CN103488756B - Picture classification method and terminal - Google Patents
Picture classification method and terminal Download PDFInfo
- Publication number
- CN103488756B CN103488756B CN201310440197.6A CN201310440197A CN103488756B CN 103488756 B CN103488756 B CN 103488756B CN 201310440197 A CN201310440197 A CN 201310440197A CN 103488756 B CN103488756 B CN 103488756B
- Authority
- CN
- China
- Prior art keywords
- picture
- similarity
- level
- classification
- storage position
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
Landscapes
- Engineering & Computer Science (AREA)
- Library & Information Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The embodiment of the invention discloses a picture classification method. The picture classification method comprises determining the similarity of reference pictures of designated features and non-reference pictures and dividing the similarity into N levels, wherein N is a natural number larger than 1; respectively storing the non-reference pictures into different picture storage positions according to the different levels of the similarity; storing a first picture into the picture storage position of a corresponding level according to a classification adjustment instruction when the classification adjustment instruction of the first picture at a storage position of the pictures with a K-level similarity is obtained, wherein K is a natural number smaller than N and the first picture is any one or more pictures at the storage position of the pictures with the K-level similarity and selected by a terminal user. The embodiment of the invention further discloses a terminal. Through the application of the picture classification method and the terminal, advantages that the efficiency of picture classification can be improved and the convenience of picture browsing of the terminal user can be improved can be obtained.
Description
Technical field
The present invention relates to electronic technology field, particularly relate to method and the terminal of a kind of picture classification.
Background technology
Currently along with development and the reduction of mobile phone production cost of mobile phone research and development technology, mobile phone becomes increasingly popular people
Daily life in, mobile phone camera pixel and the raising of hardware, also make mobile phone photograph become personage raw
One beautiful landscape in the middle of work.In daily life, cellphone subscriber likes recording each essence with mobile phone
Color moment or the moment that each is important, record the people oneself got close to, the sound of the people liked with mobile phone
Holding smiling expression, this makes the picture number in mobile phone picture library more and more huger.Additionally, along with cell-phone function
Becoming stronger day by day, also become one of the most frequently used function of mobile phone by surfing Internet with cell phone, surfing Internet with cell phone then may be
Mobile phone picture library increases and browses the picture of webpage, by the picture etc. of mobile phone-downloaded, the picture in mobile phone picture library
Kind is the most various, and lookup is browsed and the most increasingly bothered.
Personage's picture during cellphone subscriber searches mobile phone picture library for convenience in prior art, can be by mobile phone picture library
In there is the photo finishing of face characteristic out, be placed in a file, for user search, browse.So
And, in the prior art, when the picture classification in mobile phone picture library is made mistakes, need the picture that classification is made mistakes to enter
When row classification adjusts, need user to be operated manually, manually search corresponding picture storage file in picture library
Folder, picture classification made mistakes is transferred in correct file.Picture classification method of the prior art exists
When picture number in mobile phone picture library is huger, needs to adjust if there being picture classification to make mistakes, then need user's hands
Dynamic go to browse a large amount of picture searching corresponding picture storage folder, manually adjust the storage position of picture, behaviour
Bother, workload big, and classification effectiveness is low.
Summary of the invention
The embodiment of the present invention provides a kind of method of picture classification and a kind of terminal, improves picture classification and adjustment
Efficiency, improve end-user browses picture convenience.
A kind of method embodiments providing picture classification, including:
Determine the similarity of reference picture and the non-reference picture with specific characteristic, and described similarity is divided
For N number of level, wherein N is the natural number more than 1；
Described non-reference picture is stored to different picture storage positions respectively according to the different layers level of similarity；
When on the storage position get the picture that similarity is K level, the classification of the first picture adjusts instruction
Time, adjust instruction according to described classification and described first picture stored the picture storage position to corresponding level,
Wherein K is the natural number less than N, described first picture be the described similarity that terminal use chooses be K
Any one or plurality of pictures on the storage position of the picture of level.
The embodiment of the present invention additionally provides a kind of terminal, including:
Determine module, for determining the similarity of reference picture and the non-reference picture with specific characteristic, and
Described similarity is divided into N number of level, and wherein N is the natural number more than 1；
Sort module, for storing described non-reference picture respectively to difference according to the different layers level of similarity
Picture storage position；
Adjusting module, for the first picture on the storage position getting the picture that similarity is K level
Classification adjust instruction time, according to described classification adjust instruction described first picture is stored to corresponding level
Picture storage position, wherein K is the natural number less than N, and described first picture is the institute that terminal use chooses
State any one or plurality of pictures on the storage position of the picture that similarity is K level.
The embodiment of the present invention can the similarity of picture of computing terminal primary importance storage, and similarity is divided into
Multiple levels, and then can quickly picture be carried out classification storage, and when needs carry out classification adjustment to picture
Quickly picture is adjusted and store to corresponding storage position, improve the efficiency of picture classification, enhance end
The Consumer's Experience of picture browsing in end picture library.
Accompanying drawing explanation
For the technical scheme being illustrated more clearly that in the embodiment of the present invention, institute in embodiment being described below
The accompanying drawing used is needed to be briefly described, it should be apparent that, the accompanying drawing in describing below is the one of the present invention
A little embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, also
Other accompanying drawing can be obtained according to these accompanying drawings.
Fig. 1 is the embodiment schematic flow sheet of the method for the picture classification that the embodiment of the present invention provides；
Fig. 2 is a structural representation of the embodiment of the terminal that the embodiment of the present invention provides；
Fig. 3 is another structural representation of the embodiment of the terminal that the embodiment of the present invention provides.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clearly
Chu, it is fully described by, it is clear that described embodiment is a part of embodiment of the present invention rather than all
Embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creative labor
The every other embodiment obtained under dynamic premise, broadly falls into the scope of protection of the invention.
