CN109284666B - A kind of dress ornament recommended method based on image recognition and big data - Google Patents
A kind of dress ornament recommended method based on image recognition and big data Download PDFInfo
- Publication number
- CN109284666B CN109284666B CN201810822167.4A CN201810822167A CN109284666B CN 109284666 B CN109284666 B CN 109284666B CN 201810822167 A CN201810822167 A CN 201810822167A CN 109284666 B CN109284666 B CN 109284666B
- Authority
- CN
- China
- Prior art keywords
- dress ornament
- face
- image
- data
- recommended
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- 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
- 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
- G06V40/168—Feature extraction; Face representation
-
- 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)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention will be inputted by the acquisition of objective data and subjective artificial demand, provide a kind of perfect intelligent dress ornament recommended method；The present invention provides a kind of image-recognizing method for human body external feature and common dress ornament for recommending field suitable for intelligent dress ornament, a kind of acquisition of dress ornament recommending data, analysis, processing method, the specific service and implementation method of a kind of synthetic image identification and big data based on the above method；The present invention has good beneficial effect, can build bridge for both user and businessman, is also able to use family and obtains more satisfied dress ornament recommendation results to the full extent.
Description
Technical field
The present invention relates to the intelligent recommendation fields of dress ornament, and in particular to a kind of about human body barment tag and clothes to be recommended
The image-recognizing method for adoring feature is further related to a kind of recommended method acquired based on big data with analysis, is mentioned based on this present invention
For a kind of specific recommended method of dress ornament of synthetic image identification and big data.
Background technique
With the improvement of the quality of life, very more people has been not content with single dress ornament demand, but pursues more
Meet the dress ornament of oneself demand and feature, however many times need oneself it is artificial in the dress ornament of magnanimity select because not
Be profession clothing matching teacher, so from demand with aesthetically for cannot select to the greatest extent to meet trip requirements and
The best dress ornament of colour match finds improper, is more artificially from magnanimity to cause the waste of resource after leading to purchase
It is selected in dress ornament, this way not only wastes time but also do not meet the trend that science and technology instantly changes life.
To solve above content, the prior art provides associated solutions:
Such as notification number is the Chinese invention patent of 104484450 A of CN, discloses that " clothing matching based on image pushes away
Recommend method and clothing matching recommendation apparatus ", technical solution are as follows: first obtain first group of dress ornament image, identify the master of the dress ornament
Then tone or domain color match corresponding and appropriate with first group of dress ornament image domain color from the database that it had previously been established
Second group of dress ornament image, to realize the optimal collocation of dress ornament.This technical solution suffers from the drawback that first is that needing first to select one
Group dress ornament image, then by the method for identification image, so that clothing matching is carried out in established dress ornament database, it is this to do
Method is relatively complicated, and is suitble to the first group of dress ornament also bad determination of oneself；Second is that the invention only for dress ornament color this
Principal element is arranged in pairs or groups and is recommended, and seems too simple from the point of view of in content and service, is unsatisfactory for people and takes to existing dress ornament
With the demand with recommendation.
Such as notification number is the Chinese invention patent of 106952142 A of CN, discloses " clothing matching intelligent recommendation system
System, device and method ", core technology scheme are as follows: obtain the human body image of wearer by camera shooting, and pass through identifying processing list
Member obtains characteristics of human body's parameter, and recommends several clothing matching samples from clothing matching sample database by characteristics of human body's parameter
This, and generate expected dress ornament image.This technical solution suffers from the drawback that adopts first is that carrying out image only for characteristics of human body
Collection, identification, processing do not carry out the Image Acquisition of dress ornament, identification, processing to the dress ornament that can recommend collocation, therefore the way is excessive
Deviation characteristics of human body's parameter, cannot adequately to dress ornament carry out information collection and analysis, so cannot more expand dress ornament
Dress ornament sample size in sample database of arranging in pairs or groups, causes that dress ornament quantity can be recommended few；Second is that in order to be directed to user individual dress ornament demand
And the information acquired, it is limited only to human body barment tag, it is clear that, it is single according to human body for dress ornament recommendation
Barment tag is recommended, and recommendation results cannot utmostly suit user's real demand.
It can find, the research field nowadays recommended dress ornament is mainly concentrated in be pushed away using collaborative filtering
Technology is recommended, the behavioral data of user is analyzed and processed, to carry out dress ornament recommendation.
To sum up, for more comprehensive and accurate acquisition, analysis, processing dress ornament and user information, and more improve ground
Implement dress ornament recommendation, it is desirable to provide a kind of more optimal dress ornament suggested design.
Summary of the invention
The technical problem to be solved by the present invention is
How by human body external feature, garment ornament to be recommended, big data acquisition and analysis method, thus comprehensively and
Accurately acquisition, analysis, processing dress ornament, user and Business Information, and more perfect implement dress ornament recommendation.
The present invention will be inputted by the acquisition of objective data and subjective artificial demand, provide a kind of perfect intelligent dress ornament
Recommended method.
The present invention provides a kind of based on human body external feature image-recognizing method, the image-recognizing method of common dress ornament, clothes
The dress ornament for adoring the acquisition of recommending data, analysis, the synthetic image identification of processing method and big data is recommended specific service and is realized
Method.
Elder generation's employment of the present invention recommends related human body extrinsic information in characteristic image recognition methods acquisition and dress ornament in vitro, then
Dress ornament information is acquired by common dress ornament image-recognizing method, then acquires relevant dress ornament recommending data information and it is divided
Analysis and processing, finally in summary method carries out last dress ornament recommendation process.
Human body external feature information image recognition methods of the present invention, comprising:
The identification of human body face characteristic image；The identification of Human Height characteristic image.
Specifically, human body face characteristic image identifies, comprising:
Facial image is obtained by camera；The main colour of skin of preliminary judgement face and aging degree；Establish facial characteristics to
Amount；It is compared and analyzes in image surface library；Obtain human body face characteristic results.
