CN108305146A - A kind of hair style recommendation method and system based on image recognition - Google Patents
A kind of hair style recommendation method and system based on image recognition Download PDFInfo
- Publication number
- CN108305146A CN108305146A CN201810091195.3A CN201810091195A CN108305146A CN 108305146 A CN108305146 A CN 108305146A CN 201810091195 A CN201810091195 A CN 201810091195A CN 108305146 A CN108305146 A CN 108305146A
- Authority
- CN
- China
- Prior art keywords
- image
- hair style
- user
- dimensional
- head
- 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
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000011161 development Methods 0.000 claims abstract description 49
- 230000001815 facial effect Effects 0.000 claims abstract description 46
- 238000012545 processing Methods 0.000 claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 230000002708 enhancing effect Effects 0.000 claims description 15
- 238000003708 edge detection Methods 0.000 claims description 10
- 238000004321 preservation Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 3
- 210000003128 head Anatomy 0.000 description 80
- 230000000694 effects Effects 0.000 description 18
- 230000003796 beauty Effects 0.000 description 5
- 238000012935 Averaging Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 235000013399 edible fruits Nutrition 0.000 description 4
- 230000008092 positive effect Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 241001270131 Agaricus moelleri Species 0.000 description 2
- 206010019049 Hair texture abnormal Diseases 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 210000000056 organ Anatomy 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000001131 transforming effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
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
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Economics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Processing Or Creating Images (AREA)
- Geometry (AREA)
- Image Processing (AREA)
- General Health & Medical Sciences (AREA)
Abstract
The invention belongs to field of computer technology, and in particular to a kind of hair style recommendation method and system based on image recognition include the following steps：S1 obtains the shooting image of user, carries out preliminary treatment to shooting image, obtains the head image of user and show head image；S2 carries out digitization analyzing processing to head image and obtains analysis result, and recommends a variety of hair styles to user according to analysis result；S3 obtains display type, hair style and the color development of user's selection, replaces the original hair style of head image and color development according to the hair style and color development and carry out two-dimentional display and Three-dimensional Display.The present invention is that user recommends suitable hair style, and is shown in user interface according to the hair style and color development of user's selection, the suitable hair style that can select oneself to like thereby using family according to the facial image that user shoots.
Description
Technical field
The invention belongs to field of computer technology, and in particular to a kind of hair style based on image recognition recommends method and is
System.
Background technology
If got a haircut in beauty parlor, when most, customer selects needs from cut-and-dried magazine and hair type
Hair style.In this approach, considerable customer complains hair style after haircut.This is because understanding and the customer of designer
It understands differently caused by cause.
According to the institute in Japanese patent application Laid-Open No.2001-344492 (hereinafter referred to as first prior art)
The beauty salon of description selects system, if the information terminal input that the search criterion of such as hair style possesses from customer, is searched
Device searches the beauty salon for meeting search criterion from the first database for storing store information, designer's information etc..Therefore,
Which depict customers can select the beauty salon that required hair style can be satisfied.
But such technology has following problem：Client can only select beauty salon, but nothing according to desired hair style
Method confirms whether oneself selected hair style meets oneself.
Invention content
For the defects in the prior art, the hair style based on image recognition that the present invention provides a kind of recommending method and is
System, the facial image according to user's shooting are that user recommends suitable hair style, and according to the hair style of user's selection and color development with
Family interface shown, the suitable hair style that can select oneself to like thereby using family.
In a first aspect, the present invention provides a kind of, the hair style based on image recognition recommends method, includes the following steps：
S1 obtains the shooting image of user, carries out preliminary treatment to shooting image, obtains the head image of user and display
Head image；
S2 carries out digitization analyzing processing to head image and obtains analysis result, and recommended to user according to analysis result
A variety of hair styles；
S3 obtains display type, hair style and the color development of user's selection, head image is replaced according to the hair style and color development
Original hair style and color development simultaneously carry out two-dimentional display and Three-dimensional Display.
Preferably, the S1 is specially：
The shooting image for obtaining user carries out denoising to shooting image and obtains smoothed image, increased by enhancing operator
The edge of strong smoothed image, carries out the smoothed image after enhancing edge the edge detection of face, hair and neck, to
Obtain include face, hair and neck head image, and user interface show head image.
Preferably, the S2 is specially：
Recognition of face and analyzing processing are carried out to head image, head image progress recognition of face is obtained into facial image,
Facial image is inputted into hair style recommended device, hair style recommended device carries out several hair styles in facial image and hair style template library
Match, obtain matching angle value, top n hair style is ranked up according to the sequence of matching angle value from big to small, hair style is obtained and recommends row
Table, and on a user interface, head image side show hair style recommendation list.
Preferably, the user interface includes for showing the head image sub-interface of head image, for showing N number of push away
Recommend the hair style recommendation list sub-interface of the matching angle value of hair style and each hair style, for showing that the color development of several color development selects son
Interface and display type sub-interface for showing display type.
Preferably, the S3 is specially：
The display type of user's selection is obtained, the display type includes two dimension display and Three-dimensional Display；
After display type determines, facial image is handled to obtain full two-dimensional faceform or Complete three-dimensional head mould
Type；
Obtain the hair style and color development of user's selection；
Increase hair style, color development and facial image on full two-dimensional faceform or Complete three-dimensional head model, obtains two
Dimension shows image or three dimensional rendered images；
Two-dimensional display image or three dimensional rendered images are shown on a user interface.