Terminal described in the embodiment of the present invention, it may include: mobile phone, panel computer, notebook computer etc.,
Above-mentioned terminal is only citing, and non exhaustive, including but not limited to above-mentioned terminal.Below will as a example by mobile phone,
Method and terminal to the volume picture classification described in the embodiment of the present invention are specifically described.
See Fig. 1, be the embodiment schematic flow sheet of the method for the picture classification that the embodiment of the present invention provides.This
The method of the picture classification described in embodiment, including step:
S101, determines the similarity of reference picture and the non-reference picture with specific characteristic, and by described phase
It is divided into N number of level like degree.
In the embodiment that some are feasible, the picture in the present embodiment with specific characteristic can be to include
The picture of character facial feature, can also be the picture etc. of picture or the specific background comprising particular scene.
Character facial feature comprises the steps that character facial profile, character facial five official rank；Particular scene can be flower or
Other plant of person, animal or object, as a example by flower, comprise colored color, profile etc..
S102, stores described non-reference picture respectively to different pictures according to the different layers level of similarity and deposits
Storage space is put.
Terminal primary importance concretely hands in the embodiment that some are feasible, described in the present embodiment
Machine picture library, when in the mobile phone picture library of user, the picture number of storage is huger, when the kind of picture is more,
When user searches the photo of a certain individual or some people in picture library, difficulty can be bigger, it is desirable to by a certain individual
All pictures all to find out difficulty bigger, so, some during cellphone subscriber checks mobile phone for convenience
People or the photo of some people, first can choose all of personage's picture from mobile phone picture library, and will choose
Personage's picture is classified, and is stored respectively in different storage positions (the most different files), user
Corresponding file can be searched according to the actual requirements to browse the picture (photo) of related person.In implementing,
Specific characteristic described in the present embodiment is mainly character facial feature, wherein, above-mentioned character facial feature
Comprise the steps that character facial profile, character facial five official rank, can by identify personage face mask or
Five official rank features judge whether this picture is personage's picture.Concrete, face recognition technology can be passed through, from
All pictures of mobile phone picture library are chosen and there is the picture of specific characteristic (i.e. there is the personage of character facial feature
Picture), and these pictures are separately deposited to designated storage location (such as new folder etc.).In implementing,
Personage's picture in mobile phone picture library potentially includes the picture of multiple personage, such as, it may include first, second, third,
The photo of many individuals such as fourth, so after have chosen personage's picture from all pictures of mobile phone picture library, also may be used
Identified further by face recognition technology, judge, the personage's picture chosen is carried out according to different personages
Classification, storage.
In the embodiment that some are feasible, after have chosen personage's picture from mobile phone picture library, then can calculate
The similarity of each personage's picture in the personage's picture chosen, also can preset multiple (the most N number of,
N is the natural number more than 1) predetermined threshold value of similarity, and the predetermined threshold value according to default similarity will
Similarity is divided into N number of level, and then passes through the similarity between picture by each personage's picture according to similarity
Different layers level classify, and store respectively to different storage positions.In implementing, mobile phone calculates
First can choose any one from all of personage's picture during the similarity of each personage's picture got (to include
Any one pictures of designated person) as reference picture, according to the character facial feature of reference picture and
Each character facial feature of character facial feature calculation of other personage's pictures (i.e. each non-reference picture)
Similarity, then the similarity of each character facial feature comprehensive determines reference picture and each non-reference picture
Similarity.Wherein, the similarity of above-mentioned character facial feature comprises the steps that the phase of ratio of character facial profile
Like degree, character facial face shape, size, the similarity etc. of ratio, such as, between personage two away from
From, the height of nose, width, the degree of depth of eye socket, the height of cheekbone, profile of lower jaw and lower jaw etc. each
The similarity of character facial feature.Such as, the predetermined threshold value of multiple personage's picture similarity can be preset,
And with any one pictures in all personage's pictures as reference, calculate other personage's pictures similar to this picture
Degree, and according to the result of calculation of similarity, the similarity with this picture is reached the first predetermined threshold value and (be assumed to be
The value that similarity is the highest) all personage's pictures store to first specify position (this picture also can be given tacit consent to and deposit
Store up to this first appointment position), all personage's pictures of K predetermined threshold value will be reached with the similarity of this picture
Storing and specify position to K, wherein, K is the natural number less than N, can set the predetermined threshold value of similarity
In the similarity of the first predetermined threshold value the highest, the similarity of N predetermined threshold value is minimum, such sort descending.
Concrete, above-mentioned first specifies position to specify position can be specifically some or multiple file to K.
Such as, a total of 40 of the personage's picture obtained from mobile phone picture library, 10, the photo including first is (false
It is set to numbering 1-10), 10, the photo of second (being assumed to be numbering 11-20), 10, the photo of third (is assumed to be
Numbering 21-30), 10, the photo of fourth (being assumed to be numbering 31-40).Mobile phone get these personage's pictures it
After, with the picture of numbering 1 as reference, calculate the similarity of other personage's pictures and this picture, if judging
Know that 9 personage's pictures of numbering 2-10 and the similarity of this picture reach above-mentioned first predetermined threshold value (similarity
It is the first level), then can determine whether to learn that these 10 personage's pictures are picture (the i.e. first with same characteristic features
Photo), and then the file 1(that can be stored to mobile phone picture library by these 10 personage's pictures is the newest
The file built or default file), when user wants the photo browsing first, then can directly open file
Folder 1.In like manner, can by the calculating of the similarity of personage's picture, if judge to learn the picture of numbering 11-20 and
The similarity of the picture of numbering 1 reaches the second predetermined threshold value (similarity is the second level), then can will number
Picture (photo of the second) classification of 11-20 stores the file 2 to mobile phone picture library；If judging to learn numbering
The picture (photo of third) of 21-30 reaches the 3rd predetermined threshold value (similarity with the similarity of the picture of numbering 1
For third layer level), then the picture classification of numbering 21-30 can be stored the file 3 to mobile phone picture library；If
Judge that the picture (photo of third) learning numbering 31-40 and the similarity of the picture of numbering 1 reach the 4th default
Threshold value (similarity is the 4th level), then can store the picture classification of numbering 31-40 to mobile phone picture library
File 4, and so on by all of personage's picture according to the different layers level of similarity carry out classifying and store to
Different picture storage positions, user can open corresponding file according to the demand of oneself and browse related person
Photo.