It is optionally, described that facial image is obtained by camera, comprising:
First time shooting is carried out, face resolution requirement in recognition of face is such as met, terminates shooting, determines facial image,
Otherwise taken the photograph image is discarded, and prompts to be shot next time, is just terminated until meeting face resolution requirement in recognition of face
Prompt；
First time shooting is carried out, such as meets facial angle requirement in recognition of face, terminates shooting, determine facial image, it is no
Taken the photograph image is then discarded, and prompts to be shot next time, requires just to terminate to prompt until meeting facial angle in recognition of face；
First time shooting is carried out, face context request in recognition of face is such as met, terminates shooting, determines facial image, it is no
Taken the photograph image is then discarded, and prompts to be shot next time, just terminates to prompt until meeting face context request in recognition of face；
First time shooting is carried out, human face light requirement in recognition of face is such as met, terminates shooting, determines facial image, it is no
Taken the photograph image is then discarded, and prompts to be shot next time, just terminates to prompt until meeting human face light requirement in recognition of face.
Preferably, the face resolution requirement is 36 × 27(ppi)；
Preferably, the facial angle requirement is that shoulder gradient section is -15 ° to+15 ° to the left and right on head, and head is bowed
Facing upward angular interval is -20 ° to+20 °, and it is -10 ° to+10 ° that head, which is swung left and right and rotates angular interval,.
It should illustrate, the face context request is without excessively sundries information related with face, the human face light
It, also cannot there is a serious shortage of brightness, both of the above explanation can be adjusted according to the actual situation it is required that for without excessive strong light filling
With decision.
Optionally, the main colour of skin of the preliminary judgement face and aging degree, comprising:
Establish colour of skin sample database and aging degree sample database；Face's sample point is established, data are acquired；Carry out face texture with
Color analysis；It compares established sample database and obtains judgement result.
It is optionally, described to establish facial characteristics vector, comprising:
Locating human face's image major organs；With structure attribute feature representation face major organs；Using can feature identification
Model carries out information processing；Determine substantially shape of face.
Preferably, locating human face's image major organs are eyes, nose, eyebrow, ear, mouth, face's outer profile,
Chin, forehead.
Preferably, described to be characterized between each organ with structure attribute in structure attribute feature representation face major organs
Distance, angle, position, size, shape.
Preferably, it is described using can feature identification model carry out information processing in can feature identification model be based on sky
Between rectangular coordinate system and the polynary three-dimensional vector established.
It is optionally, described to be compared and analyze in image surface library, comprising:
Common shape of face sample is compared and is analyzed；Common face complexion sample is compared and is analyzed；To difference
The skin quality sample of age bracket is compared and analyzes.
Preferably, the common shape of face sample can be 1. round shapes of face；2. oval shape of face；3. oval shape of face；4.
The shape of face of falling oval；5. rectangular shape of face；6. rectangle shape of face；7. trapezoidal shape of face；8. inverted trapezoidal shape of face；9. diamond shape shape of face；10. five
Angular shape of face;It should be noted that can all be expressed together in those skilled in the art by the proprietary term of other shapes of face
The sample meaning.
Preferably, the common colour of skin can be the 1. neutral colours of skin；2. the cold shallow colour of skin；3. the cold depth colour of skin；4. warming up the shallow colour of skin；
5 warm up the gorgeous colour of skin；6. the warm depth colour of skin；It should be noted that passing through the dedicated art of other colours of skin in those skilled in the art
Language can all express the same meaning.
Preferably, the skin quality of the common age bracket can be 1. early childhoods；2. Development in Puberty period；3. when young
Phase；4. the middle age；5. senectitude；It should be noted that the professional technician in the art passes through other skin quality
Proprietary term can all express the same meaning.
It is optionally, described to obtain facial characteristics result, comprising:
Obtain basic shape of face feature；Obtain basic colour of skin result；User is tentatively obtained by face skin quality age characteristics
Age；
Specifically, Human Height characteristic image identifies, comprising:
Whole Body image is obtained by camera；Height identification is carried out using existing height identification technology；Obtain human body
Height characteristic results.
It is optionally, described that Whole Body image is obtained by camera, comprising:
First time shooting is carried out, height identification technology intermediate-resolution requirement is such as met, terminates shooting, determines Whole Body figure
Otherwise picture discards taken the photograph image, and prompts to be shot next time, just tie until meeting the requirement of height identification technology intermediate-resolution
Beam prompt；
First time shooting is carried out, angle requirement in height identification technology is such as met, terminates shooting, determines Whole Body figure
Otherwise picture discards taken the photograph image, and prompts to be shot next time, just terminate until meeting angle requirement in height identification technology
Prompt；
First time shooting is carried out, context request in height identification technology is such as met, terminates shooting, determines Whole Body figure
Otherwise picture discards taken the photograph image, and prompts to be shot next time, just terminate until meeting context request in height identification technology
Prompt；
First time shooting is carried out, illumination requirement in height identification technology is such as met, terminates shooting, determines Whole Body figure
Otherwise picture discards taken the photograph image, and prompts to be shot next time, just terminate until meeting illumination requirement in height identification technology
Prompt.
It is optionally, described that height identification is carried out using existing height identification technology, comprising:
Height, which delimited, according to face characteristic identifies reference line；Height mapping relationship is established according to reference line；According to highly reflecting
It penetrates relationship and calculates current individual object height；A upper limit reference line and a lower limit reference line delimited according to same face characteristic
Two reference lines make the corresponding face characteristic of the human body to be identified between two reference lines, and record described two
The ordinate of reference line；It is vertical according to mapping relations of the corresponding face characteristic for the human body for being actually subjected to obtain between two reference lines
Coordinate obtains the height of the human body；
It is optionally, described to obtain Human Height characteristic results, comprising:
It obtains height values and classifies, a kind of height values range is less than 160cm, and two class height values are more than or equal to
160cm is simultaneously less than or equal to 170cm, and three classes height values are greater than 170cm and are less than 175cm, and four class height values are greater than and are equal to
175cm is less than 180cm, and five class height values are greater than and are equal to 180cm.
The image-recognizing method of common dress ornament of the present invention, comprising:
Determine whether dress ornament can use bar code recognition；Dress ornament bar code recognition；Body color identification and its colour system are sentenced
It is fixed；The identification of dress ornament style.
