CN108022161A - Clothing matching commending system based on image recognition and big data analysis - Google Patents
Clothing matching commending system based on image recognition and big data analysis Download PDFInfo
- Publication number
- CN108022161A CN108022161A CN201711437438.6A CN201711437438A CN108022161A CN 108022161 A CN108022161 A CN 108022161A CN 201711437438 A CN201711437438 A CN 201711437438A CN 108022161 A CN108022161 A CN 108022161A
- Authority
- CN
- China
- Prior art keywords
- module
- clothes
- feature
- portrait
- data
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0641—Shopping interfaces
- G06Q30/0643—Graphical representation of items or shoppers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- 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
- G06V40/172—Classification, e.g. identification
-
- 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
- G06V40/174—Facial expression recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06V40/178—Human faces, e.g. facial parts, sketches or expressions estimating age from face image; using age information for improving recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of clothing matching commending system based on image recognition and big data analysis, including picture acquisition module, Identification of Images module, clothes identification module, portrait feature output module, garment feature output module, first server, second server, the 3rd server, portrait biometric data storehouse and clothes biometric data storehouse.The present invention provides a kind of clothing matching commending system based on image recognition and big data analysis, it is not necessary to which clothes, which can be achieved, in user data recommends, real-time, recommends matching degree high.
Description
Technical field
The present invention relates to software technology field, more specifically, it is related to a kind of based on image recognition and big data analysis
Clothing matching commending system.
Background technology
In the prior art, such as the Chinese patent application of Publication No. CN105913297A discloses and a kind of is based on big data
Fitting recommend method and system, including data acquisition interface and clothes recommended engine, clothes recommended engine is used to do big data
Analysis phase, recommends suitable clothes by data mining technology for user.Technical side disclosed in the specification of the patent application
Case can increase user and be tested in purchase clothes body on the net, while be capable of the paving goods of convenience store's furniture body.In another example Publication No.
The Chinese patent application of CN106649300A discloses a kind of intelligent clothing matching based on cloud platform and recommends method, the patent Shen
Technical solution disclosed in specification please, which can be automatically selected and arranged in pairs or groups out, meets the clothes of user demand.For another example publication number
A kind of garment size commending system based on multilayer neural network is disclosed for the Chinese patent application of CN107180375A, is used
One six layers of neutral net, including a data input layer, four hidden layers and an output layer；The explanation of the patent application
Technical solution disclosed in book realizes that garment size is recommended intelligent and diversified.
With the fast development of national consumption upgrading, apparel industry is also upgraded to experience consumption, and user is purchasing
Buy clothes can run into often do not know select which type of clothes, clothes how to arrange in pairs or groups, suitable the problems such as being not suitable for oneself, precisely
Garment coordination recommend to contribute to user quickly to select commodity, clothes of the prior art recommend method to pass through frequently with collaborative filtering
Method or commending contents algorithm, such as the commodity bought according to user are recommended, or the people similar to user interest
The commodity bought recommend etc., but these methods need user related data that accumulation could be recommended, real-time
Matching degree is poor with recommending.
The content of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of based on image recognition and big data analysis
Clothing matching commending system, it is not necessary to which clothes, which can be achieved, in user data recommends, real-time, recommends matching degree high.