S103, when on the storage position get the picture that similarity is K level, the classification of the first picture is adjusted
During whole instruction, adjust instruction according to described classification and described first picture is stored the picture storage to corresponding level
Position.
In the embodiment that some are feasible, when mobile phone calculates the similarity of each personage's picture, probably due to
A certain pictures or the reason such as plurality of pictures is unintelligible or picture processing program is made mistakes so that the phase of picture
Make mistakes like degree judgement, thus a certain pictures or plurality of pictures (the i.e. first picture) are stored to mistake
Storage position.Such as, user is handshaking when shooting photo (the such as numbering 11,12 etc.) of second makes to shine
Sheet shooting is unintelligible, or during shooting, angle is not provided with so that the character facial None-identified in photo,
Or because the hair style of second has been blocked face and made the photo of second cannot the reason such as identification well during shooting so that
When the photo of second is carried out Similarity Measure, wrong by the similarity of the photo (numbering 11,12) of second and first
The similarity of photo is judged to that similarity reaches the first predetermined threshold value (similarity is the first level), and then by second
The storage position (file 1) that stores to first of photo in, now, the photo (numbering 13-20) of second with
The similarity of the photo (numbering 11,12) of second is judged as similarity and reaches the second predetermined threshold value.User exists
When browsing photo, then can find the photo of the second occurred in the storage position (file 1) of the photo of first, this
Time, user then needs to second 2 photos (numbering 11,12) to carry out classification adjustment, by these 2 photos
Store the storage position (file 2) of the photo to second.In implementing, picture classification can be preset and adjust
Whole edit pattern, and set corresponding edit operation button and (physical button, virtual key, cunning can be included
Dynamic button), i.e. when having the user discover that a certain or plurality of pictures (the i.e. first picture) sorts out and staggers the time, then
The optional edit pattern (can click on or long by the picture entrance made mistakes) entering picture classification adjustment, and point
Hit corresponding edit operation button (including slip mobile phone screen) and this picture is carried out classification adjustment, by this figure
Sheet stores to correct storage position.Such as, arrive when the photo (numbering 11,12) that the user discover that second stores
During the file of the photo of first, then optional enter the edit pattern that picture classification adjusts, concrete, can point
Hit or length enters into picture classification adjustment modes by this picture (photo (numbering 11,12) of second), and select
Select edit key corresponding to this picture and send the classification adjustment instruction of this picture to embedded in mobile phone processing module.Hands
The built-in processing module of machine gets the classification of the photo (numbering 11,12) of second in file 1 and adjusts instruction
After, then this picture can be automatically adjusted, it is not necessary to user is manually adjusted.In implementing, above-mentioned
It can be the glide direction that terminal use controls when the first picture moves that classification adjusts instruction, can be according to the cunning of user
Picture is stored to storing position accordingly by dynamic direction.Concrete, it is K level that mobile phone gets similarity
Picture storage folder on one or plurality of pictures classification adjust instruction time, if according to getting
Classification adjust instruction judgement learn when glide direction is first direction, then can be by above-mentioned one or multiple figures
Sheet stores to similarity that (such as K-1 level, wherein K level is than K-1 layer than K level high-level
Level a low level) picture storage position, wherein, above-mentioned first direction can include being parallel to mobile phone screen
To the direction at mobile phone top, or it is parallel to mobile phone screen and is perpendicular to mobile phone top to the right etc..If judging to learn
Glide direction is second direction, then can store to similarity above-mentioned one or plurality of pictures than K level
The picture storage of low-level (such as K+1 level, wherein a K level level higher than K+1 level)
Position, wherein, above-mentioned second direction can include being parallel to the mobile phone screen direction to mobile phone end part, or flat
Row is perpendicular to mobile phone top to the left etc. in mobile phone screen.Such as, when personage's picture memory error, mobile phone obtains
After the classification of the photo (numbering 11,12) getting above-mentioned second adjusts instruction, it is judged that learn user's slip figure
The glide direction of sheet is above-mentioned second direction, then can be stored by this picture (photo (numbering 11,12) of second)
(file 2, with the picture that similarity is the second level of the picture of numbering 1 to specify position to the most above-mentioned second
Storage position), it is not necessary to user carries out manual classification adjustment.In implementing, it is that mobile phone gets similarity
When the classification of in the storage folder of the picture of K level or plurality of pictures adjusts instruction, it is possible to straight
Connect and this pictures or plurality of pictures are stored to a level lower than this similarity level (or high level)
The storage position of picture, such as K+1 level (a K level level higher than K+1 level).
In implementing, when personage's picture memory error, mobile phone gets the photo (numbering 11,12) of above-mentioned second
Classification adjust instruction after, due to the photo (numbering 11,12) of second current store position be the photograph of first
The storage position of the sheet storage position of the picture that similarity is the first level of the picture of numbering 1 (i.e. with), therefore
This, then can be directly by this figure after the classification of the picture of mobile phone gets above-mentioned numbering 11,12 adjusts instruction
Sheet (photo (numbering 11,12) of second) stores to above-mentioned second specifies that (file 2, with numbering 1 in position
The storage position of the picture that similarity is the second level of picture), it is not necessary to user carries out manual classification adjustment.