Specifically, determining whether dress ornament can use bar code recognition, comprising:
User observes whether the dress ornament contains bar code relevant to the application, if any associated bar code, then
Dress ornament barcode recognition method is taken to omit body color identification and its colour system judgement, dress ornament style two steps of identification, directly
It summarizes classification；
User observes whether the dress ornament contains bar code relevant to the application, such as without associated bar code, then
It takes body color to identify and its colour system determines, dress ornament style identifies two steps, omit dress ornament barcode recognition method, directly
It summarizes classification.
Specifically, dress ornament bar code recognition, comprising:
Establish dress ornament bar code data library；Generate dress ornament bar code data；Bar code is scanned, obtains bar code data, and
Correspond to the corresponding dress ornament information in dress ornament bar code data library.
It is optionally, described to establish dress ornament bar code data library, comprising:
Each bar code corresponds to corresponding dress ornament information, and style is contained in dress ornament information, wears occasion, yardage, the place of production
Deng, under the premise of without prejudice to present invention thought information professional and technical personnel in the field relevant to dress ornament can show and
It is understood and appreciated that with being clear to, this will not be repeated here.
Specifically, body color identification and its colour system determine, comprising:
The body color of dress ornament is determined by image recognition；Colour system judgement is carried out to body color after determining body color；
Finally sort out from cool colour, warm colour, intermediate colour system three's one of them.
Specifically, dress ornament style identifies, comprising:
Establish dress ornament Database of Garment；Common dress ornament style image information and affiliated clothes are inputted in dress ornament Database of Garment
Adorn style；It identifies dress ornament image, extracts pertinent image information and dress ornament Database of Garment is matched；As dress ornament Database of Garment can match
To success, then corresponding dress ornament style is exported, is not otherwise exported.
A kind of acquisition in relation to dress ornament recommending data of the present invention, analysis, processing method, comprising:
Artificially it is actively entered data needed for dress ornament is recommended；The public information of access internet passively inputs dress ornament and recommends institute
Need data；The analysis and processing of data needed for dress ornament is recommended.
Specifically, being artificially actively entered data needed for dress ornament is recommended, comprising:
It is actively entered user's gender；It is actively entered user's age；It is actively entered the purchase dress ornament that user can bear
Cost range；It is actively entered the dress ornament style of user's hobby；It is actively entered the occasion that user will wear；Being actively entered makes
The style that user wants, finally it should be noted that those skilled in the art finds out artificial under the premise of not making the creative labor
Ground is actively entered dress ornament and required data content is recommended to belong to the invention thought content to be expressed.
Specifically, the public information of access internet passively inputs data needed for dress ornament is recommended, comprising:
Passive input user locality real-time weather situation；The local humane custom feature of geography of passive input user；
Disclosed social information in the social software of passive input user；The passive style information for inputting trend dress ornament instantly, should
What is illustrated is the public information quilt for the access internet that those skilled in the art finds out under the premise of not making the creative labor
Dynamic input dress ornament recommends required data content to belong to the invention thought content to be expressed.
Optionally, the passive input user locality real-time weather situation, comprising:
User locality real time temperature；User locality real-time humidity；User local wind direction and wind-force in real time；User
Local real-time air quality.
Specifically, dress ornament recommends the analysis and processing of required data, comprising:
Data needed for the dress ornament being actively entered is recommended recommend required data to integrate with the dress ornament passively inputted；Such as master
Data needed for the dress ornament of dynamic input is recommended recommend required data to have differences with the dress ornament passively inputted, and preferential selection is actively entered
Dress ornament recommend needed for data；By treated, information categorization feeds back to recommender system.
It is of the present invention that the dress ornament recommendation specific service and implementation method of a kind of synthetic image identification and big data are provided,
Include:
Determine human body barment tag result；Determine garment ornament result；It determines and is actively entered data knot needed for dress ornament is recommended
Fruit；Determine that passive input dress ornament recommends required data result；User behavior habituation mechanism；Individual subscriber is pushed away with businessman's dress ornament
Recommend the selection mechanism of required database；The operating mechanism of database needed for individual subscriber dress ornament is recommended；Needed for businessman's dress ornament is recommended
The operating mechanism of database；Dress ornament recommendation process unit carries out last recommendation results and presents.
Optionally, the user behavior habituation mechanism, comprising:
By each human body barment tag result, garment ornament result, it is actively entered data result needed for dress ornament is recommended, passive
Input dress ornament recommends required data result to be counted, and same or like result occurs more, recommends link can be in dress ornament
Preferentially stress correlated results, reduces unnecessary repetitive operation.
Optionally, the selection mechanism of database needed for the individual subscriber and businessman's dress ornament are recommended, comprising:
There are two kinds of dress ornaments for a user and recommends selection method, it, can if user needs to wear related dress ornament at once
To select individual subscriber dress ornament to recommend required database；If user needs to buy and not wear at once, quotient can choose
Recommending data library needed for family's dress ornament.
Optionally, the operating mechanism in recommending data library needed for the individual subscriber dress ornament, comprising:
User carries out dress ornament image recognition to the existing dress ornament of individual；Dress ornament image recognition result is inputted personal dress ornament to recommend
Required database；Database generalization user's human body barment tag result, individual subscriber garment ornament knot needed for personal dress ornament is recommended
Fruit is actively entered data result needed for dress ornament is recommended, passively inputs data result, user behavior habituation needed for dress ornament is recommended
Mechanism provides integrated data to processing unit.
Optionally, the operating mechanism of database needed for businessman's dress ornament is recommended, comprising:
Businessman will sell dress ornament and carry out dress ornament image recognition；Dress ornament image recognition result input businessman's dress ornament is recommended into institute
Need database；Library synthetic user human body barment tag result, businessman's garment ornament result, active needed for businessman's dress ornament recommending data
It inputs data result needed for dress ornament is recommended, passively input result, user behavior habituation mechanism needed for dress ornament is recommended to processing
Unit provides integrated data.