The purpose of the present invention is what is be achieved through the following technical solutions：It is a kind of based on image recognition and big data analysis
Clothing matching commending system, including picture acquisition module, Identification of Images module, clothes identification module, portrait feature output module,
Garment feature output module, first server, second server, the 3rd server, portrait biometric data storehouse and clothes are special
Levy identification database, be provided with first server portrait feature extraction program module, portrait characteristic matching program module and
First transport module, be provided with second server garment feature extraction procedure module, garment feature matcher module and
Second transport module, is provided with analysis program module, calculation procedure module, collator module and recommendation in the 3rd server
Module；The picture acquisition module is used to obtain picture, and after picture is got, the Identification of Images module is to getting
Picture carries out Identification of Images processing, and the clothes identification module carries out clothes identifying processing to the picture of acquisition, and the portrait is special
Sign output module is used to the figure information after the Identification of Images resume module being transmitted to the first server, described the
In one server, the portrait feature extraction program module is used to extract portrait characteristic, the portrait biometric data
Storehouse is used to store portrait characteristic, and the portrait characteristic matching program module is used to match the portrait characteristic and pass through
For first transport module by portrait matching result data transfer to the 3rd server, the garment feature output module is used for will
Clothing information after the clothes identification module processing is transmitted to the second server, in second server, the clothes
Feature extraction program module is used to extract garment feature data, and the garment feature identification database is used to store garment feature number
According to the garment feature matcher module is used to match the garment feature data and will be taken by second transport module
Matching result data transfer is filled to the 3rd server；In the 3rd server, the analysis program module is used for according to portrait
Data control is carried out with result data and clothes matching result data, the calculation procedure module is used to calculate data contrast ratio
Rate, the collator module are used to data control ratio being ranked up, and the recommending module is used to present to user
Clothes recommendation information.
Further, including first artificial intelligence program's module, the first artificial intelligence program module are arranged on first
In server, for intelligence learning portrait feature recognition.
Further, including second artificial intelligence program's module, the second artificial intelligence program module are arranged on two clothes
It is engaged in device, is identified for intelligence learning garment feature.
Further, including third party's work intelligent program module, third party's work intelligent program module are arranged on three clothes
It is engaged in device, is compareed for intelligence learning data, and recommends for intelligence learning.
Further, including trend clothing information database, for storing trend clothing information data.
Further, including storage program module, picture feature data are stored in the way of four-tuple, for preserving
Depth network carries out data during forward process, and for be stored in it is backward seek gradient procedure when gradient data.
Further, including the first artificial intelligence program module, second artificial intelligence program's module and third party's work
Intelligent program module, the first artificial intelligence program module are arranged in first server, for intelligence learning portrait feature
Identification, the second artificial intelligence program module is arranged in two servers, is identified for intelligence learning garment feature, and described the
Three artificial intelligence program's modules are arranged in three servers, are compareed for intelligence learning data, and are recommended for intelligence learning.
Further, the Identification of Images module includes head feature recognizer module, trunk feature recognition program mould
Block, stature feature recognition program module, the head feature recognizer module are used to identify the portrait head feature on picture,
The trunk feature recognition program module is used to identify the portrait trunk feature on picture, the stature feature recognition program module
For identifying the portrait stature feature on picture.
Further, the clothes identification module includes clothes fashion recognizer module, model recognizer mould
Block and clothes dress material recognizer module, the clothes fashion recognizer module are used to identify the clothes fashion on picture, institute
State model recognizer module to be used to identify the clothes fashion on picture, the clothes dress material recognizer module is used to know
Clothes dress material on other picture.
Further, the portrait matching result data include age data, skin color data, gender data and expression
Data；The clothing matching result data includes trend clothing information data.
The beneficial effects of the invention are as follows：
(1) picture is identified in the present invention, and wear clothes to the human body on picture and human body is identified jointly, compares
In conventional clothes recommended technology, it is not necessary to which accumulating user data can realize that clothes are recommended, real-time, recommend matching degree
It is high.
(2) characteristic information of acquisition is carried out depth by the present invention during to picture recognition by neutral net
Practise, expansion and unmanned supervised learning are carried out to clothes identification storehouse based on artificial intelligence technology, more perfect identification can be formed
Model and identification storehouse, and then improve and recommend matching degree.
(3) the present invention provides a kind of clothes to recommend method and system, by image recognition by the current clothes of user with taking
Dress is matched in storehouse, quickly recommends the other clothes mutually arranged in pairs or groups with itself clothes for user.
(4) present invention is learnt using big data analysis, artificial intelligence, can carry out trend clothes recommendation to user.