If additionally, in above-mentioned file 1 the third of the numbering 21,22 of the picture of memory error photo time, according to
The picture of numbering 21,22 is adjusted after specifying position (i.e. file 2) to second and finds also by said method
Being wrong, user can again tap on or length is entered into by this picture (photo (numbering 11,12) of second)
Picture classification adjustment modes, and select the edit key that this picture is corresponding to be somebody's turn to do to the transmission of embedded in mobile phone processing module
The classification of picture adjusts instruction, and mobile phone gets corresponding after adjusting instruction, then can be by numbering 21,22
Picture continues to move backward, stores to the similarity with the picture of numbering 1 and reaches the picture (third of third layer level
Photo) storage position (file 3).The like, can be by picture corresponding for each similarity level
The picture of storage position memory error is adjusted, until being stored by picture to correct storage position.Specifically
, when personage's picture memory error, when needing to carry out classification adjustment, mobile phone gets the classification of certain picture and adjusts
After whole instruction, it is possible to representative picture list corresponding for the similarity level of picture is exported to current user
Browser interface, user can select a picture from this list, and adjust the picture of memory error to this figure
The picture storage position that sheet is corresponding.Wherein, above-mentioned representative picture is that each similarity level correspondence picture is deposited
Storage space reaches the picture of highest level, above-mentioned representative with the similarity of most pictures in this storage position in putting
Just list is that the representative picture of each similarity level is carried out according to the similarity hierarchic sequence with numbering 1
The list rearranged.I.e. cellphone subscriber can directly select the similarity with this picture according to representative picture list
The highest picture (naked eyes judgement), and the storage position of the picture the highest with the similarity of this picture is chosen to be
Target storage position, and then this picture is stored to this target storage position.
The method of the picture classification described in the present embodiment can will be from mobile phone figure according to the similarity of personage's picture
The personage's picture chosen in storehouse is classified, and is classified according to different people object plane hole by personage's picture, convenient
User browses, searches, and the figure made mistakes classification in the similarity sorted out between staggering the time according to personage's picture
Sheet is classified adjustment automatically, it is not necessary to user manually adjusts, and can reduce personage's picture classification and classification
Adjust workload, improve personage's picture classification and regulated efficiency, improve mobile phone user picture
Convenience.
See Fig. 2, be the example structure schematic diagram of the terminal that the embodiment of the present invention provides.Institute in the present embodiment
The terminal described, including:
Determine module 20, for determining the similarity of reference picture and the non-reference picture with specific characteristic,
And described similarity is divided into N number of level.
Sort module 30, for storing described non-reference picture respectively to not according to the different layers level of similarity
Same picture storage position.
Adjusting module 40, for the first figure on the storage position getting the picture that similarity is K level
When the classification of sheet adjusts instruction, adjust instruction according to described classification and described first picture is stored to corresponding level
Picture storage position.
In the embodiment that some are feasible, the terminal (such as Fig. 3) described in the present embodiment, also include:
Choose module 10, for choosing from all pictures of terminal primary importance storage, there is specific characteristic
Picture.
In the embodiment that some are feasible, above-mentioned determine module 20 determining have specific characteristic with reference to figure
During the similarity of sheet and non-reference picture, specifically for:
According to each the non-reference picture of character facial feature calculation in described each picture with specific characteristic
The similarity of character facial feature of character facial feature and reference picture, it is comprehensive that described each is non-with reference to figure
Sheet determines each non-reference picture described and described ginseng with the similarity of the character facial feature of described reference picture
The similarity of photograph and picture.
In the embodiment that some are feasible, above-mentioned classification adjusts instruction and controls described for described terminal use
Glide direction when one picture moves, above-mentioned adjusting module 40 is adjusting instruction by described the according to described classification
When one picture stores the picture storage position to corresponding level, specifically for:
When described glide direction is first direction, described first picture is stored to similarity than described K
The picture storage position of level high-level；
When described glide direction is second direction, described first picture is stored to similarity than described K
The picture storage position of level low-level.
In the embodiment that some are feasible, the picture in the present embodiment with specific characteristic can be to include
The picture of character facial feature, can also be the picture etc. of picture or the specific background comprising particular scene.
Character facial feature comprises the steps that character facial profile, character facial five official rank；Particular scene can be flower or
Other plant of person, animal or object, as a example by flower, comprise colored color, profile etc..
Terminal primary importance concretely hands in the embodiment that some are feasible, described in the present embodiment
Machine picture library, when in the mobile phone picture library of user, the picture number of storage is huger, when the kind of picture is more,
When user searches the photo of a certain individual or some people in picture library, difficulty can be bigger, it is desirable to by a certain individual
All pictures all to find out difficulty bigger, so, some during cellphone subscriber checks mobile phone for convenience
People or the photo of some people, first can choose all of personage's picture from mobile phone picture library, and will choose
Personage's picture is classified, and is stored respectively in different storage positions (the most different files), user
Corresponding file can be searched according to the actual requirements to browse the picture (photo) of related person.In implementing,
Specific characteristic described in the present embodiment is mainly character facial feature, wherein, above-mentioned character facial feature
Comprise the steps that character facial profile, character facial five official rank, can by identify personage face mask or
Five official rank features judge whether this picture is personage's picture.Concrete, can first pass through and choose module 10 from hands
Machine picture library is chosen personage's picture, concrete, choose module 10 and can pass through face recognition technology, from mobile phone figure
All pictures in storehouse are chosen the picture (i.e. having personage's picture of character facial feature) with specific characteristic,
And these pictures are separately deposited to designated storage location (such as new folder etc.).In implementing, mobile phone figure
Personage's picture in storehouse potentially includes the picture of multiple personage, such as, it may include first, second, third, Ding Dengduo
The photo of individual, so mobile phone have chosen figure map by choosing module 10 from all pictures of mobile phone picture library
After sheet, identify further also by face recognition technology, judge, by personage's picture of choosing according to
Different personages carry out classifying, storing.In implementing, choose how module 10 chooses figure map from picture library
Step in the embodiment of the method that the process that implements of sheet can be found in the picture classification that the embodiment of the present invention provides
Rapid S101, does not repeats them here.
In the embodiment that some are feasible, choose after module 10 have chosen personage's picture from mobile phone picture library,
Determine that module 20 then can calculate the similarity of each personage's picture in the personage's picture chosen, also can set in advance
The predetermined threshold value of fixed multiple (the most N number of, N is the natural number more than 1) similarity, and according to default
Similarity is divided into N number of level by the predetermined threshold value of similarity, and sort module 30 then can be by the phase between picture
Like degree by each personage's picture according to the different levels of similarity, and store respectively to different storage positions.