Optionally, the dress ornament recommendation process unit carries out last recommendation results presentation, comprising:
Determine the type of database needed for dress ornament is recommended；Data provided by database needed for recommending dress ornament are handled
And then it completes dress ornament pairing and recommends；Dress ornament recommendation results are presented to user
Preferably, it is described recommend dress ornament required data provided by database handle so that complete dress ornament pairing with
Recommend, comprising:
Suitable basic clothing type is determined by user's human body external feature, then by being actively entered needed for dress ornament recommendation
Data result further reduces specifically suitable dress ornament, recommends required data result to carry out into one referring again to passive input dress ornament
The dress ornament that step reduction is specifically suitble to finally recommends the dress ornament in required database to be screened and matched, really with existing dress ornament
Fixed consequently recommended dress ornament.
It is preferably, described that dress ornament recommendation results are presented to user, comprising:
It carries out first time dress ornament recommendation results to present, if met requirement of the user to recommendation dress ornament, terminates dress ornament and recommend,
It determines dress ornament recommendation results, otherwise discards dress ornament recommendation results, and prompt to carry out dress ornament next time and recommend, until meeting user couple
The requirement of recommended dress ornament just terminates to prompt.
Compared with prior art, the present invention has the following advantages:
1. it will be laid stress on merely among the collocation of dress ornament and dress ornament during carrying out dress ornament recommendation, it is further excellent
It turns to and implements specific matching between the different characteristics between dress ornament and user；
2. establishing database needed for dress ornament is recommended during dress ornament is recommended, database needed for which recommends is used
Completely new theory, the dress ornament for being not merely confined to businessman recommend user, and individual subscriber also can establish to one's name
Dress ornament recommend needed for database, thus the perfect personal experience of user；
3. being not limited solely to the recognition methods of the body color of dress ornament in the step for dress ornament identifies, and increase
The recognition methods of dress ornament style has also introduced bar code identification technology, in the conceived case may be used as more optimal process
Preferentially to select bar code recognition dress ornament, the realization difficulty of entire dress ornament image recognition is reduced；
4. data fields needed for recommending in dress ornament are actively entered mode and passive input mode using completely new, both increased
The reliability and acquisition convenience of dress ornament recommending data, also utmostly from the subjective particular demands for meeting user.
Detailed description of the invention
Fig. 1 is the dress ornament recommended flowsheet figure of the application；Fig. 2 is the acquisition human body external feature information Step figure of the application；
Fig. 3 is that the application acquires garment ornament information Step figure；Fig. 4 is that the application acquires data step figure needed for dress ornament is recommended.
Specific embodiment
In order to make the invention's purpose, features and advantages of the invention more obvious and easy to understand, below in conjunction with the present invention
The technical method in the embodiment of the present invention is clearly and completely described in attached drawing in embodiment, but described
Embodiment be only a portion embodiment of the present invention, and not all embodiments.The application can be to be much different from this
The other way of description is implemented, and those skilled in the art can do similar popularization without prejudice to the application intension,
Therefore the application is not limited by following public specific implementation.
The present invention will be inputted by the acquisition of objective data and subjective artificial demand, provide a kind of perfect intelligent dress ornament
Recommended method.
The present invention provides a kind of image-recognizing method for human body external feature and common dress ornament；Present invention simultaneously provides
A kind of acquisition of dress ornament recommending data, analysis, processing method；Present invention simultaneously provides a kind of identifications of synthetic image and big data
Specific service and implementation method.
Objectively data acquisition as described above refers to two parts, and a part is the acquisition of garment ornament information, another
Part is the acquisition of userspersonal information；Subjective artificial demand input as described above refers to two parts, and a part is basis
Related data content, another part are to select businessman's clothes according to their own needs with dress ornament recommendation for oneself feature and demand input
The operating mechanism in recommending data library needed for the operating mechanism in recommending data library needed for adoring and personal dress ornament.
Human body external feature image-recognizing method as described above, comprising: human body face characteristic image identification；Human Height
Characteristic image identification.
The image-recognizing method of common dress ornament as described above, comprising: determine whether dress ornament can use bar code recognition；
Dress ornament bar code recognition；Body color identification and its colour system determine；The identification of dress ornament style.
Acquisition, the analysis, processing method of dress ornament recommending data as described above, comprising: be artificially actively entered dress ornament and push away
Recommend required data；The public information of access internet passively inputs data needed for dress ornament is recommended；Point of data needed for dress ornament is recommended
Analysis and processing.
The specific service and implementation method of synthetic image identification and big data as described above, comprising: determine that human body appearance is special
Levy result；Determine garment ornament result；It determines and is actively entered data result needed for dress ornament is recommended；Determine that passive input dress ornament is recommended
Required data result；User behavior habituation mechanism；The selection mechanism of database needed for individual subscriber and businessman's dress ornament are recommended；
The operating mechanism of database needed for individual subscriber dress ornament is recommended；The operating mechanism of database needed for businessman's dress ornament is recommended；Dress ornament pushes away
It recommends processing unit and carries out last recommendation results presentation.
It should be noted that the identification of human body face characteristic image mentioned by summary of the invention, Human Height are special in the present invention
It is existing for levying the particular techniques such as image recognition, bar code identification technology, dress ornament color identification technology, dress ornament style identification technology
Technology is not emphasis of the invention, wherein described preference is only part reference example, those skilled in the art can root
It makes adjustment according to needs, but its technical thought realized should be a part of the invention.
It should be noted that the technical solution of part in order to solve the technical problem is certain prior arts in the present invention
Combination, functionally mutually support between each technical characteristic of the assembled scheme, new technical effect, and the combination can be obtained
Whether the right and wrong for person of ordinary skill in the field are it will be apparent that and complete specific to each individual technical characteristic
Or evaluation to the invention can not be influenced known to part, component invention should be met to creative requirement.
Embodiments herein is described in detail below.Please refer to Fig. 1, Fig. 2, Fig. 3, Fig. 4.Fig. 1 is of the invention
Described method includes following steps for overview flow chart:
Step S1, acquisition recommends required information to complete dress ornament.
In this step step by step comprising three, comprising: S11, acquisition human body external feature information, S12, acquisition garment ornament
Information, data needed for S13, acquisition dress ornament are recommended, first progress step S11 carries out step S12 again and finally carries out step S13.