Preferably, described that facial image is handled to obtain full two-dimensional faceform or Complete three-dimensional head model tool
Body is：
The label of face mask and face profile is carried out to facial image；
Each image characteristic point in original two-dimensional faceform is iterated along the normal direction of profile, each
Continuous adjusting parameter in iteration so that entire model constantly adapts to target shape and keeps the Reasonable Shape of face, is most connect
The full two-dimensional faceform of person of modern times's face image；
Model characteristic point corresponding with image characteristic point is marked in initial three-dimensional head model, and according to each figure
As the depth information of characteristic point optimizes initial three-dimensional head model, the Complete three-dimensional head closest to user's head is obtained
Model.
Preferably, the original two-dimensional faceform is obtained after carrying out alignment normalized to several image patterns
's；The initial three-dimensional head model is obtained by three-dimensional head sample training several different.
Preferably, the step further includes：
S4 obtains the fine tuning instruction of user, is carried out to two-dimensional display image or three dimensional rendered images according to fine tuning instruction micro-
It adjusts, the two-dimensional display image after being optimized or three dimensional rendered images.
Preferably, the step further includes：
S5 obtains the preservation instruction of user, is instructed to the two-dimensional display image or three-dimensional display after optimization according to preserving
As being preserved.
Second aspect, the present invention provides a kind of hair style commending system based on image recognition are suitable for first aspect institute
The hair style based on image recognition stated recommends method, including：
Image processing unit, the shooting image for obtaining user carry out preliminary treatment to shooting image, obtain user's
Head image simultaneously shows head image；
Hair style recommendation unit obtains analysis result for carrying out digitization analyzing processing to head image, and according to analysis
As a result recommend a variety of hair styles to user；
Display unit is selected, display type, hair style and color development for obtaining user's selection, according to the hair style and color development
It replaces the original hair style of head image and color development and carries out two-dimentional display and Three-dimensional Display.
Beneficial effects of the present invention are：The present invention is that user recommends suitable hair style according to the facial image that user shoots,
And shown in user interface according to the hair style and color development of user's selection, it is suitble to thereby using what family can select oneself to like
Hair style.
Description of the drawings
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described.In all the appended drawings, similar element
Or part is generally identified by similar reference numeral.In attached drawing, each element or part might not be drawn according to actual ratio.
Fig. 1 is the flow chart that the hair style based on image recognition recommends method in the present embodiment；
Fig. 2 is the structure chart of the hair style commending system based on image recognition in the present embodiment.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
It should be appreciated that ought use in this specification and in the appended claims, term " comprising " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, element, component and/or its presence or addition gathered.
It is also understood that the term used in this description of the invention is merely for the sake of the mesh for describing specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singulative, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combinations and all possible combinations of one or more of associated item listed, and includes these combinations.
Embodiment one：
A kind of hair style recommendation method based on image recognition is present embodiments provided, as shown in Figure 1, the method for the present embodiment
It is run on user terminal and remote server.User terminal can download this system APP, and the user terminal includes intelligent hand
The intelligent terminals such as machine, tablet computer.
The method of the present embodiment includes the following steps：
S1 obtains the shooting image of user, carries out preliminary treatment to shooting image, obtains the head image of user and display
Head image.
The S1 is specially：
The shooting image for obtaining user carries out denoising to shooting image and obtains smoothed image, increased by enhancing operator
The edge of strong smoothed image, carries out the smoothed image after enhancing edge the edge detection of face, hair and neck, to
Obtain include face, hair and neck head image, and user interface show head image.
After user enters this system APP, hair style design control key is clicked, last time figure is prompted in the user interface of user terminal
Picture, user can select the image having had been taken to upload in photograph album, or be shot originally by the camera function of user terminal
Image after user terminal obtains the shooting image of user, can also occur locally carrying out image procossing to remote service
Device carries out image procossing.
Image procossing is carried out if remote server, then user terminal is sent to remote server by image is shot, remotely
Server is smoothed shooting image.Image shoot and transmission process in, due to being influenced by various interference, or pass
Defeated channel, sampling system are second-rate, and cause image crude, at this time, it is necessary to be smoothed to image.The present embodiment
Image procossing use neighborhood averaging, the neighborhood averaging be it is a kind of carrying out smooth technology directly in spatial domain, should
Technology is to be based on such a hypothesis：Image is made of the constant fritter of many gray scales, there is very high space between adjacent pixel
Correlation, and noise is then relatively independent, based on as above it is assumed that being averaged all pixels in a pixel and its neighborhood
Gray value is assigned to corresponding pixel in smoothed image, smooth to achieve the purpose that, also known as mean filter or local smoothing method method.
Because after having carried out denoising smooth processing to image, although being the reduction of picture noise, the marginal information of image
Also it can lose, the edge of image can thicken.In order to reduce this unfavorable effect, usually figure is sharpened using enhancing operator
Picture makes edge be apparent from, to facilitate subsequent edge detection.The enhancing operator is commonly used in adjustment brightness of image, right
Than degree or grey level distribution, the pixel value of output image as a result is taken on a new look according to the transforming function transformation function applied.