The technique effect of the present invention is not limited to the described above.Above technique effect is merely exemplary to illustrate, of the invention
Other features and its effect etc. will be described in detail in subsequent specific embodiment part.Those skilled in the art can root
It can directly or indirectly know remaining technique effect etc. according to the description of the present application, details are not described herein again.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also
To obtain other attached drawings according to these attached drawings.
Fig. 1 is the schematic diagram of the flow in the present invention.
Embodiment
Technical scheme is described in further detail below in conjunction with the accompanying drawings, but protection scope of the present invention is not limited to
It is as described below.All features disclosed in this specification, or implicit disclosed all methods or during the step of, except mutual
Beyond the feature and/or step of repulsion, it can combine in any way.
Any feature disclosed in this specification (including any accessory claim, summary and attached drawing), except non-specifically chatting
State, can be replaced by other alternative features that are equivalent or have similar purpose.I.e., unless specifically stated, each feature
It is an example in a series of equivalent or similar characteristics.
The specific embodiment of the present invention is described more fully below, it should be noted that the embodiments described herein is served only for illustrating
Illustrate, be not intended to limit the invention.In the following description, in order to provide a thorough understanding of the present invention, a large amount of spies are elaborated
Determine details.It will be apparent, however, to one skilled in the art that：This hair need not be carried out using these specific details
It is bright.In other instances, in order to avoid obscuring the present invention, known circuit, software or method are not specifically described.
As shown in Figure 1, a kind of clothing matching commending system based on image recognition and big data analysis, including picture obtain
Module, Identification of Images module, clothes identification module, portrait feature output module, garment feature output module, first server,
Second server, the 3rd server, portrait biometric data storehouse and clothes biometric data storehouse, set in first server
Portrait feature extraction program module, portrait characteristic matching program module and the first transport module are equipped with, is set in second server
Garment feature extraction procedure module, garment feature matcher module and the second transport module are equipped with, is set in the 3rd server
It is equipped with analysis program module, calculation procedure module, collator module and recommending module；The picture acquisition module is used to obtain
Picture, after picture is got, the Identification of Images module carries out Identification of Images processing, the clothes to the picture got
Identification module carries out clothes identifying processing to the picture of acquisition, and the portrait feature output module is used for the Identification of Images mould
Figure information after block processing is transmitted to the first server, in the first server, the portrait feature extraction journey
Sequence module is used to extract portrait characteristic, and the portrait biometric data storehouse is used to store portrait characteristic, the people
As characteristic matching program module is used to match the portrait characteristic and match portrait by first transport module to tie
To the 3rd server, the garment feature output module is used for the clothes after the clothes identification module processing fruit data transfer
Information is transmitted to the second server, and in second server, the garment feature extraction procedure module is used to extract clothes
Characteristic, the garment feature identification database are used to store garment feature data, the garment feature matcher module
For matching the garment feature data and clothing matching result data being transmitted to the 3rd clothes by second transport module
Business device；In the 3rd server, the analysis program module is used for according to portrait matching result data and clothes matching result number
According to data control is carried out, the calculation procedure module is used to calculate data control ratio, and the collator module is used for institute
State data control ratio to be ranked up, the recommending module is used to clothes recommendation information be presented to user.
Alternatively, including first artificial intelligence program's module, the first artificial intelligence program module are arranged on the first clothes
It is engaged in device, for intelligence learning portrait feature recognition.
Alternatively, including second artificial intelligence program's module, the second artificial intelligence program module are arranged on two services
In device, identified for intelligence learning garment feature.
Alternatively, including third party's work intelligent program module, third party's work intelligent program module are arranged on three services
In device, compare for intelligence learning data, and recommend for intelligence learning.
Alternatively, including trend clothing information database, for storing trend clothing information data.
Alternatively, including storage program module, picture feature data are stored in the way of four-tuple, for preserving depth
Spend data when network carries out forward process, and for be stored in it is backward seek gradient procedure when gradient data.
