20180314850A1 (en) * | 2017-05-01 | 2018-11-01 | International Business Machines Corporation | Method for showing faces in digital images only to friends |
| CN109766786A (en) * | 2018-12-21 | 2019-05-17 | 深圳云天励飞技术有限公司 | Character relation analysis method and Related product |
| CN109948447A (en) * | 2019-02-21 | 2019-06-28 | 山东科技大学 | The discovery of personage's cyberrelationship and evolution rendering method based on video image identification |
| CN109961325A (en) * | 2019-03-21 | 2019-07-02 | 刘昊洋 | Advertisement recommended method, device, system and mobile TV based on character relation |
-
2019
- 2019-07-16 CN CN201910642997.3A patent/CN110263765A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8321358B2 (en) * | 2009-10-26 | 2012-11-27 | Hon Hai Precision Industry Co., Ltd. | Interpersonal relationships analysis system and method which computes distances between people in an image |
| CN102103697A (en) * | 2009-12-22 | 2011-06-22 | 索尼公司 | Information processing device, method, and program |
| CN105183758A (en) * | 2015-07-22 | 2015-12-23 | 深圳市万姓宗祠网络科技股份有限公司 | Content recognition method for continuously recorded video or image |
| US20170124387A1 (en) * | 2015-10-30 | 2017-05-04 | Canon Kabushiki Kaisha | Control apparatus and control method for determining relation of persons included in an image, and storage medium storing a program therefor |
| CN106803056A (en) * | 2015-11-26 | 2017-06-06 | 华为技术有限公司 | The method of estimation and device of a kind of limbs relation |
| CN106709765A (en) * | 2017-01-11 | 2017-05-24 | 北京图知天下科技有限责任公司 | Advertisement delivery management method |
| CN106951858A (en) * | 2017-03-17 | 2017-07-14 | 中国人民解放军国防科学技术大学 | A kind of recognition methods of personage's affiliation and device based on depth convolutional network |
| CN106980830A (en) * | 2017-03-17 | 2017-07-25 | 中国人民解放军国防科学技术大学 | One kind is based on depth convolutional network from affiliation recognition methods and device |
| CN106980831A (en) * | 2017-03-17 | 2017-07-25 | 中国人民解放军国防科学技术大学 | Based on self-encoding encoder from affiliation recognition methods |
| US20180314850A1 (en) * | 2017-05-01 | 2018-11-01 | International Business Machines Corporation | Method for showing faces in digital images only to friends |
| CN109766786A (en) * | 2018-12-21 | 2019-05-17 | 深圳云天励飞技术有限公司 | Character relation analysis method and Related product |
| CN109948447A (en) * | 2019-02-21 | 2019-06-28 | 山东科技大学 | The discovery of personage's cyberrelationship and evolution rendering method based on video image identification |
| CN109961325A (en) * | 2019-03-21 | 2019-07-02 | 刘昊洋 | Advertisement recommended method, device, system and mobile TV based on character relation |
Non-Patent Citations (3)
| Title |
|---|
| ILYA SOCHENKOV等: "Visual Duplicates Image Search for a Non-cooperative Person Recognition at a Distance", 《PROCEDIA ENGINEERING》 * |
| 许庆勇著: "《基于深度学习理论的纹身图像识别与检测研究》", 31 December 2018, 武汉：华中科技大学出版社 * |
| 陈大海: "固定摄像头图像中测算目标距离和尺寸的算法", 《大众科技》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113724454A (en) * | 2021-08-25 | 2021-11-30 | 上海擎朗智能科技有限公司 | Interaction method of mobile equipment, device and storage medium |
| WO2023147764A1 (en) * | 2022-02-04 | 2023-08-10 | Huawei Technologies Co., Ltd. | Systems, methods, and media for main group identification in images via social relation recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5994397B2 (en) | Information processing apparatus, information processing method, and program | |
| WO2021082635A1 (en) | Region of interest detection method and apparatus, readable storage medium and terminal device | |
| CN104103030B (en) | Image analysis method, camera apparatus, control apparatus and control method | |
| CN110874583A (en) | Passenger flow statistics method and device, storage medium and electronic equipment | |
| CN104813339A (en) | Methods, devices and systems for detecting objects in a video | |
| US11688077B2 (en) | Adaptive object tracking policy | |
| CN107194361A (en) | Two-dimentional pose detection method and device | |
| JP2007201556A (en) | Congestion information providing system and method therefor | |
| CN109902551A (en) | The real-time stream of people's statistical method and device of open scene | |
| CN110263765A (en) | Image processing method, device and electronic equipment | |
| Radaelli et al. | Using cameras to improve wi-fi based indoor positioning | |
| CN108229418A (en) | Human body critical point detection method and apparatus, electronic equipment, storage medium and program | |
| WO2021249306A1 (en) | Crowd overdensity prediction method and device | |
| CN108647587A (en) | Demographic method, device, terminal and storage medium | |
| CN112668525A (en) | People flow counting method and device, electronic equipment and storage medium | |
| CN110245645A (en) | Face vivo identification method, device, equipment and storage medium | |
| CN109872407A (en) | A kind of face identification method, device, equipment and punch card method, device and system | |
| Xie et al. | Event-based stereo matching using semiglobal matching | |
| CN110264497B (en) | Method and device for determining tracking duration, storage medium and electronic device | |
| CN113344906B (en) | Camera evaluation method and device in vehicle-road cooperation, road side equipment and cloud control platform | |
| CN114882251A (en) | Internet of things big data intelligent video security monitoring method and device | |
| JP7484492B2 (en) | Radar-based attitude recognition device, method and electronic device | |
| EP4198772A1 (en) | Method and device for making music recommendation | |
| WO2021138893A1 (en) | Vehicle license plate recognition method and apparatus, electronic device, and storage medium | |
| Chow | A crowdsourcing–geocomputational framework of mobile crowd simulation and estimation |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190920 |
|
| RJ01 | Rejection of invention patent application after publication |