Behind the edge for enhancing image, the edge of face, hair and neck is detected using edge detection method, to
Obtain include face, hair and neck head image.There are many type of institute's edge detection method, such as method of differential operator, model
With method, Wavelet Detection method, neural network etc..And it is shown in the user interface of user terminal.
The user interface includes multiple small interfaces, such as showing the head image sub-interface of head image, for showing
The hair style recommendation list sub-interface of showing the matching angle value that five kinds are recommended hair styles and each hair style, the hair for showing several color development
Color sorting selects sub-interface, the display type sub-interface etc. for showing display type.
S2 carries out digitization analyzing processing to head image and obtains analysis result, and recommended to user according to analysis result
A variety of hair styles.
The S2 is specially：
Recognition of face and analyzing processing are carried out to head image, head image progress recognition of face is obtained into facial image,
Facial image is inputted into hair style recommended device, hair style recommended device carries out several hair styles in facial image and hair style template library
Match, obtain matching angle value, top n hair style is ranked up according to the sequence of matching angle value from big to small, hair style is obtained and recommends row
Table, and on a user interface, head image side show hair style recommendation list.
The limb recognition that the remote server of the present embodiment carries out head image facial contour extracts facial image, then
Facial image is inputted into hair style recommended device, hair style recommended device uses neural network algorithm, and hair style recommended device is by facial image and sends out
Several hair styles in type template library are matched, and obtain matching angle value.If be stored in hair style module library several shapes of face and
There are one dry hair style, each hair style and each hair style tools matches angle value, and hair style recommended device is identified point according to facial image
Analysis, finds immediate shape of face therewith, the matching angle value with each hair style is obtained according to immediate shape of face, further according to matching degree
Top n hair style is ranked up by the sequence of value from big to small, obtains hair style recommendation list, and hair style recommended device exports the hair style and pushes away
List is recommended, the hair style recommendation list includes N number of hair style and corresponding N number of matching angle value.Remote server recommends hair style
List is sent to user terminal, user terminal on a user interface, head image side show hair style recommendation list, for user into
Row selection.
S3 obtains display type, hair style and the color development of user's selection, head image is replaced according to the hair style and color development
Original hair style and color development simultaneously carry out two-dimentional display and Three-dimensional Display；
The S3 is specially：
The display type of user's selection is obtained, the display type includes two dimension display and Three-dimensional Display；
After display type determines, facial image is handled to obtain full two-dimensional faceform or Complete three-dimensional head mould
Type；
Obtain the hair style and color development of user's selection；
Increase hair style, color development and facial image on full two-dimensional faceform or Complete three-dimensional head model, obtains two
Dimension shows image or three dimensional rendered images；
Two-dimensional display image or three dimensional rendered images are shown on a user interface.
In the present embodiment, it is equipped with two dimension in the display type sub-interface of user interface and shows control key and Three-dimensional Display control
Key.After user selects two dimension to show control key or Three-dimensional Display control key, user terminal is by two-dimentional idsplay order or 3-D display command
It is sent to remote server.Original two-dimensional faceform and initial three-dimensional face model are prestored in remote server, it is described first
Beginning two-dimension human face model obtains after carrying out alignment normalized to several image patterns；The initial three-dimensional head model
It is to be obtained by three-dimensional head sample training several different.
The remote server of the present embodiment carries out facial image the label of face mask and face profile；
Each image characteristic point in original two-dimensional faceform is iterated along the normal direction of profile, each
Continuous adjusting parameter in iteration so that entire model constantly adapts to target shape and keeps the Reasonable Shape of face, is most connect
Person of modern times's face image full two-dimensional faceform (direct picture in the case of full two-dimensional faceform, that is, people's shaven head,
Because of facial image the blocking due to hair identified from the shooting image that user uploads, it is most likely that be incomplete).
Model characteristic point corresponding with image characteristic point is marked in initial three-dimensional head model, and according to each figure
As the depth information of characteristic point optimizes initial three-dimensional head model, the Complete three-dimensional head closest to user's head is obtained
Model.
The two-dimension human face model or three-dimensional head model are sent to user terminal by the present embodiment remote server, and
It is shown in user terminal.The matching angle value of each hair style shown on user's display interface is used to make reference to user,
Since everyone hobby is different, perhaps the most suitable hair style recommended according to this system is not user preferences, described reality
It applies and recommends a variety of hair styles by matching angle value in the hair style recommendation list of example, selected convenient for user.
If what user front selected is two-dimentional display, the present embodiment increases face figure on full two-dimensional faceform
The hair style and color development that picture, user select, to obtain two-dimensional display image, and show two-dimensional display image on a user interface,
The effect of this hair style of do-it-yourself is seen convenient for user.After user reselects hair style and color development, regenerate to obtain two-dimentional aobvious
Diagram picture, and show on a user interface, the effect of this hair style of do-it-yourself is seen convenient for user.