In implementing, mobile phone can be determined by can when module 20 calculates the similarity of each personage's picture got
First choose any one (including any one pictures of designated person) as reference from all of personage's picture
Picture, according to character facial feature and other personage's pictures (i.e. each non-reference picture) of reference picture
The similarity of each character facial feature of character facial feature calculation, more comprehensive each character facial feature
Similarity determines the similarity of reference picture and each non-reference picture.Wherein, above-mentioned character facial feature
Similarity comprises the steps that the similarity of the ratio of character facial profile, character facial face shape, size, ratio
The similarity etc. of example, such as, the distance between personage two, the height of nose, width, the degree of depth of eye socket,
The similarity of each character facial features such as the profile of the height of cheekbone, lower jaw and lower jaw.Such as, mould is determined
Block 20 can preset the predetermined threshold value of multiple personage's picture similarity, and with any one in all personage's pictures
Pictures is reference, calculates the similarity of other personage's pictures and this picture, and sort module 30 then can be according to phase
Like the result of calculation spent, the similarity with this picture is reached the first predetermined threshold value and (be assumed to be similarity the highest
Value) all personage's pictures store to first specify position (can by this picture also default storage to this first refer to
Location is put), all personage's pictures reaching K predetermined threshold value with the similarity of this picture are stored to K
Specifying position, wherein, K is the natural number less than N, can set in the predetermined threshold value of similarity first and preset
The similarity of threshold value is the highest, and the similarity of N predetermined threshold value is minimum, such sort descending.Concrete, on
Stating the first appointment position specifies position can be specifically some or multiple file to K.Such as, choose
A total of 40 of personage's picture that module 10 obtains from mobile phone picture library, 10, the photo including first is (false
It is set to numbering 1-10), 10, the photo of second (being assumed to be numbering 11-20), 10, the photo of third (is assumed to be
Numbering 21-30), 10, the photo of fourth (being assumed to be numbering 31-40).Mobile phone obtains by choosing module 10
After these personage's pictures, then can be determined by the module 20 picture with numbering 1 as reference, calculate other
Personage's picture and the similarity of this picture, and then by sort module 30, these personage's pictures are sorted out.
If sort module 30 judges to learn 9 personage's pictures of numbering 2-10 and the similarity of this picture reaches above-mentioned the
One predetermined threshold value (similarity is the first level), then can determine whether to learn that these 10 personage's pictures are and have phase
With the picture (i.e. the photo of first) of feature, and then these 10 personage's pictures can be stored to mobile phone picture library
File that file 1(is the most newly-built or default file), user wants to browse the photo of first
Time, then can direct opened file folder 1.In like manner, sort module 30 passes through the judgement of the similarity of personage's picture,
If it is (similar to judge to learn that the picture of numbering 11-20 reaches the second predetermined threshold value to the similarity of the picture of numbering 1
Degree is the second level), then the picture (photo of second) of numbering 11-20 can be classified and store to mobile phone picture library
File 2；If judging to learn that the picture (photo of third) of numbering 21-30 is similar to the picture of numbering 1
Degree reaches the 3rd predetermined threshold value (similarity is third layer level), then can be stored by the picture classification of numbering 21-30
File 3 to mobile phone picture library；If judging the picture (photo of third) learning numbering 31-40 and numbering 1
The similarity of picture reach the 4th predetermined threshold value (similarity is the 4th level), then can be by numbering 31-40
Picture classification stores the file 4 to mobile phone picture library, and so on by all of personage's picture according to similarity
Different layers level carry out classifying and store to different picture storage positions, user can beat according to the demand of oneself
Open corresponding file and browse the photo of related person.In implementing, determine that how module is to choosing module
The similarity of the personage's picture chosen from picture library carries out counting, how sort module is to choosing module from picture library
In personage's picture of choosing process that implements of carrying out classifying picture of can be found in the embodiment of the present invention and providing divide
Step S101-S102 in the embodiment of the method for class, does not repeats them here.
In the embodiment that some are feasible, when determining the similarity that module 20 calculates each personage's picture, can
Can be because a certain pictures or the reason such as plurality of pictures is unintelligible or picture processing program is made mistakes so that figure
The similarity of sheet judges to make mistakes, thus a certain pictures or plurality of pictures (the i.e. first picture) are stored to
The storage position of mistake.Such as, user shoots during photo (the such as numbering 11,12 etc.) of second handshaking
Make photograph taking unintelligible, or during shooting, angle is not provided with so that the character facial in photo cannot
Identify, or during shooting because the hair style of second blocked face make second photo cannot the reason such as identification well,
Make to determine when module 20 carries out Similarity Measure to the photo of second, wrong by the photo (numbering 11,12) of second
The similarity of photo of similarity and first be judged to similarity reach the first predetermined threshold value (similarity be first
Level) so that in the storage position (file 1) that the photo of second is stored to first by sort module 30, now,
The similarity of the photo (numbering 13-20) of second and the photo (numbering 11,12) of second is judged as similarity and reaches
To the second predetermined threshold value.User when browsing photo, then can find the storage position (file 1) of the photo of first
The photo of the second of middle appearance, now, user then needs to second 2 photos (numbering 11,12) to carry out point
Class adjusts, and these 2 photos store the storage position (file 2) of the photo to second.In implementing,
The storage position of the picture that sort module 30 classification makes mistakes can be adjusted by mobile phone by adjusting module 40,
Concrete, adjusting module 40 can preset the edit pattern that picture classification adjusts, and sets corresponding editor
Operation button (physical button, virtual key, slip button can be included), i.e. when the user discover that a certain or
Person's plurality of pictures (i.e. first personage's picture) sorts out and staggers the time, then the optional volume entering picture classification adjustment
Volume pattern (can click on or long enter by the picture made mistakes), and click on corresponding edit operation button and (include
Slip mobile phone screen) this picture is carried out classification adjustment, this picture is stored to correct storage position.Example
As, when the photo (numbering 11,12) that the user discover that second stores the file of the photo of first, the most optional
Select and enter the edit pattern that picture classification adjusts, concrete, can click on or long by this picture (photo of second
(numbering 11,12)) enter into picture classification adjustment modes, and select edit key that this picture is corresponding to hands
The built-in processing module of machine sends the classification of this picture and adjusts instruction.Adjusting module 40 gets second in file 1
Photo (numbering 11,12) classification adjust instruction after, then this picture can be automatically adjusted, nothing
User is needed manually to be adjusted.In implementing, it can be that terminal use controls first that above-mentioned classification adjusts instruction
Glide direction when picture moves, can store picture to storing position accordingly according to the glide direction of user.