Alternatively, including the first artificial intelligence program module, second artificial intelligence program's module and third party's work intelligence
Energy program module, the first artificial intelligence program module are arranged in first server, are known for intelligence learning portrait feature
Not, the second artificial intelligence program module is arranged in two servers, is identified for intelligence learning garment feature, the described 3rd
Artificial intelligence program's module is arranged in three servers, is compareed for intelligence learning data, and is recommended for intelligence learning.
Alternatively, the Identification of Images module include head feature recognizer module, trunk feature recognition program module,
Stature feature recognition program module, the head feature recognizer module are used to identify the portrait head feature on picture, institute
State trunk feature recognition program module to be used to identify the portrait trunk feature on picture, the stature feature recognition program module is used
Portrait stature feature on identification picture.
Alternatively, the clothes identification module includes clothes fashion recognizer module, model recognizer module
With clothes dress material recognizer module, the clothes fashion recognizer module is used to identify the clothes fashion on picture, described
Model recognizer module is used to identify the clothes fashion on picture, and the clothes dress material recognizer module is used to identify
Clothes dress material on picture.
Alternatively, the portrait matching result data include age data, skin color data, gender data and expression number
According to；The clothing matching result data includes trend clothing information data.
The present invention provides a kind of clothing matching method based on image recognition, and the people in image is judged by image recognition
Body, cliping and pasting extraction head, trunk, the data such as stature and pass recognition of face server back by image, into face recognition, is returned
Return age, skin color, gender, expression data etc..Can also by image clip and paste interception clothing section extract style, pattern,
The data such as dress material return to the contrast generation big data analysis that dress material server carries out the conditions such as same pattern, pattern.
The present invention combines recognition of face server and the analysis data of dress material server pass identification server back, is taken in identification
Be engaged in device in formed gender, stature, mood, cloth, clothes fashion, etc. data control, the ratio row of being formed from high to low can be compareed
Row, the artificial intelligence by identifying server match fashion trend database, by rational Algorithm Analysis, are formed to clothes
Matching is recommended.
Alternatively, the present invention can be marked first manually, after mark part sample manually, be carried out certainly by deep learning
Dynamic mark more multisample, further optimizes feature vector, so as to fulfill the purpose of iterative learning.Learn skill by artificial intelligence
Art, the service of recognition of face server and dress material server as big data analysis, is formed by neuroid recognizer
Self study, has been put into 5,000,000 face samples in recognition of face server and has been trained, and 60 data are taken to face
Region is extracted, forms the algorithm identification of expression, age, gender.Dress material pattern style server is by the analysis to picture
Color, comparison diagram, decorative pattern, local feature are extracted, form the deep learning control training of 5,000,000 parts of clothes.
Of the invention to be with conventional clothes commending system difference, we are to be based on image recognition technology, with human body
And worn clothes are identified jointly.Human body, clothes have been carried out labeling by us, are rolled up by the depth under GPU parallel architectures
Product neutral net, Deep Learning scheduling algorithms, the structure of various convolutional neural networks is defined according to frame, is designed
Rational algorithm, during training pattern, we store picture feature data in the way of four-tuple, for preserving depth
Data when network carries out forward process are spent, is also used for being stored in gradient data when seeking gradient procedure backward, generates lmdb
Data set, builds CNN networks.By model training, we form the deep learning control training and 5,000,000 of 5,000,000 clothes
The deep learning control training of human body.When picture clothes identify during, the garment feature of acquisition is also passed through into nerve net
Network, to clothes identification storehouse expand and unmanned supervised learning, the more perfect identification model of formation and identification storehouse.
Remaining technical characteristic in the present embodiment, those skilled in the art can flexibly be selected according to actual conditions
With with to meet different specific actual demands.It will be apparent, however, to one skilled in the art that：It need not use
These specific details carry out the present invention.In other instances, in order to avoid obscuring the present invention, known calculation is not specifically described
Method, method or system etc., limit within technical protection scope in the claimed technical solution of claims of the present invention,
Details are not described herein again.