If user's selection is Three-dimensional Display, the present embodiment increases facial image on Complete three-dimensional faceform, uses
The hair style and color development of family selection, to obtain three dimensional rendered images, and show three dimensional rendered images, Yong Huke on a user interface
To be clicked in user interface, the operations such as stroke can not only see to realize three dimensional rendered images from a different perspective
Positive effect after to do-it-yourself this hair style can also be seen that lateral effect and back side effect.User sends out after reselecting hair style
After color, regenerates to obtain three dimensional rendered images, and show on a user interface, see convenient for user after having changed another hair style
Effect.
S4 obtains the fine tuning instruction of user, is carried out to two-dimensional display image or three dimensional rendered images according to fine tuning instruction micro-
It adjusts, the two-dimensional display image after being optimized or three dimensional rendered images.
In the present embodiment, the fine tuning instruction of the user includes the instruction of the facial light and shade of adjustment, the instruction of removal spot, adjusts
The instruction etc. of whole certain organs local detail, such as increase eyes, increase the bridge of the nose, thicken lip.Two-dimensional display image or three-dimensional
After showing that image is finely tuned after fine tuning instructs, customer satisfaction system two-dimensional display image or three dimensional rendered images after being optimized.
S5 obtains the preservation instruction of user, is instructed to the two-dimensional display image or three-dimensional display after optimization according to preserving
As being preserved.
In the present embodiment, after user has finely tuned, the preservation instruction of user interface can be clicked, to by two-dimensional display image or
Three dimensional rendered images are stored in local user terminal, so that user's later stage uses, such as QQ head portraits, wechat head portrait.
In the present embodiment, the facial image according to user's shooting is that user recommends suitable hair style, and is selected according to user
Hair style and color development carry out two-dimentional display or Three-dimensional Display in user interface, the effect of the different hair style of do-it-yourself is seen convenient for user
Fruit.For Three-dimensional Display, the operations such as user can click in user interface, stroke, to realize from a different perspective
Three dimensional rendered images not can be only seen the positive effect after do-it-yourself this hair style and can also be seen that lateral effect and back side effect
Fruit, being capable of the suitable hair style liked of better choice oneself thereby using family.
Embodiment two：
Present embodiments provide a kind of hair style commending system based on image recognition, be suitable for embodiment one described in based on
The hair style of image recognition recommends method, as shown in Fig. 2, including image processing unit, hair style recommendation unit, selection display unit,
Fine-adjusting unit, storage unit etc..
Image processing unit, the shooting image for obtaining user carry out preliminary treatment to shooting image, obtain user's
Head image simultaneously shows head image.
Hair style recommendation unit obtains analysis result for carrying out digitization analyzing processing to head image, and according to analysis
As a result recommend a variety of hair styles to user.
Display unit is selected, display type, hair style and color development for obtaining user's selection, according to the hair style and color development
It replaces the original hair style of head image and color development and carries out two-dimentional display and Three-dimensional Display.
Fine-adjusting unit, the fine tuning for obtaining user instructs, according to fine tuning instruction to two-dimensional display image or Three-dimensional Display
Image is finely adjusted, the two-dimensional display image after being optimized or three dimensional rendered images.
Storage unit, for obtain user preservation instruct, according to preserve instruction to after optimization two-dimensional display image or
Three dimensional rendered images are preserved.
As shown in Figure 1, the hair style recommends method to include the following steps：
S1 obtains the shooting image of user, carries out preliminary treatment to shooting image, obtains the head image of user and display
Head image.
The S1 is specially：
The shooting image for obtaining user carries out denoising to shooting image and obtains smoothed image, increased by enhancing operator
The edge of strong smoothed image, carries out the smoothed image after enhancing edge the edge detection of face, hair and neck, to
Obtain include face, hair and neck head image, and user interface show head image.
After user enters this system APP, hair style design control key is clicked, last time figure is prompted in the user interface of user terminal
Picture, user can select the image having had been taken to upload in photograph album, or be shot originally by the camera function of user terminal
Image after user terminal obtains the shooting image of user, can also occur locally carrying out image procossing to remote service
Device carries out image procossing.
Image procossing is carried out if remote server, then user terminal is sent to remote server by image is shot, remotely
Server is smoothed shooting image.Image shoot and transmission process in, due to being influenced by various interference, or pass
Defeated channel, sampling system are second-rate, and cause image crude, at this time, it is necessary to be smoothed to image.The present embodiment
Image procossing use neighborhood averaging, the neighborhood averaging be it is a kind of carrying out smooth technology directly in spatial domain, should
Technology is to be based on such a hypothesis：Image is made of the constant fritter of many gray scales, there is very high space between adjacent pixel
Correlation, and noise is then relatively independent, based on as above it is assumed that being averaged all pixels in a pixel and its neighborhood
Gray value is assigned to corresponding pixel in smoothed image, smooth to achieve the purpose that, also known as mean filter or local smoothing method method.
Because after having carried out denoising smooth processing to image, although being the reduction of picture noise, the marginal information of image
Also it can lose, the edge of image can thicken.In order to reduce this unfavorable effect, usually figure is sharpened using enhancing operator
Picture makes edge be apparent from, to facilitate subsequent edge detection.The enhancing operator is commonly used in adjustment brightness of image, right
Than degree or grey level distribution, the pixel value of output image as a result is taken on a new look according to the transforming function transformation function applied.