Concrete, the adjusting module 40 of mobile phone gets in the storage folder of the picture that similarity is K level
When the classification of or plurality of pictures adjusts instruction, learn if adjusting instruction judgement according to the classification got
When glide direction is first direction, then above-mentioned one or plurality of pictures can be stored to similarity than K layer
The picture of level high-level (such as K-1 level, wherein a K level level lower than K-1 level) is deposited
Storage space is put, and wherein, above-mentioned first direction can include the direction being parallel to mobile phone screen to mobile phone top, or
It is parallel to mobile phone screen and is perpendicular to mobile phone top to the right etc..If judging to learn that glide direction is second direction, then
Above-mentioned one or plurality of pictures can be stored to similarity than K level low-level (such as K+1 level,
A wherein K level level higher than K+1 level) picture storage position, wherein, above-mentioned second party
To including being parallel to the mobile phone screen direction to mobile phone end part, or it is parallel to mobile phone screen and is perpendicular to mobile phone
Top is to the left etc..Such as, when sort module 30 carries out classifying memory error to personage's picture, adjusting module
The classification of 40 photos (numbering 11,12) getting above-mentioned second judges to learn that user slides after adjusting instruction
The glide direction of picture is above-mentioned first direction, then can this picture (photo (numbering 11,12) of second) be deposited
Storage specifies position to above-mentioned second, and (file 2, with the picture that similarity is the second level of the picture of numbering 1
Storage position), it is not necessary to user carries out manual classification adjustment.In implementing, mobile phone gets similarity and is
When the classification of in the storage folder of the picture of K level or plurality of pictures adjusts instruction, it is possible to
Directly this pictures or plurality of pictures are stored to a level lower than this similarity level (or high level)
The storage position of picture, such as K+1 level (a K level level higher than K+1 level).
In implementing, when sort module 30 carries out classifying memory error to personage's picture, adjusting module 40 obtains
After the classification of the photo (numbering 11,12) getting above-mentioned second adjusts instruction, due to the photo (numbering of second
11,12) current storage position is that the storage position of photo of first is (i.e. with the similarity of the picture of numbering 1
It is the storage position of the picture of the first level), so when adjusting module 40 gets above-mentioned numbering 11,12
After the classification of picture adjusts instruction, then can directly this picture (photo (numbering 11,12) of second) be stored
(file 2, with the picture that similarity is the second level of the picture of numbering 1 to specify position to the most above-mentioned second
Storage position), it is not necessary to user carries out manual classification adjustment.If additionally, memory error in above-mentioned file 1
Picture numbering 21,22 third photo time, adjusting module 40 picture of numbering 21,22 is adjusted to
Second specify position (i.e. file 2) after the user discover that or mistake, then can again tap on or long by
This picture (photo (numbering 11,12) of second) enters into picture classification adjustment modes, and selects this picture pair
The classification that the edit key answered sends this picture to embedded in mobile phone processing module adjusts instruction, the adjustment mould of mobile phone
Block 40 gets corresponding after adjusting instruction, then can continue to move by the picture of numbering 21,22 backward, deposit
The similarity of the picture of Chu Zhiyu numbering 1 reaches the storage position (literary composition of the picture (photo of third) of third layer level
Part folder 3).The like, can be by the figure of picture storage position memory error corresponding for each similarity level
Sheet is adjusted, until being stored by picture to correct storage position.Concrete, when personage's picture stores out
Mistake, when needing to carry out classification adjustment, after adjusting module 40 gets the classification adjustment instruction of certain picture, also
Representative picture list corresponding for the similarity level of picture can be exported to current user's browser interface, user
Can select a picture from this list, storage is also gone out by picture that adjusting module 40 then can be selected according to user
Wrong picture adjusts to picture storage position corresponding to this picture.Wherein, above-mentioned representative picture is each phase
Like in degree level correspondence picture storage position, the similarity of most pictures reaches top with this storage position
Level picture, above-mentioned representative picture list be each similarity level representative picture according to numbering 1
Similarity hierarchic sequence carries out the list rearranged.I.e. cellphone subscriber directly can select according to representative picture list
Select the picture (naked eyes judgement) the highest with the similarity of this picture, and will be with this picture by adjusting module 40
The storage position of the highest picture of similarity be chosen to be target storage position, and then this picture is stored to this
Target storage position.In implementing, how adjusting module is by the personage's picture after sort module classification storage
The process that implements that is adjusted of storage position can be found in the side of the picture classification that the embodiment of the present invention provides
Step S103 in the embodiment of method, does not repeats them here.
The people that described in the present embodiment, terminal will can be chosen according to the similarity of personage's picture from mobile phone picture library
Thing picture is classified, and is classified according to different people object plane hole by personage's picture, facilitates user to browse, looks into
Look for, and the picture made mistakes classification in the similarity sorted out between staggering the time according to personage's picture divides automatically
Class adjusts, it is not necessary to user manually adjusts, and can reduce personage's picture classification and the workload of classification adjustment,
Improve personage's picture classification and regulated efficiency, improve the convenience of mobile phone user picture.
Step in embodiment of the present invention method can carry out order according to actual needs and adjust, merges and delete.