In the description of the present invention, unless otherwise clearly defined and limited, term " setting ", " installation ", " connected ",
" connection " is broader sense, and those skilled in the art should broadly understood.For example, it may be it is fixedly connected or lives
Dynamic connection, or be entirely connected, or partly connect, can mechanically connect or be electrically connected, can be direct phase
It is indirectly connected with even or by intermediary, connection inside two elements etc. is can also be, for the technology of this area
For personnel, can understand the concrete meaning of above-mentioned term in the present invention as the case may be, i.e. the expression of word language with
The implementation of actual techniques can be corresponded to flexibly, and the expression of the word language (including attached drawing) of specification of the invention is not formed to power
Any restrictions that profit requires are explained.
One of ordinary skill in the art will appreciate that realizing all or part of flow in the method in above-described embodiment, it is
Relevant hardware can be instructed to complete by computer program, the program can be stored in computer-readable storage and be situated between
In matter, the program is upon execution, it may include such as the flow of the embodiment of above-mentioned each method.Wherein, the storage medium can be
Magnetic disc, CD, ROM, RAM etc..For foregoing each embodiment of the method, in order to be briefly described, therefore it is all expressed as to a system
The combination of actions of row, but those skilled in the art should know, the present invention and from the limitation of described sequence of movement, because
For according to the present invention, certain some step can use other orders or be carried out at the same time.Secondly, those skilled in the art also should
Know, embodiment described in this description belongs to preferred embodiment, involved action and unit not necessarily this hair
Necessary to bright.
Those skilled in the art combine each exemplary unit and algorithm steps of the embodiments described herein description, energy
It is enough to be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually with hardware or software
Mode performs, application-specific and design constraint depending on technical solution.Professional technician can be to each specific
Application realize described function using distinct methods, but this realization should not exceed the scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed system, module and method, can pass through
Other modes are realized.For example, device embodiment described above is only illustrative, for example, the division of the module, can
To be only a kind of division of logic function, there can be other dividing mode when actually realizing, such as multiple module or components can be with
With reference to or be desirably integrated into another system, or some features can be ignored, or not perform.It is another, it is shown or discussed
Mutual coupling, direct-coupling or communication connection it may be said that by some interfaces, the INDIRECT COUPLING of device or unit or
Communication connection, can be electrical, machinery or other forms.The unit that the discrete parts illustrates can be or can not also
Receive physically separate, can be as the component that unit is shown or can not receive physical location, you can with positioned at a ground
Side, or can also be distributed in multiple network unit, some or all of list therein can be selected according to the actual needs
Member realizes the purpose of the present embodiment.In addition, each functional unit in each embodiment of the present invention can be integrated at one
Reason unit in or unit be individually physically present, can also two or more units be integrated in a unit
In.
If the function is realized in the form of SFU software functional unit and is used as independent production marketing or in use, can be with
It is stored in a computer-readable recording medium.Based on such understanding, technical scheme is substantially right in other words
The part or the part of the technical solution that the prior art contributes can be embodied in the form of software product, the calculating
Machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be personal
Computer, server, or network equipment etc.) perform all or part of step of each embodiment the method for the present invention.And
Foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (Read-Only Memory, ROM), random access memory
Device (Random Access Memory, RAM), magnetic disc or CD etc. are various can be with the medium of store program codes.
The above is only the preferred embodiment of the present invention, it should be understood that the present invention is not limited to described herein
Form, is not to be taken as the exclusion to other embodiment, and can be used for various other combinations, modification and environment, and can be at this
In the text contemplated scope, it is modified by the technology or knowledge of above-mentioned teaching or association area.And those skilled in the art institute into
Capable modifications and changes do not depart from the spirit and scope of the present invention, then all should be in the protection domain of appended claims of the present invention
It is interior.