Behind the edge for enhancing image, the edge of face, hair and neck is detected using edge detection method, to
Obtain include face, hair and neck head image.There are many type of institute's edge detection method, such as method of differential operator, model
With method, Wavelet Detection method, neural network etc..And it is shown in the user interface of user terminal.
The user interface includes multiple small interfaces, such as showing the head image sub-interface of head image, for showing
The hair style recommendation list sub-interface of showing the matching angle value that five kinds are recommended hair styles and each hair style, the hair for showing several color development
Color sorting selects sub-interface, the display type sub-interface etc. for showing display type.
S2 carries out digitization analyzing processing to head image and obtains analysis result, and recommended to user according to analysis result
A variety of hair styles.
The S2 is specially：
Recognition of face and analyzing processing are carried out to head image, head image progress recognition of face is obtained into facial image,
Facial image is inputted into hair style recommended device, hair style recommended device carries out several hair styles in facial image and hair style template library
Match, obtain matching angle value, top n hair style is ranked up according to the sequence of matching angle value from big to small, hair style is obtained and recommends row
Table, and on a user interface, head image side show hair style recommendation list.
The limb recognition that the remote server of the present embodiment carries out head image facial contour extracts facial image, then
Facial image is inputted into hair style recommended device, hair style recommended device uses neural network algorithm, and hair style recommended device is by facial image and sends out
Several hair styles in type template library are matched, and obtain matching angle value.If be stored in hair style module library several shapes of face and
There are one dry hair style, each hair style and each hair style tools matches angle value, and hair style recommended device is identified point according to facial image
Analysis, finds immediate shape of face therewith, the matching angle value with each hair style is obtained according to immediate shape of face, further according to matching degree
Top n hair style is ranked up by the sequence of value from big to small, obtains hair style recommendation list, and hair style recommended device exports the hair style and pushes away
List is recommended, the hair style recommendation list includes N number of hair style and corresponding N number of matching angle value.Remote server recommends hair style
List is sent to user terminal, user terminal on a user interface, head image side show hair style recommendation list, for user into
Row selection.
S3 obtains display type, hair style and the color development of user's selection, head image is replaced according to the hair style and color development
Original hair style and color development simultaneously carry out two-dimentional display and Three-dimensional Display；
The S3 is specially：
The display type of user's selection is obtained, the display type includes two dimension display and Three-dimensional Display；
After display type determines, facial image is handled to obtain full two-dimensional faceform or Complete three-dimensional head mould
Type；
Obtain the hair style and color development of user's selection；
Increase hair style, color development and facial image on full two-dimensional faceform or Complete three-dimensional head model, obtains two
Dimension shows image or three dimensional rendered images；
Two-dimensional display image or three dimensional rendered images are shown on a user interface.
In the present embodiment, it is equipped with two dimension in the display type sub-interface of user interface and shows control key and Three-dimensional Display control
Key.After user selects two dimension to show control key or Three-dimensional Display control key, user terminal is by two-dimentional idsplay order or 3-D display command
It is sent to remote server.Original two-dimensional faceform and initial three-dimensional face model are prestored in remote server, it is described first
Beginning two-dimension human face model obtains after carrying out alignment normalized to several image patterns；The initial three-dimensional head model
It is to be obtained by three-dimensional head sample training several different.
The remote server of the present embodiment carries out facial image the label of face mask and face profile；
Each image characteristic point in original two-dimensional faceform is iterated along the normal direction of profile, each
Continuous adjusting parameter in iteration so that entire model constantly adapts to target shape and keeps the Reasonable Shape of face, is most connect
Person of modern times's face image full two-dimensional faceform (direct picture in the case of full two-dimensional faceform, that is, people's shaven head,
Because of facial image the blocking due to hair identified from the shooting image that user uploads, it is most likely that be incomplete).
Model characteristic point corresponding with image characteristic point is marked in initial three-dimensional head model, and according to each figure
As the depth information of characteristic point optimizes initial three-dimensional head model, the Complete three-dimensional head closest to user's head is obtained
Model.
The two-dimension human face model or three-dimensional head model are sent to user terminal by the present embodiment remote server, and
It is shown in user terminal.The matching angle value of each hair style shown on user's display interface is used to make reference to user,
Since everyone hobby is different, perhaps the most suitable hair style recommended according to this system is not user preferences, described reality
It applies and recommends a variety of hair styles by matching angle value in the hair style recommendation list of example, selected convenient for user.
If what user front selected is two-dimentional display, the present embodiment increases face figure on full two-dimensional faceform
The hair style and color development that picture, user select, to obtain two-dimensional display image, and show two-dimensional display image on a user interface,
The effect of this hair style of do-it-yourself is seen convenient for user.After user reselects hair style and color development, regenerate to obtain two-dimentional aobvious
Diagram picture, and show on a user interface, the effect of this hair style of do-it-yourself is seen convenient for user.
If user's selection is Three-dimensional Display, the present embodiment increases facial image on Complete three-dimensional faceform, uses
The hair style and color development of family selection, to obtain three dimensional rendered images, and show three dimensional rendered images, Yong Huke on a user interface
To be clicked in user interface, the operations such as stroke can not only see to realize three dimensional rendered images from a different perspective
Positive effect after to do-it-yourself this hair style can also be seen that lateral effect and back side effect.User sends out after reselecting hair style
After color, regenerates to obtain three dimensional rendered images, and show on a user interface, see convenient for user after having changed another hair style
Effect.