Wherein, human body external feature information reference Fig. 2 step S11, is acquired, step S111 is first carried out and carries out step again
In S112, the step S111 human body face characteristic image identification include in have to the main colour of skin 6, shape of face 7, aging degree
The image recognition of 8 threes.
Wherein, step S12, acquire garment ornament information reference Fig. 3, first carry out step S121, to step S121 result into
Row judgement, "Yes" carries out step S122 as the result is shown, and "No" carries out step S123 as the result is shown, continues after completing step S123
Carry out step S124.
Wherein, step S13, acquisition dress ornament recommends required data reference Fig. 4, advanced step S131 to carry out step S132 again
Finally carry out step S133.
Step S2, it handles information collected and obtains dress ornament recommendation results.
This step includes that five contents are respectively: user behavior habituation mechanism 1, individual subscriber and businessman's dress ornament are recommended
Needed for the operating mechanism 3 of database needed for the selection mechanism 2 of required database, individual subscriber dress ornament are recommended, businessman's dress ornament are recommended
The operating mechanism 4 of database, dress ornament recommendation process unit carry out last recommendation results and are presented 5.
Step S3, user is confirmed whether to meet the requirements.
When in this step show " meeting the requirements " then terminate dress ornament recommend, retract step if " being unsatisfactory for requiring " if shown
Content before S1 repetition.
Technical solution of the content that above-mentioned steps and step are included in the specification summary of the invention is existing specifically
It is bright that therefore not to repeat here.
Recommended based on the dress ornament of image recognition and big data with a concrete application scene in the embodiment of the present invention below
Method is illustrated.In the embodiment of the present invention, the dress ornament recommended method based on image recognition and big data is applied in smart phone
On.
The camera that personal user opens smart phone shoots face, after meeting after face image identification requires
Stop shooting, it is assumed that the main colour of skin of personal user is to warm up the shallow colour of skin, and round shape of face, aging degree is nonage；Continue to use
Camera carries out the shooting of whole body, meets after height image recognition requires and stops shooting, it is assumed that personal user meets three classes height
Numerical requirements；
Personal user and businessman user can be carried out acquisition garment ornament information Step, and user and businessman can be by oneself
All clothes establish one's own dress ornament recommending data library with dress ornament image-recognizing method, and personal dress ornament information is stored in individual
Dress ornament recommending data library；Personal user takes out clothes, carries out judging no bar code to it, with the camera pair of smart phone
Dress ornament is identified, it is assumed that show that body color is that yellow belongs to warm colour, it is assumed that show that style is surplus；With cell phone software
Form artificially inputs data needed for dress ornament is recommended, it is assumed that input male assumes the age 20 years old, it is assumed that the dress ornament that can be born spends expense
It is 500 yuan or less with range, it is assumed that be friend's party by the occasion attended, access internet passively inputs information, it is assumed that uses
The linking Internet information at family shows that locating place is Guangdong Province Guangzhou, it is assumed that the period locating for locating ground be outdoor low temperature and
Have slight wind, these comprehensive information go out the result is that the men's clothing of young style should be worn, and the men's clothing meets weather demand
It is the autumn and winters such as long sleeves trousers dress；Personal user repeatedly carry out finding that same or like result occurs after above-mentioned steps it is more,
Recommend link that can preferentially stress correlated results in dress ornament, reduce unnecessary repetitive operation, individual subscriber is according to oneself need
Seek database needed for selecting different dress ornaments to recommend, it is assumed that user is not purchase clothes but selects from the dress ornament that oneself has possessed
It selects, then database needed for selecting individual subscriber dress ornament to recommend, it is assumed that user will buy clothes and select in the clothes of businessman, then
Database needed for selecting businessman individual's dress ornament to recommend；Then according to the selection of disparate databases, different databases is run；Assuming that
Businessman's customer data base is selected, according to warming up the shallow colour of skin, round shape of face, nonage, male, three classes height, the dress ornament flower that can be born
Expense cost range is 500 yuan or less, the occasion that will attend is that friend meets, the period locating for locating ground is outdoor low temperature and has
Slight wind, these comprehensive opinions, recommend out men's clothing warm colour surplus yardage tentatively meet three classes height, then user selects
It selects, satisfaction then terminates, and is unsatisfactory for continuing to repeat.
For the various method embodiments described above, it describes for simplicity, therefore, it is stated as a series of action combinations, but
Be those skilled in the art should understand that, the present invention is not limited by the sequence of acts described because according to the present invention, certain
A little steps can use other sequences or carry out simultaneously.Secondly, those skilled in the art should also know that, it is retouched in specification
The embodiment stated belongs to preferred embodiment, and related actions and modules might not all be necessary to the present invention.