Claims (10)
1. a kind of clothing matching commending system based on image recognition and big data analysis, it is characterised in that obtained including picture
Module, Identification of Images module, clothes identification module, portrait feature output module, garment feature output module, first server,
Second server, the 3rd server, portrait biometric data storehouse and clothes biometric data storehouse, set in first server
Portrait feature extraction program module, portrait characteristic matching program module and the first transport module are equipped with, is set in second server
Garment feature extraction procedure module, garment feature matcher module and the second transport module are equipped with, is set in the 3rd server
It is equipped with analysis program module, calculation procedure module, collator module and recommending module；
The picture acquisition module is used to obtain picture, and after picture is got, the Identification of Images module is to getting
Picture carries out Identification of Images processing, and the clothes identification module carries out clothes identifying processing to the picture of acquisition, and the portrait is special
Sign output module is used to the figure information after the Identification of Images resume module being transmitted to the first server, described the
In one server, the portrait feature extraction program module is used to extract portrait characteristic, the portrait biometric data
Storehouse is used to store portrait characteristic, and the portrait characteristic matching program module is used to match the portrait characteristic and pass through
For first transport module by portrait matching result data transfer to the 3rd server, the garment feature output module is used for will
Clothing information after the clothes identification module processing is transmitted to the second server, in second server, the clothes
Feature extraction program module is used to extract garment feature data, and the garment feature identification database is used to store garment feature number
According to the garment feature matcher module is used to match the garment feature data and will be taken by second transport module
Matching result data transfer is filled to the 3rd server；In the 3rd server, the analysis program module is used for according to portrait
Data control is carried out with result data and clothes matching result data, the calculation procedure module is used to calculate data contrast ratio
Rate, the collator module are used to data control ratio being ranked up, and the recommending module is used to present to user
Clothes recommendation information.
2. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is, including first artificial intelligence program's module, and the first artificial intelligence program module is arranged in first server, uses
In intelligence learning portrait feature recognition.
3. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is, including second artificial intelligence program's module, and the second artificial intelligence program module is arranged in two servers, is used for
Intelligence learning garment feature identifies.
4. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is, including third party's work intelligent program module, and third party's work intelligent program module is arranged in three servers, is used for
Intelligence learning data compare, and recommend for intelligence learning.
5. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is, including trend clothing information database, for storing trend clothing information data.
6. system is recommended based on the clothing matching of image recognition and big data analysis according to claim 2-4 any one of them is a kind of
System, it is characterised in that including storage program module, picture feature data are stored in the way of four-tuple, for preserving depth
Spend data when network carries out forward process, and for be stored in it is backward seek gradient procedure when gradient data.
7. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is, including the first artificial intelligence program module, second artificial intelligence program's module and third party's work intelligent program mould
Block, the first artificial intelligence program module are arranged in first server, for intelligence learning portrait feature recognition, described
Two artificial intelligence program's modules are arranged in two servers, are identified for intelligence learning garment feature, third party's work intelligence
Program module is arranged in three servers, is compareed for intelligence learning data, and is recommended for intelligence learning.
8. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is that the Identification of Images module includes head feature recognizer module, trunk feature recognition program module, stature feature
Recognizer module, the head feature recognizer module are used to identify the portrait head feature on picture, and the trunk is special
Sign recognizer module is used to identify the portrait trunk feature on picture, and the stature feature recognition program module, which is used to identify, to be schemed
The portrait stature feature of on piece.
9. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is that the clothes identification module includes clothes fashion recognizer module, model recognizer module and clothes clothing
Expect recognizer module, the clothes fashion recognizer module is used to identify the clothes fashion on picture, the model
Recognizer module is used to identify the clothes fashion on picture, and the clothes dress material recognizer module is used to identify on picture
Clothes dress material.