S4 obtains the fine tuning instruction of user, is carried out to two-dimensional display image or three dimensional rendered images according to fine tuning instruction micro-
It adjusts, the two-dimensional display image after being optimized or three dimensional rendered images.
In the present embodiment, the fine tuning instruction of the user includes the instruction of the facial light and shade of adjustment, the instruction of removal spot, adjusts
The instruction etc. of whole certain organs local detail, such as increase eyes, increase the bridge of the nose, thicken lip.Two-dimensional display image or three-dimensional
After showing that image is finely tuned after fine tuning instructs, customer satisfaction system two-dimensional display image or three dimensional rendered images after being optimized.
S5 obtains the preservation instruction of user, is instructed to the two-dimensional display image or three-dimensional display after optimization according to preserving
As being preserved.
In the present embodiment, after user has finely tuned, the preservation instruction of user interface can be clicked, to by two-dimensional display image or
Three dimensional rendered images are stored in local user terminal, so that user's later stage uses, such as QQ head portraits, wechat head portrait.
In the present embodiment, the facial image according to user's shooting is that user recommends suitable hair style, and is selected according to user
Hair style and color development carry out two-dimentional display or Three-dimensional Display in user interface, the effect of the different hair style of do-it-yourself is seen convenient for user
Fruit.For Three-dimensional Display, the operations such as user can click in user interface, stroke, to realize from a different perspective
Three dimensional rendered images not can be only seen the positive effect after do-it-yourself this hair style and can also be seen that lateral effect and back side effect
Fruit, being capable of the suitable hair style liked of better choice oneself thereby using family.
Those of ordinary skill in the art may realize that lists described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and method and step, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, depend on the specific application and design constraint of technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
In several embodiments provided herein, it should be understood that disclosed method and system can pass through it
Its mode is realized.For example, in addition the division of the above unit, only a kind of division of logic function can have in actual implementation
Dividing mode, such as multiple units or component can be combined or can be integrated into another system or some features can be with
Ignore, or does not execute.Said units may or may not be physically separated, and the component shown as unit can be with
It is or may not be physical unit, you can be located at a place, or may be distributed over multiple network units.It can
The purpose of the embodiment of the present invention is realized to select according to the actual needs some or all of unit therein.
Finally it should be noted that：The above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Present invention has been described in detail with reference to the aforementioned embodiments for pipe, it will be understood by those of ordinary skill in the art that：Its according to
So can with technical scheme described in the above embodiments is modified, either to which part or all technical features into
Row equivalent replacement；And these modifications or replacements, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme should all cover in the claim of the present invention and the range of specification.
Claims (10)
1. a kind of hair style based on image recognition recommends method, which is characterized in that include the following steps：
S1 obtains the shooting image of user, carries out preliminary treatment to shooting image, obtains the head image of user and show head
Image；
S2 carries out digitization analyzing processing to head image and obtains analysis result, and a variety of to user's recommendation according to analysis result
Hair style；
S3 obtains display type, hair style and the color development of user's selection, it is original to replace head image according to the hair style and color development
Hair style and color development and carry out two-dimentional display and Three-dimensional Display.
2. hair style according to claim 1 based on image recognition recommends method, which is characterized in that the S1 is specially：
The shooting image for obtaining user carries out denoising to shooting image and obtains smoothed image, flat by enhancing operator enhancing
The edge of sliding image, carries out the smoothed image after enhancing edge the edge detection of face, hair and neck, to obtain
Include the head image of face, hair and neck, and head image is shown in user interface.
3. hair style according to claim 2 based on image recognition recommends method, which is characterized in that the S2 is specially：
Recognition of face and analyzing processing are carried out to head image, head image progress recognition of face is obtained into facial image, by people
Face image inputs hair style recommended device, and hair style recommended device matches facial image with several hair styles in hair style template library,
Matching angle value is obtained, top n hair style is ranked up according to the sequence of matching angle value from big to small, obtains hair style recommendation list,
And on a user interface, head image side show hair style recommendation list.
4. the hair style according to claim 3 based on image recognition recommends method, which is characterized in that the user interface packet
Include head image sub-interface for showing head image, the matching angle value for showing N number of recommendation hair style and each hair style
Hair style recommendation list sub-interface, the color development selection sub-interface for showing several color development and the display for showing display type
Type sub-interface.
5. hair style according to claim 4 based on image recognition recommends method, which is characterized in that the S3 is specially：
The display type of user's selection is obtained, the display type includes two dimension display and Three-dimensional Display；
After display type determines, facial image is handled to obtain full two-dimensional faceform or Complete three-dimensional head model；
Obtain the hair style and color development of user's selection；
Increase hair style, color development and facial image on full two-dimensional faceform or Complete three-dimensional head model, it is aobvious to obtain two dimension
Diagram picture or three dimensional rendered images；
Two-dimensional display image or three dimensional rendered images are shown on a user interface.