Claims (5)
1. a kind of dress ornament recommended method based on image recognition and big data characterized by comprising
It is obtained by human body external feature image-recognizing method and recommends related human body extrinsic information with dress ornament, then pass through common clothes
It adorns image-recognizing method and acquires dress ornament information, then use a kind of acquisition in relation to dress ornament recommending data, analysis, processing method, most
Afterwards in summary a kind of acquisition in relation to dress ornament recommending data, analysis, processing method take a kind of synthetic image identify and
The dress ornament of big data recommends specific service and implementation method；
It is described by include: in human body external feature image-recognizing method human body face characteristic image identification；Human Height feature
Image recognition can be derived that human body face characteristic results and Human Height characteristic results；
Described acquired in dress ornament information by common dress ornament image-recognizing method includes: to determine whether dress ornament can use bar code
Identification；Dress ornament bar code recognition；Body color identification and its colour system determine；The identification of dress ornament style can be derived that related dress ornament letter
Breath；
The acquisition in relation to dress ornament recommending data, analysis include: artificially to be actively entered in processing method needed for dress ornament recommends
Data；The public information of access internet passively inputs data needed for dress ornament is recommended；The analysis of data needed for dress ornament is recommended and place
Reason；
In summary the method take dress ornament recommendation specific service and the realization of a kind of synthetic image identification and big data
Human body barment tag result is comprised determining that in method；Determine garment ornament result；It determines and is actively entered data needed for dress ornament is recommended
As a result；Determine that passive input dress ornament recommends required data result；User behavior habituation mechanism；Individual subscriber and businessman's dress ornament
The selection mechanism of database needed for recommending；The operating mechanism of database needed for individual subscriber dress ornament is recommended；Businessman's dress ornament recommends institute
Need the operating mechanism of database；Dress ornament recommendation process unit carries out last recommendation results and presents,
Its feature, is:
The dress ornament of a kind of synthetic image identification and big data recommends specific service and implementation method, comprising:
Determine human body barment tag result；Determine garment ornament result；It determines and is actively entered data result needed for dress ornament is recommended；Really
Data result needed for fixed passive input dress ornament is recommended；User behavior habituation mechanism；Individual subscriber and businessman's dress ornament recommend institute
Need the selection mechanism of database；The operating mechanism of database needed for individual subscriber dress ornament is recommended；Data needed for businessman's dress ornament is recommended
The operating mechanism in library；Dress ornament recommendation process unit carries out last recommendation results and presents；
Wherein, the user behavior habituation mechanism, comprising:
By each human body barment tag result, garment ornament result, it is actively entered data result needed for dress ornament is recommended, passive input
Dress ornament recommend needed for data result counted, same or like result occur it is more, dress ornament recommend link can be preferential
Stress correlated results, reduces unnecessary repetitive operation；
Wherein, the selection mechanism of database needed for the individual subscriber and businessman's dress ornament are recommended, comprising:
There are two kinds of dress ornaments for a user and recommend selection scheme, if user needs to wear related dress ornament, Ke Yixuan at once
Select database needed for individual subscriber dress ornament is recommended；If user needs to buy and not wear at once, it can choose businessman's clothes
Recommending data library needed for adoring；
Wherein, the operating mechanism in recommending data library needed for the individual subscriber dress ornament, comprising:
User carries out dress ornament image recognition to the existing dress ornament of individual；Dress ornament image recognition result is inputted needed for personal dress ornament recommendation
Database；Database generalization user's human body barment tag result, individual subscriber garment ornament result, master needed for personal dress ornament is recommended
Data result, user behavior habituation mechanism needed for data result needed for dynamic input dress ornament is recommended, passive input dress ornament are recommended
Integrated data is provided to processing unit；
Wherein, the operating mechanism of database needed for businessman's dress ornament is recommended, comprising:
Businessman will sell dress ornament and carry out dress ornament image recognition；Number needed for dress ornament image recognition result input businessman's dress ornament is recommended
According to library；Library synthetic user human body barment tag result needed for businessman's dress ornament recommending data, is actively entered businessman's garment ornament result
Data result needed for dress ornament is recommended passively inputs result, user behavior habituation mechanism needed for dress ornament is recommended to processing unit
Integrated data is provided；
Wherein, the dress ornament recommendation process unit carries out last recommendation results presentation, comprising:
Determine the type of database needed for dress ornament is recommended；Data provided by database needed for recommending dress ornament are handled in turn
It completes dress ornament pairing and recommends；Dress ornament recommendation results are presented to user；
It is wherein, described to recommend data provided by required database handle on dress ornament and then complete dress ornament pairing and recommendation,
Include:
Suitable basic clothing type is determined by user's human body external feature, then recommends required data by being actively entered dress ornament
As a result the dress ornament that further reduction is specifically suitble to recommends required data result further to be contracted referring again to passive input dress ornament
Subtract specifically suitable dress ornament, the dress ornament in database needed for finally recommending with existing dress ornament is screened and matched, and is determined most
The dress ornament recommended eventually；
It is wherein, described that dress ornament recommendation results are presented to user, comprising:
It carries out first time dress ornament recommendation results to present, if met requirement of the user to recommendation dress ornament, terminates dress ornament and recommend, determine
Otherwise dress ornament recommendation results discard dress ornament recommendation results, and prompt to carry out dress ornament next time and recommend, until meeting user to being pushed away
The requirement for recommending dress ornament just terminates to prompt.
2. the method according to claim 1, which is characterized in that the Human Height characteristic image identification, comprising:
Whole Body image is obtained by camera；Height identification is carried out using existing height identification technology；Obtain Human Height
Characteristic results；
It is wherein, described that Whole Body image is obtained by camera, comprising:
First time shooting is carried out, height identification technology intermediate-resolution requirement is such as met, terminates shooting, determines Whole Body image,
Otherwise taken the photograph image is discarded, and prompts to be shot next time, is just terminated until meeting height identification technology intermediate-resolution requirement
Prompt；
First time shooting is carried out, angle requirement in height identification technology is such as met, terminates shooting, determines Whole Body image, it is no
Taken the photograph image is then discarded, and prompts to be shot next time, just terminates to prompt until meeting angle requirement in height identification technology；
First time shooting is carried out, context request in height identification technology is such as met, terminates shooting, determines Whole Body image, it is no
Taken the photograph image is then discarded, and prompts to be shot next time, just terminates to prompt until meeting context request in height identification technology；
First time shooting is carried out, illumination requirement in height identification technology is such as met, terminates shooting, determines Whole Body image, it is no
Taken the photograph image is then discarded, and prompts to be shot next time, just terminates to prompt until meeting illumination requirement in height identification technology；
It is wherein, described that height identification is carried out using existing height identification technology, comprising:
Height, which delimited, according to face characteristic identifies reference line；Height mapping relationship is established according to reference line；It is closed according to height mapping
System calculates current individual object height；A upper limit reference line and one two, lower limit reference line delimited according to same face characteristic
Reference line makes the corresponding face characteristic of the human body to be identified between two reference lines, and records two references
The ordinate of line；It is sat according to mapping relations of the corresponding face characteristic for the human body for being actually subjected to obtain between two reference lines are vertical
Mark, obtains the height of the human body；
It is wherein, described to obtain Human Height characteristic results, comprising:
It obtains height values and classifies, a kind of height values range is less than 160cm, and two class height values are more than or equal to
160cm is simultaneously less than or equal to 170cm, and three classes height values are greater than 170cm and are less than 175cm, and four class height values are greater than and are equal to
175cm is less than 180cm, and five class height values are greater than and are equal to 180cm.