10. a kind of clothing matching commending system based on image recognition and big data analysis according to claim 1, it is special
Sign is that the portrait matching result data include age data, skin color data, gender data and expression data；It is described
Clothing matching result data includes trend clothing information data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711437438.6A CN108022161A (en) | 2017-12-26 | 2017-12-26 | Clothing matching commending system based on image recognition and big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711437438.6A CN108022161A (en) | 2017-12-26 | 2017-12-26 | Clothing matching commending system based on image recognition and big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108022161A true CN108022161A (en) | 2018-05-11 |
Family
ID=62070975
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711437438.6A Pending CN108022161A (en) | 2017-12-26 | 2017-12-26 | Clothing matching commending system based on image recognition and big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108022161A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108960499A (en) * | 2018-06-27 | 2018-12-07 | 东华大学 | A kind of Fashion trend predicting system merging vision and non-vision feature |
| CN109124206A (en) * | 2018-08-15 | 2019-01-04 | 河北中晟易通科技有限公司 | The big data analysis processing system of clothes recommendation is carried out based on intelligent interaction fitting mirror |
| WO2019237721A1 (en) * | 2018-06-14 | 2019-12-19 | 深圳码隆科技有限公司 | Garment dimension data identification method and device, and user terminal |
| CN111008626A (en) * | 2018-10-04 | 2020-04-14 | 斯特拉德视觉公司 | Method and device for detecting object based on R-CNN |
| CN111142671A (en) * | 2019-12-27 | 2020-05-12 | 江西服装学院 | Intelligent garment with intelligent interaction system |
| CN111144850A (en) * | 2019-12-30 | 2020-05-12 | 江西服装学院 | Intelligent clothing data analysis and management system |
| WO2020119311A1 (en) * | 2018-12-14 | 2020-06-18 | 深圳市商汤科技有限公司 | Neural network training method and image matching method and device |
| CN117648493A (en) * | 2023-12-13 | 2024-03-05 | 南京航空航天大学 | Cross-domain recommendation method based on graph learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096335A (en) * | 2015-09-17 | 2015-11-25 | 无锡天脉聚源传媒科技有限公司 | Evaluation information transmission method and device |
| US20160189431A1 (en) * | 2014-12-25 | 2016-06-30 | Kabushiki Kaisha Toshiba | Virtual try-on system, virtual try-on terminal, virtual try-on method, and computer program product |
| CN105760999A (en) * | 2016-02-17 | 2016-07-13 | 中山大学 | Method and system for clothes recommendation and management |
| CN106952142A (en) * | 2017-03-17 | 2017-07-14 | 合肥龙图腾信息技术有限公司 | Clothing matching intelligent recommendation systems, devices and methods |
| CN106951448A (en) * | 2017-02-20 | 2017-07-14 | 竹间智能科技（上海）有限公司 | A kind of personalization, which is worn, takes recommendation method and system |
| US20170277365A1 (en) * | 2016-03-28 | 2017-09-28 | Intel Corporation | Control system for user apparel selection |
| CN107248099A (en) * | 2017-06-01 | 2017-10-13 | 深圳天珑无线科技有限公司 | Clothing matching recommends method, system and storage medium |
-
2017
- 2017-12-26 CN CN201711437438.6A patent/CN108022161A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160189431A1 (en) * | 2014-12-25 | 2016-06-30 | Kabushiki Kaisha Toshiba | Virtual try-on system, virtual try-on terminal, virtual try-on method, and computer program product |
| CN105096335A (en) * | 2015-09-17 | 2015-11-25 | 无锡天脉聚源传媒科技有限公司 | Evaluation information transmission method and device |
| CN105760999A (en) * | 2016-02-17 | 2016-07-13 | 中山大学 | Method and system for clothes recommendation and management |
| US20170277365A1 (en) * | 2016-03-28 | 2017-09-28 | Intel Corporation | Control system for user apparel selection |
| CN106951448A (en) * | 2017-02-20 | 2017-07-14 | 竹间智能科技（上海）有限公司 | A kind of personalization, which is worn, takes recommendation method and system |
| CN106952142A (en) * | 2017-03-17 | 2017-07-14 | 合肥龙图腾信息技术有限公司 | Clothing matching intelligent recommendation systems, devices and methods |