6. the hair style according to claim 5 based on image recognition recommends method, which is characterized in that described to facial image
It is handled to obtain full two-dimensional faceform or Complete three-dimensional head model is specially：
The label of face mask and face profile is carried out to facial image；
Each image characteristic point in original two-dimensional faceform is iterated along the normal direction of profile, in each iteration
In continuous adjusting parameter so that entire model constantly adapts to target shape and the Reasonable Shape of holding face, obtains closest to people
The full two-dimensional faceform of face image；
Model characteristic point corresponding with image characteristic point is marked in initial three-dimensional head model, and according to each image spy
The depth information of sign point optimizes initial three-dimensional head model, obtains the Complete three-dimensional head mould closest to user's head
Type.
7. the hair style according to claim 6 based on image recognition recommends method, which is characterized in that the original two-dimensional people
Face model obtains after carrying out alignment normalized to several image patterns；If the initial three-dimensional head model be by
A dry different three-dimensional head sample training obtains.
8. the hair style according to claim 1 based on image recognition recommends method, which is characterized in that the step is also wrapped
It includes：
S4 obtains the fine tuning instruction of user, is finely adjusted, is obtained to two-dimensional display image or three dimensional rendered images according to fine tuning instruction
Two-dimensional display image after to optimization or three dimensional rendered images.
9. the hair style according to claim 8 based on image recognition recommends method, which is characterized in that the step is also wrapped
It includes：
S5, obtain user preservation instruction, according to preserve instruction to after optimization two-dimensional display image or three dimensional rendered images into
Row preserves.
10. a kind of hair style commending system based on image recognition, which is characterized in that be suitable for any one institute of claim 1-9
The hair style based on image recognition stated recommends method, including：
Image processing unit, the shooting image for obtaining user carry out preliminary treatment to shooting image, obtain the head of user
Image simultaneously shows head image；
Hair style recommendation unit obtains analysis result for carrying out digitization analyzing processing to head image, and according to analysis result
Recommend a variety of hair styles to user；
Display unit is selected, display type, hair style and color development for obtaining user's selection are replaced according to the hair style and color development
Fall the original hair style of head image and color development and carries out two-dimentional display and Three-dimensional Display.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810091195.3A CN108305146A (en) | 2018-01-30 | 2018-01-30 | A kind of hair style recommendation method and system based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810091195.3A CN108305146A (en) | 2018-01-30 | 2018-01-30 | A kind of hair style recommendation method and system based on image recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108305146A true CN108305146A (en) | 2018-07-20 |
Family
ID=62867284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810091195.3A Pending CN108305146A (en) | 2018-01-30 | 2018-01-30 | A kind of hair style recommendation method and system based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108305146A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109167914A (en) * | 2018-09-25 | 2019-01-08 | 维沃移动通信有限公司 | A kind of image processing method and mobile terminal |
| CN109190574A (en) * | 2018-09-13 | 2019-01-11 | 郑州云海信息技术有限公司 | A kind of hair style recommended method, device, terminal and storage medium based on big data |
| CN109191255A (en) * | 2018-09-04 | 2019-01-11 | 中山大学 | A kind of commodity alignment schemes based on the detection of unsupervised characteristic point |
| CN109544262A (en) * | 2018-09-30 | 2019-03-29 | 百度在线网络技术（北京）有限公司 | Item recommendation method, device, electronic equipment, system and readable storage medium storing program for executing |
| CN110009708A (en) * | 2019-04-10 | 2019-07-12 | 上海大学 | Color development transform method, system and terminal based on color segmentation |
| CN111507791A (en) * | 2019-01-31 | 2020-08-07 | 北京奇虎科技有限公司 | Image-based hair style transformation method and device, computer equipment and storage medium |
| CN112015934A (en) * | 2020-08-27 | 2020-12-01 | 华南理工大学 | Intelligent hair style recommendation method, device and system based on neural network and Unity |
| CN112102148A (en) * | 2019-06-18 | 2020-12-18 | 北京陌陌信息技术有限公司 | Figure hair style replacing method, device, equipment and medium based on neural network |
| CN113112616A (en) * | 2021-04-06 | 2021-07-13 | 济南大学 | Three-dimensional generation method based on improved editable |
| CN113724366A (en) * | 2020-05-25 | 2021-11-30 | 北京新氧科技有限公司 | 3D model generation method, device and equipment |
| CN113744367A (en) * | 2021-09-10 | 2021-12-03 | 电子科技大学 | System and method for editing portrait hairstyle in two-dimensional image |
| CN117389676A (en) * | 2023-12-13 | 2024-01-12 | 成都白泽智汇科技有限公司 | Intelligent hairstyle adaptive display method based on display interface |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663820A (en) * | 2012-04-28 | 2012-09-12 | 清华大学 | Three-dimensional head model reconstruction method |
| CN103065360A (en) * | 2013-01-16 | 2013-04-24 | 重庆绿色智能技术研究院 | Generation method and generation system of hair style effect pictures |
| CN104794275A (en) * | 2015-04-16 | 2015-07-22 | 北京联合大学 | Face and hair style matching model for mobile terminal |
| CN106326815A (en) * | 2015-06-30 | 2017-01-11 | 芋头科技（杭州）有限公司 | Human face image recognition method |
-
2018
- 2018-01-30 CN CN201810091195.3A patent/CN108305146A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663820A (en) * | 2012-04-28 | 2012-09-12 | 清华大学 | Three-dimensional head model reconstruction method |
| CN103065360A (en) * | 2013-01-16 | 2013-04-24 | 重庆绿色智能技术研究院 | Generation method and generation system of hair style effect pictures |
| CN104794275A (en) * | 2015-04-16 | 2015-07-22 | 北京联合大学 | Face and hair style matching model for mobile terminal |
| CN106326815A (en) * | 2015-06-30 | 2017-01-11 | 芋头科技（杭州）有限公司 | Human face image recognition method |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109191255B (en) * | 2018-09-04 | 2022-04-15 | 中山大学 | Commodity alignment method based on unsupervised feature point detection |
| CN109191255A (en) * | 2018-09-04 | 2019-01-11 | 中山大学 | A kind of commodity alignment schemes based on the detection of unsupervised characteristic point |
| CN109190574A (en) * | 2018-09-13 | 2019-01-11 | 郑州云海信息技术有限公司 | A kind of hair style recommended method, device, terminal and storage medium based on big data |
| CN109167914A (en) * | 2018-09-25 | 2019-01-08 | 维沃移动通信有限公司 | A kind of image processing method and mobile terminal |
| CN109544262A (en) * | 2018-09-30 | 2019-03-29 | 百度在线网络技术（北京）有限公司 | Item recommendation method, device, electronic equipment, system and readable storage medium storing program for executing |
| CN111507791A (en) * | 2019-01-31 | 2020-08-07 | 北京奇虎科技有限公司 | Image-based hair style transformation method and device, computer equipment and storage medium |
| CN110009708A (en) * | 2019-04-10 | 2019-07-12 | 上海大学 | Color development transform method, system and terminal based on color segmentation |
| CN112102148A (en) * | 2019-06-18 | 2020-12-18 | 北京陌陌信息技术有限公司 | Figure hair style replacing method, device, equipment and medium based on neural network |
| CN113724366B (en) * | 2020-05-25 | 2024-02-27 | 北京新氧科技有限公司 | 3D model generation method, device and equipment |
| CN113724366A (en) * | 2020-05-25 | 2021-11-30 | 北京新氧科技有限公司 | 3D model generation method, device and equipment |
| CN112015934B (en) * | 2020-08-27 | 2022-07-26 | 华南理工大学 | Intelligent hair style recommendation method, device and system based on neural network and Unity |
| CN112015934A (en) * | 2020-08-27 | 2020-12-01 | 华南理工大学 | Intelligent hair style recommendation method, device and system based on neural network and Unity |
| CN113112616A (en) * | 2021-04-06 | 2021-07-13 | 济南大学 | Three-dimensional generation method based on improved editable |
| CN113744367A (en) * | 2021-09-10 | 2021-12-03 | 电子科技大学 | System and method for editing portrait hairstyle in two-dimensional image |
| CN113744367B (en) * | 2021-09-10 | 2023-08-08 | 电子科技大学 | System and method for editing portrait hairstyle in two-dimensional image |
| CN117389676A (en) * | 2023-12-13 | 2024-01-12 | 成都白泽智汇科技有限公司 | Intelligent hairstyle adaptive display method based on display interface |
| CN117389676B (en) * | 2023-12-13 | 2024-02-13 | 成都白泽智汇科技有限公司 | Intelligent hairstyle adaptive display method based on display interface |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108305146A (en) | A kind of hair style recommendation method and system based on image recognition | |
| Shih et al. | Style transfer for headshot portraits | |
| CN107527318B (en) | Hair style replacement method based on generation countermeasure network model | |
| Baskan et al. | Projection based method for segmentation of human face and its evaluation | |
| Mechrez et al. | Saliency driven image manipulation | |
| Xiao et al. | Fast shadow removal using adaptive multi‐scale illumination transfer | |
| US9589178B2 (en) | Image processing with facial features | |
| Shackleton et al. | Classification of facial features for recognition | |
| CN109598210B (en) | Picture processing method and device | |
| CN105512617B (en) | Method for screening images and device | |
| Keren | Painter identification using local features and naive bayes | |
| CN101505659A (en) | System and method for counting follicular units | |
| GB2379823A (en) | Selecting images based on the similarity between images | |
| US10217265B2 (en) | Methods and systems of generating a parametric eye model | |
| KR20010041177A (en) | Automatic determination of preset positions corresponding to participants in video-conferences | |
| CN106127181A (en) | One is virtual tries manicure method, system on | |
| CN108319894A (en) | Fruit recognition methods based on deep learning and device | |
| CN104346136A (en) | Method and device for picture processing | |
| CN108537126A (en) | A kind of face image processing system and method | |
| KR101967814B1 (en) | Apparatus and method for matching personal color | |
| CN106203286A (en) | The content acquisition method of a kind of augmented reality, device and mobile terminal | |
| Wang et al. | An effective background estimation method for shadows removal of document images | |
| Battiato et al. | Automatic image enhancement by content dependent exposure correction | |
| CN109389076A (en) | Image partition method and device | |
| Gritzman et al. | Comparison of colour transforms used in lip segmentation algorithms |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180720 |
|
| RJ01 | Rejection of invention patent application after publication |