3. the method according to claim 1, wherein the image-recognizing method of the common dress ornament, comprising:
Determine whether dress ornament can use bar code recognition；Dress ornament bar code recognition；Body color identification and its colour system determine；Clothes
Adorn style identification；
Wherein, determine whether dress ornament can use bar code recognition, comprising:
Whether user observes the dress ornament contains bar code relevant to the application is then taken if any associated bar code
Dress ornament barcode recognition method omits body color identification and its colour system determines, dress ornament style identifies two steps, directly carries out
It summarizes and sorts out；
Whether user observes the dress ornament contains bar code relevant to the application is then taken such as without associated bar code
Body color identification and its colour system determine, dress ornament style identifies two steps, omit dress ornament barcode recognition method, directly carry out
It summarizes and sorts out；
Wherein, dress ornament bar code recognition, comprising:
Establish dress ornament bar code data library；Generate dress ornament bar code data；Bar code is scanned, obtains bar code data, and corresponding
The corresponding dress ornament information in dress ornament bar code data library；
It is wherein, described to establish dress ornament bar code data library, comprising:
Each bar code corresponds to corresponding dress ornament information, and style is contained in dress ornament information, wears occasion, yardage, the place of production；
Wherein, body color identification and its colour system determine, comprising:
The body color of dress ornament is determined by image recognition；Colour system judgement is carried out to body color after determining body color；Finally
Sort out from cool colour, warm colour, intermediate colour system three's one of them；
Wherein, dress ornament style identifies, comprising:
Establish dress ornament Database of Garment；Common dress ornament style image information and affiliated dress ornament money are inputted in dress ornament Database of Garment
Formula；It identifies dress ornament image, extracts pertinent image information and dress ornament Database of Garment is matched；As dress ornament Database of Garment can be paired into
Function then exports corresponding dress ornament style, does not otherwise export.
4. the method according to claim 1, wherein a kind of acquisition in relation to dress ornament recommending data, point
Analysis, processing method, comprising:
Artificially it is actively entered data needed for dress ornament is recommended；The public information of access internet passively inputs number needed for dress ornament is recommended
According to；The analysis and processing of data needed for dress ornament is recommended；
Wherein, data needed for dress ornament is recommended artificially are actively entered, comprising:
It is actively entered user's gender；It is actively entered user's age；It is actively entered the purchase dress ornament expense that user can bear
Range；It is actively entered the dress ornament style of user's hobby；It is actively entered the occasion that user will wear；It is actively entered user
Desired style；
Wherein, the public information for accessing internet passively inputs data needed for dress ornament is recommended, comprising:
Passive input user locality real-time weather situation；The local humane custom feature of geography of passive input user；Passively
Input disclosed social information in the social software of user；The passive style information for inputting trend dress ornament instantly；
Wherein, the passive input user locality real-time weather situation, comprising:
User locality real time temperature；User locality real-time humidity；User local wind direction and wind-force in real time；User is local
Real-time air quality；
Wherein, the analysis and processing of data needed for dress ornament is recommended, comprising:
Data needed for the dress ornament being actively entered is recommended recommend required data to integrate with the dress ornament passively inputted；As active is defeated
Data needed for the dress ornament entered is recommended recommend required data to have differences with the dress ornament passively inputted, preferentially select the clothes being actively entered
Data needed for decorations are recommended；By treated, information categorization feeds back to recommender system.
5. the method according to claim 1, wherein the human body face characteristic image identifies, comprising: pass through
Camera obtains facial image；The main colour of skin of preliminary judgement face and aging degree；Establish facial characteristics vector；Image surface library into
Row compares and analysis；Obtain human body face characteristic results；
It is wherein, described that facial image is obtained by camera, comprising:
First time shooting is carried out, face resolution requirement in recognition of face is such as met, terminates shooting, determines facial image, otherwise
Discarded taken the photograph image, and prompt to be shot next time, just terminate to prompt until meeting face resolution requirement in recognition of face；
First time shooting is carried out, such as meets facial angle requirement in recognition of face, terminates shooting, determine facial image, otherwise give up
Taken the photograph image is abandoned, and prompts to be shot next time, requires just to terminate to prompt until meeting facial angle in recognition of face；
First time shooting is carried out, face context request in recognition of face is such as met, terminates shooting, determines facial image, otherwise give up
Taken the photograph image is abandoned, and prompts to be shot next time, just terminates to prompt until meeting face context request in recognition of face；
First time shooting is carried out, human face light requirement in recognition of face is such as met, terminates shooting, determines facial image, otherwise give up
Taken the photograph image is abandoned, and prompts to be shot next time, just terminates to prompt until meeting human face light requirement in recognition of face；
Wherein, the face resolution requirement is 36 × 27(ppi)；
Wherein, the facial angle requirement is that shoulder gradient section is -15 ° to+15 ° to the left and right on head, head pitch angle
Section is -20 ° to+20 °, and it is -10 ° to+10 ° that head, which is swung left and right and rotates angular interval,；
Wherein, the main colour of skin of the preliminary judgement face and aging degree, comprising:
Establish colour of skin sample database and aging degree sample database；Face's sample point is established, data are acquired；Carry out face texture and color
Analysis；It compares established sample database and obtains judgement result；
It is wherein, described to establish facial characteristics vector, comprising:
Locating human face's image major organs；With structure attribute feature representation face major organs；Using can feature identification model
Carry out information processing；Determine substantially shape of face；
Wherein, locating human face's image major organs are eyes, nose, eyebrow, ear, mouth, face's outer profile, chin,
Forehead；