Module or unit in embodiment of the present invention device can merge according to actual needs, divide and delete
Subtract.
Module in all embodiments of the invention or unit, can pass through universal integrated circuit, such as CPU
(Central Processing Unit, central processing unit), or by ASIC (Application Specific
Integrated Circuit, special IC) realize.
One of ordinary skill in the art will appreciate that all or part of flow process realizing in above-described embodiment method,
Can be by computer program and complete to instruct relevant hardware, described program can be stored in computer
In read/write memory medium, this program is upon execution, it may include such as the flow process of the embodiment of above-mentioned each method.
Wherein, described storage medium can be magnetic disc, CD, read-only store-memory body (Read-Only Memory,
Or random store-memory body (Random Access Memory, RAM) etc. ROM).
Above disclosed be only present pre-ferred embodiments, certainly can not with this limit the present invention it
Interest field, the equivalent variations therefore made according to the claims in the present invention, still belong to the scope that the present invention is contained.
Claims (10)
1. the method for a picture classification, it is characterised in that including:
Determine the similarity of reference picture and the non-reference picture with specific characteristic, and described similarity is divided
For N number of level, wherein N is the natural number more than 1；
Described non-reference picture is stored to different picture storage positions respectively according to the different layers level of similarity；
When on the storage position get the picture that similarity is K level, the classification of the first picture adjusts instruction
Time, adjust instruction according to described classification and determine the classification adjustment direction corresponding with described classification adjustment instruction, root
According to described classification adjustment direction, described first picture is stored to corresponding level according to described classification adjustment direction
Picture storage position, wherein K is the natural number less than N, and described first picture is the institute that terminal use chooses
State any one or plurality of pictures on the storage position of the picture that similarity is K level.
2. the method for claim 1, it is characterised in that described determine the reference with specific characteristic
Before the similarity of picture and non-reference picture, also include:
The picture with specific characteristic is chosen from all pictures of terminal primary importance storage.
3. method as claimed in claim 2, it is characterised in that described specific characteristic is personage's face feature；
Described character facial feature, including: in character facial profile, character facial face at least one.
4. method as claimed in claim 3, it is characterised in that described determine the reference with specific characteristic
Picture and the similarity of non-reference picture, including:
According to each the non-reference picture of character facial feature calculation in described each picture with specific characteristic
The similarity of character facial feature of character facial feature and reference picture；
Comprehensive each non-reference picture described determines institute with the similarity of the character facial feature of described reference picture
State the similarity of each non-reference picture and described reference picture；
Wherein, the similarity of described character facial feature, including: the similarity of the ratio of character facial profile,
Or character facial face shape, size or the similarity of ratio.
5. the method as described in claim 1-4 any one, it is characterised in that described classification adjusts instruction
Glide direction when described first picture moves is controlled for described terminal use, described according to described classification adjustment
Described first picture is stored the picture storage position to corresponding level by instruction, including:
When described glide direction is first direction, described first picture is stored to similarity than described K
The picture storage position of level high-level；
When described glide direction is second direction, described first picture is stored to similarity than described K
The picture storage position of level low-level.
6. a terminal, it is characterised in that including:
Determine module, for determining the similarity of reference picture and the non-reference picture with specific characteristic, and
Described similarity is divided into N number of level, and wherein N is the natural number more than 1；
Sort module, for storing described non-reference picture respectively to difference according to the different layers level of similarity
Picture storage position；
Adjusting module, for the first picture on the storage position getting the picture that similarity is K level
Classification when adjusting instruction, adjust instruction according to described classification and determine and described classification adjusts that instruction is corresponding to divide
Class adjustment direction, deposits described first picture according to described classification adjustment direction according to described classification adjustment direction
Storing up the picture storage position to corresponding level, wherein K is the natural number less than N, and described first picture is eventually
The picture that described similarity is K level that end subscriber is chosen storage position on any one or multiple
Picture.
7. terminal as claimed in claim 6, it is characterised in that described terminal, also includes:
Choose module, for choosing the figure with specific characteristic from all pictures of terminal primary importance storage
Sheet.
8. terminal as claimed in claim 7, it is characterised in that described specific characteristic is personage's face feature；
Described character facial feature, including: in character facial profile, character facial face at least one.
9. terminal as claimed in claim 8, it is characterised in that described determine that module is determining have appointment
During the similarity of the reference picture of feature and non-reference picture, specifically for:
According to each the non-reference picture of character facial feature calculation in described each picture with specific characteristic
The similarity of character facial feature of character facial feature and reference picture, it is comprehensive that described each is non-with reference to figure
Sheet determines each non-reference picture described and described ginseng with the similarity of the character facial feature of described reference picture
The similarity of photograph and picture；
Wherein, the similarity of described character facial feature, including: the similarity of the ratio of character facial profile,
Or character facial face shape, size or the similarity of ratio.