| CN107248099A (en) * | 2017-06-01 | 2017-10-13 | 深圳天珑无线科技有限公司 | Clothing matching recommends method, system and storage medium |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019237721A1 (en) * | 2018-06-14 | 2019-12-19 | 深圳码隆科技有限公司 | Garment dimension data identification method and device, and user terminal |
| CN108960499A (en) * | 2018-06-27 | 2018-12-07 | 东华大学 | A kind of Fashion trend predicting system merging vision and non-vision feature |
| CN108960499B (en) * | 2018-06-27 | 2021-11-23 | 东华大学 | Garment fashion trend prediction system integrating visual and non-visual features |
| CN109124206A (en) * | 2018-08-15 | 2019-01-04 | 河北中晟易通科技有限公司 | The big data analysis processing system of clothes recommendation is carried out based on intelligent interaction fitting mirror |
| CN111008626A (en) * | 2018-10-04 | 2020-04-14 | 斯特拉德视觉公司 | Method and device for detecting object based on R-CNN |
| CN111008626B (en) * | 2018-10-04 | 2023-10-13 | 斯特拉德视觉公司 | Method and device for detecting object based on R-CNN |
| WO2020119311A1 (en) * | 2018-12-14 | 2020-06-18 | 深圳市商汤科技有限公司 | Neural network training method and image matching method and device |
| TWI760650B (en) * | 2018-12-14 | 2022-04-11 | 大陸商深圳市商湯科技有限公司 | Method and image matching method for neural network training and device thereof |
| CN111142671A (en) * | 2019-12-27 | 2020-05-12 | 江西服装学院 | Intelligent garment with intelligent interaction system |
| CN111144850A (en) * | 2019-12-30 | 2020-05-12 | 江西服装学院 | Intelligent clothing data analysis and management system |
| CN117648493A (en) * | 2023-12-13 | 2024-03-05 | 南京航空航天大学 | Cross-domain recommendation method based on graph learning |
| CN117648493B (en) * | 2023-12-13 | 2024-05-31 | 南京航空航天大学 | Cross-domain recommendation method based on graph learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108022161A (en) | Clothing matching commending system based on image recognition and big data analysis | |
| Luce | Artificial intelligence for fashion: How AI is revolutionizing the fashion industry | |
| Liu et al. | Hi, magic closet, tell me what to wear! | |
| US10346893B1 (en) | Virtual dressing room | |
| US20200005087A1 (en) | Cognitive automated and interactive personalized fashion designing using cognitive fashion scores and cognitive analysis of fashion trends and data | |
| Hsiao et al. | ViBE: Dressing for diverse body shapes | |
| Guan et al. | Apparel-based deep learning system design for apparel style recommendation | |
| Zhao et al. | Neo-fashion: A data-driven fashion trend forecasting system using catwalk analysis | |
| US11475500B2 (en) | Device and method for item recommendation based on visual elements | |
| KR102043440B1 (en) | Method and system for coordination searching based on coordination of a plurality of objects in image | |
| US11842378B2 (en) | Systems and methods for product identification using image analysis and trained neural network | |
| Geng et al. | One of a kind: User profiling by social curation | |
| CN107729380A (en) | Clothing matching method, terminal, terminal | |
| CN108932648A (en) | A kind of method and apparatus for predicting its model of item property data and training | |
| US20200372560A1 (en) | Method for exploring and recommending matching products across categories | |
| Song et al. | When multimedia meets fashion | |
| CN109325529A (en) | Sketch identification method and application of sketch identification method in commodity retrieval | |
| Wang et al. | Learning compatibility knowledge for outfit recommendation with complementary clothing matching | |
| Liu et al. | Person image generation with attention-based injection network | |
| CN108596646B (en) | Garment matching recommendation method integrating face attribute analysis | |
| CN113988978A (en) | Garment matching recommendation method, system and equipment based on knowledge graph | |
| CN106022860A (en) | Matching method and apparatus | |
| CN107423441A (en) | Picture association method and device and electronic equipment | |
| Liu | [Retracted] Art Painting Image Classification Based on Neural Network | |
| Angelov et al. | Advances in computational intelligence systems |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180511 |
|
| WD01 | Invention patent application deemed withdrawn after publication |