Wherein, it is described with structure attribute in structure attribute feature representation face major organs be characterized in the distance between each organ,
Angle, position, size, shape；
Wherein, it is described using can feature identification model carry out in information processing can the model of feature identification be based on space right-angle
Coordinate system and the polynary three-dimensional vector established；
It is wherein, described to be compared and analyze in image surface library, comprising:
Common shape of face sample is compared and is analyzed；Common face complexion sample is compared and is analyzed；To all ages and classes
The skin quality sample of section is compared and analyzes；
Wherein, the common shape of face sample can be 1. round shapes of face；2. oval shape of face；3. oval shape of face；4. oval
Shape shape of face；5. rectangular shape of face；6. rectangle shape of face；7. trapezoidal shape of face；8. inverted trapezoidal shape of face；9. diamond shape shape of face；10. pentagon face
Type;It should be noted that can all express the same meaning by the proprietary term of other shapes of face in those skilled in the art；
Wherein, the common face complexion sample can be the 1. neutral colours of skin；2. the cold shallow colour of skin；3. the cold depth colour of skin；4. warming up shallow skin
Color；5 warm up the gorgeous colour of skin；6. the warm depth colour of skin；
Wherein, the skin quality sample of the different age group can be 1. early childhoods；2. Development in Puberty period；3. when young
Phase；4. the middle age；5. senectitude；
It is wherein, described to obtain human body face characteristic results, comprising:
Obtain basic shape of face feature；Obtain basic colour of skin result；User's age is tentatively obtained by face skin quality age characteristics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810822167.4A CN109284666B (en) | 2018-07-24 | 2018-07-24 | A kind of dress ornament recommended method based on image recognition and big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810822167.4A CN109284666B (en) | 2018-07-24 | 2018-07-24 | A kind of dress ornament recommended method based on image recognition and big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109284666A CN109284666A (en) | 2019-01-29 |
| CN109284666B true CN109284666B (en) | 2019-07-26 |
Family
ID=65183155
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810822167.4A Expired - Fee Related CN109284666B (en) | 2018-07-24 | 2018-07-24 | A kind of dress ornament recommended method based on image recognition and big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109284666B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175247B (en) * | 2019-03-13 | 2021-06-08 | 北京邮电大学 | Method for optimizing anomaly detection model based on deep learning |
| CN113191843B (en) * | 2021-04-28 | 2023-04-07 | 北京市商汤科技开发有限公司 | Simulation clothing fitting method and device, electronic equipment and storage medium |
| CN113706244A (en) * | 2021-08-26 | 2021-11-26 | 哈尔滨工业大学（威海） | Implementation method of intelligent clothes management collocation recommendation system based on end cloud fusion |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104584033A (en) * | 2012-06-08 | 2015-04-29 | 新加坡国立大学 | Interactive clothes searching in online stores |
| CN206892617U (en) * | 2017-07-12 | 2018-01-16 | 青岛新维纺织开发有限公司 | A kind of intelligent wardrobe management and clothing commending system |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140310304A1 (en) * | 2013-04-12 | 2014-10-16 | Ebay Inc. | System and method for providing fashion recommendations |
| CN104268761A (en) * | 2014-09-29 | 2015-01-07 | 深圳市百科在线科技发展有限公司 | Background product recommendation decision-making assisting method and system based on consumption features |
| CN104281964A (en) * | 2014-09-29 | 2015-01-14 | 深圳市百科在线科技发展有限公司 | Clothing product recommendation aid decision making method and system based on real-time human model |
| CN104484816B (en) * | 2014-12-19 | 2018-04-27 | 常州云从信息科技有限公司 | Clothes purchase guiding system and method based on big data analysis |
| CN105260480A (en) * | 2015-11-09 | 2016-01-20 | 魅族科技（中国）有限公司 | Information recommendation method and terminal |
| CN106355479A (en) * | 2016-09-22 | 2017-01-25 | 京东方科技集团股份有限公司 | Virtual fitting method, virtual fitting glasses and virtual fitting system |
| CN106579786B (en) * | 2016-12-04 | 2019-04-26 | 广东劳卡家具有限公司 | Home furnishings intelligent wardrobe and its control method |
| 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 |
-
2018
- 2018-07-24 CN CN201810822167.4A patent/CN109284666B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104584033A (en) * | 2012-06-08 | 2015-04-29 | 新加坡国立大学 | Interactive clothes searching in online stores |
| CN206892617U (en) * | 2017-07-12 | 2018-01-16 | 青岛新维纺织开发有限公司 | A kind of intelligent wardrobe management and clothing commending system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109284666A (en) | 2019-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109284666B (en) | A kind of dress ornament recommended method based on image recognition and big data | |
| US11748980B2 (en) | Makeup evaluation system and operating method thereof | |
| CN106650215A (en) | Skin type detection and individuation evaluation system and method based on cloud platform | |
| CN110264299A (en) | Clothes recommended method, device and computer equipment based on recognition of face | |
| CN102426652A (en) | Traditional Chinese medicine face color identifying and retrieving method based on image analysis | |
| CN109173263A (en) | A kind of image processing method and device | |
| CN110163810A (en) | A kind of image processing method, device and terminal | |
| CN106108523A (en) | A kind of portable intelligent cosmetic mirror, cosmetic aid system and method | |
| CN103945104B (en) | Information processing method and electronic equipment | |
| Eckert et al. | Facial cosmetics database and impact analysis on automatic face recognition | |
| CN108460398A (en) | Image processing method, device, cloud processing equipment and computer program product | |
| CN109063671A (en) | Method and device for intelligent cosmetic | |
| CN102426650A (en) | Method and device of character image analysis | |
| CN113707305A (en) | Health scheme recommendation method, device, equipment and medium based on deep learning | |
| WO2021083004A1 (en) | Photo display processing method and device, and video display processing method and device | |
| CN108920509A (en) | A kind of intelligent cosmetic mirror | |
| CN105979236A (en) | Image quality adjustment method and device | |
| CN110490711A (en) | A kind of personalized visualization design platform | |
| CN101893481A (en) | Method and system for testing and determining color attribute | |
| TWI462030B (en) | An intelligent system and algorithms of facial analysis and hairstyle fitting | |
| CN109143912A (en) | A kind of intelligence whole body mirror | |
| CN110956531A (en) | Dress management and matching system and method | |
| CN109310475A (en) | System and method for automatically generating facial repair capsule and application scheme to solve the facial deviation of observable | |
| CN104317866B (en) | A kind of ready-made clothes matching process and its electronic device | |
| JP5782540B2 (en) | Beauty counseling method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190726 Termination date: 20210724 |