10. the terminal as described in claim 6-9 any one, it is characterised in that described classification adjusts instruction
Controlling glide direction when described first picture moves for described terminal use, described adjusting module is according to institute
When stating the picture storage position that described first picture is stored to corresponding level by classification adjustment instruction, specifically for:
When described glide direction is first direction, described first picture is stored to similarity than described K
The picture storage position of level high-level；
When described glide direction is second direction, described first picture is stored to similarity than described K
The picture storage position of level low-level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310440197.6A CN103488756B (en) | 2013-09-25 | 2013-09-25 | Picture classification method and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310440197.6A CN103488756B (en) | 2013-09-25 | 2013-09-25 | Picture classification method and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103488756A CN103488756A (en) | 2014-01-01 |
| CN103488756B true CN103488756B (en) | 2017-01-11 |
Family
ID=49828982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310440197.6A Active CN103488756B (en) | 2013-09-25 | 2013-09-25 | Picture classification method and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103488756B (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955481B (en) * | 2014-04-03 | 2018-04-24 | 小米科技有限责任公司 | image display method and device |
| WO2016000079A1 (en) * | 2014-07-02 | 2016-01-07 | BicDroid Inc. | Display, visualization, and management of images based on content analytics |
| CN105446637A (en) * | 2014-08-29 | 2016-03-30 | 宇龙计算机通信科技(深圳)有限公司 | Image storage method and system and terminal |
| CN105654101B (en) * | 2014-11-11 | 2019-04-26 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
| CN104461359A (en) * | 2014-11-28 | 2015-03-25 | 深圳市金立通信设备有限公司 | Terminal |
| CN104484107A (en) * | 2014-11-28 | 2015-04-01 | 深圳市金立通信设备有限公司 | Picture processing method |
| CN104536659B (en) * | 2014-12-15 | 2018-06-12 | 小米科技有限责任公司 | Object information processing method and device |
| US9911173B2 (en) | 2014-12-30 | 2018-03-06 | Xiaomi Inc. | Method and device for displaying images |
| CN104898944B (en) * | 2015-05-21 | 2017-11-24 | 广东欧珀移动通信有限公司 | A kind of photo processing method and terminal |
| CN106326908A (en) * | 2015-06-30 | 2017-01-11 | 中兴通讯股份有限公司 | Picture management method and apparatus, and terminal equipment |
| CN105183338B (en) * | 2015-07-20 | 2019-02-01 | 西安拾忆信息科技有限公司 | Object processing method |
| CN107169367A (en) * | 2017-04-07 | 2017-09-15 | 北京安云世纪科技有限公司 | A kind of storage method, device and mobile terminal for shooting file |
| CN108875797B (en) * | 2018-05-29 | 2023-04-18 | 腾讯科技（深圳）有限公司 | Method for determining image similarity, photo album management method and related equipment |
| CN110390029A (en) * | 2019-06-17 | 2019-10-29 | 宁波乐印文化有限公司 | A kind of photo classification method and system |
| CN111581419B (en) * | 2020-04-29 | 2023-08-01 | 北京金山云网络技术有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN111666262B (en) * | 2020-05-28 | 2021-06-22 | 重庆中联信息产业有限责任公司 | Working method for extracting feature points of massive medical images in network attached storage NAS state |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1770814A (en) * | 2004-10-02 | 2006-05-10 | Lg电子株式会社 | Method and apparatus for storing image files of mobile communication terminal |
| CN101009746A (en) * | 2006-01-26 | 2007-08-01 | 三星Techwin株式会社 | Method of controlling digital image processing apparatus and digital image processing apparatus |
| CN101482877A (en) * | 2008-01-07 | 2009-07-15 | 船井电机株式会社 | Content display apparatus for displaying media according to media categories |
-
2013
- 2013-09-25 CN CN201310440197.6A patent/CN103488756B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1770814A (en) * | 2004-10-02 | 2006-05-10 | Lg电子株式会社 | Method and apparatus for storing image files of mobile communication terminal |
| CN101009746A (en) * | 2006-01-26 | 2007-08-01 | 三星Techwin株式会社 | Method of controlling digital image processing apparatus and digital image processing apparatus |
| CN101482877A (en) * | 2008-01-07 | 2009-07-15 | 船井电机株式会社 | Content display apparatus for displaying media according to media categories |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103488756A (en) | 2014-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103488756B (en) | Picture classification method and terminal | |
| US10140515B1 (en) | Image recognition and classification techniques for selecting image and audio data | |
| US9329762B1 (en) | Methods and systems for reversing editing operations in media-rich projects | |
| CN110476405A (en) | For providing and shooting the method and system of related recommendation information | |
| US20130058579A1 (en) | Image information processing apparatus | |
| CN108875797B (en) | Method for determining image similarity, photo album management method and related equipment | |
| CN101706793A (en) | Method and device for searching picture | |
| US10719695B2 (en) | Method for pushing picture, mobile terminal, and storage medium | |
| CN111444366B (en) | Image classification method, device, storage medium and electronic equipment | |
| US11030236B2 (en) | Image searching by employing layered search constraints | |
| CN104915634A (en) | Image generation method based on face recognition technology and apparatus | |
| CN106202189A (en) | A kind of image search method and device | |
| CN107153708A (en) | A kind of picture inspection method and device, computer installation, computer-readable recording medium | |
| CN101515271A (en) | Method and system for photo management, photo classification code-storing computer-readable media | |
| CN106294549A (en) | A kind of image processing method and terminal | |
| CN111488477A (en) | Album processing method, apparatus, electronic device and storage medium | |
| CN106096043A (en) | A kind of photographic method and mobile terminal | |
| US20130301938A1 (en) | Human photo search system | |
| US9582160B2 (en) | Semi-automatic organic layout for media streams | |
| Ma et al. | SmartEye: assisting instant photo taking via integrating user preference with deep view proposal network | |
| JP2014092955A (en) | Similar content search processing device, similar content search processing method and program | |
| CN103260036B (en) | Image processing apparatus, image processing method, storage medium and image processing system | |
| CN105824859A (en) | Picture classification method and device as well as intelligent terminal | |
| CN107391608B (en) | Picture display method and device, storage medium and electronic equipment | |
| CN108921798A (en) | The method, apparatus and electronic equipment of image procossing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PP01 | Preservation of patent right |
Effective date of registration: 20180428 Granted publication date: 20170111 |
|
| PP01 | Preservation of patent right | ||
| PD01 | Discharge of preservation of patent |
Date of cancellation: 20210428 Granted publication date: 20170111 |
|
| PD01 | Discharge of preservation of patent | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210608 Address after: 518057 Desai Science and Technology Building, 9789 Shennan Avenue, Yuehai Street, Nanshan District, Shenzhen City, Guangdong Province, 17th Floor (15th Floor of Natural Floor) 1702-1703 Patentee after: Shenzhen Microphone Holdings Co.,Ltd. Address before: 518040 21 floor, east block, Times Technology Building, 7028 Shennan Road, Futian District, Shenzhen, Guangdong. Patentee before: DONGGUAN GOLDEX COMMUNICATION TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |