CN102982572B - A kind of intelligence image edit method and device - Google Patents
A kind of intelligence image edit method and device Download PDFInfo
- Publication number
- CN102982572B CN102982572B CN201210428871.4A CN201210428871A CN102982572B CN 102982572 B CN102982572 B CN 102982572B CN 201210428871 A CN201210428871 A CN 201210428871A CN 102982572 B CN102982572 B CN 102982572B
- Authority
- CN
- China
- Prior art keywords
- template
- picture editting
- editing
- information
- strategy
- 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
Abstract
The present invention provides a kind of intelligentized image edit method and device, wherein, method includes：Picture editting's template library is pre-established, picture editting's template-setup therein is described label and picture editting's strategy；S1, according to the crucial editor's information of acquisition of information input by user；S2, according to key edit information inquiry picture editting's template library, is matched with each description label, and obtain the obtained corresponding picture editting's template of description label of matching；S3, picture editting's strategy editor's target image in picture editting's template.The present invention can be according to information input by user, and the intelligentized editor for obtaining user is intended to, and chooses matched picture editting's template to carry out editing and processing to target image.
Description
【Technical field】
The present invention relates to image-editing technology, more particularly to a kind of intelligent image edit method and device.
【Background technology】
Nowadays, with the extensive use of digital camera technology, people can quickly and conveniently shooting digital pictures, meanwhile,
This also brings demand of the people for picture editting and beautification therewith.Existing image edit method can be roughly divided into two classes：
One kind is to provide the largely basic operation to image into edlin to the user, directly chosen by user these operate to image into
Edlin is to reach satisfied effect, such as photoshop softwares；It is another kind of to there is provided some intrinsic picture editting's templates,
By user select one of picture editting's template and directly using the template to image into edlin.Above-mentioned two is existing
Image edit method is required for clearly being operated based on user come into edlin, and the editor's meaning that intelligentized can not obtain user
Figure.
【The content of the invention】
In view of this, can be according to input by user the present invention provides a kind of intelligent image edit method and device
The editor of the acquisition user of information intelligent is intended to and carries out editing and processing to target image.
Concrete technical scheme is as follows：
A kind of intelligence image edit method, picture editting's template-setup in picture editting's template library be described label and
Picture editting's strategy, this method include：
S101, according to the crucial editor's information of acquisition of information input by user；
S102, according to it is described it is crucial edit information inquiry described image editing template storehouse, matched with each description label,
And obtain the corresponding picture editting's template of description label that matching obtains；
S103, picture editting's strategy editor's target image in described image editing template.
According to one preferred embodiment of the present invention, the information input by user is voice messaging,
The key editor's information that obtains specifically includes：The voice messaging is believed by speech recognition conversion for word
Breath, and the text information or the crucial phrase extracted from the text information are edited into information as key；
Alternatively, the emotional information in the voice messaging is obtained by voice mood identification, using the emotional information as pass
Key editor's information.
According to one preferred embodiment of the present invention, the information input by user is music information,
The key editor's information that obtains specifically includes：The music information storehouse pre-established according to the music information retrieval
Or music rhythm feature database, matched musical features are obtained, and using acquired matched musical features as key editor's letter
Breath.
According to one preferred embodiment of the present invention, the information input by user is text information, by the text information or
The crucial phrase extracted from the text information is as crucial editor's information.
According to one preferred embodiment of the present invention, the corresponding image of description label that matching obtains is obtained in the step S102
Editing template specifically includes：
Weighted value, which is obtained, in the corresponding picture editting's template of description label obtained from matching is higher than threshold value set in advance
One or more picture editting's templates, label crucial edits the higher picture editting's mould of information matches degree with described described in it
Plate weighted value is higher.
According to one preferred embodiment of the present invention, scene Editing Strategy, institute are still further comprised in described image editing template
Step S103 is stated to further comprise：
S601, check and whether there is scene Editing Strategy in described image editing template, if in the presence of continuing step
S602；Otherwise, step S604 is gone to；
S602, the set according to the situation elements for corresponding to the scene Editing Strategy, analyze the target image, detect
The situation elements having in the target image；
S603, according to the scene Editing Strategy, the field to corresponding to the scene Editing Strategy in the target image
Scape element carries out editing and processing；
S604, picture editting's strategy in described image editing template, editing and processing is carried out to the target image.
A kind of intelligence image editing apparatus, pre-establishes picture editting's template library, in described image editing template storehouse
Picture editting's template has corresponding description label, and described image editing template includes picture editting's strategy, which includes：
Pretreatment unit, for according to the crucial editor's information of acquisition of information input by user；
Query unit, for editing information inquiry described image editing template storehouse according to the key, with each description label
Matched, and obtain the corresponding picture editting's template of description label that matching obtains；
Edit cell, for picture editting's strategy editor's target image in described image editing template.
According to one preferred embodiment of the present invention, the information input by user is voice messaging,
The pretreatment unit specifically performs when obtaining crucial editor's information：The voice messaging is turned by speech recognition
Text information is changed to, and the text information or the crucial phrase extracted from the text information are edited into information as key；
Alternatively, the emotional information in the voice messaging is obtained by voice mood identification, using the emotional information as pass
Key editor's information.
According to one preferred embodiment of the present invention, the information input by user is music information,
The pretreatment unit specifically performs when obtaining crucial editor's information：Pre-established according to the music information retrieval
Music information storehouse or music rhythm feature database, obtain matched musical features, and acquired matched musical features are made
Information is edited for key.
According to one preferred embodiment of the present invention, the information input by user is text information, and the pretreatment unit obtains
Specifically performed when taking crucial editor's information：By the text information input by user or the key extracted from the text information
Phrase is as crucial editor's information.
According to one preferred embodiment of the present invention, the query unit obtains the corresponding image of description label that matching obtains and compiles
Specifically performed when collecting template：
Weighted value, which is obtained, in the corresponding picture editting's template of description label obtained from matching is higher than threshold value set in advance
One or more picture editting's templates, label crucial edits the higher picture editting's mould of information matches degree with described described in it
Plate weighted value is higher.
According to one preferred embodiment of the present invention, scene Editing Strategy, institute are still further comprised in described image editing template
Edit cell is stated specifically to perform：
S1201, check and whether there is scene Editing Strategy in described image editing template, if in the presence of continuing step
S1202；Otherwise, step S1204 is gone to；
S1202, the set according to the situation elements for corresponding to the scene Editing Strategy, analyze the target image, examine
The situation elements having are surveyed in the target image；
S1203, according to the scene Editing Strategy, the field to corresponding to the scene Editing Strategy in the target image
Scape element carries out editing and processing；
S1204, picture editting's strategy in described image editing template, Editorial Services is carried out to the target image
Reason.
As can be seen from the above technical solutions, the present invention is by voice messaging input by user, text information or music
Information is handled, and extracts crucial editor's information therein, and edit information according to the key and compiled in the image pre-established
Collect in template library and obtain matched picture editting's template, come to carry out Editorial Services to target image according to picture editting's template with this
Reason.
【Brief description of the drawings】
Fig. 1 is the intelligent image edit method flow chart that the embodiment of the present invention one provides；
Fig. 2 is the method flow diagram for application image editing template editor's target image that the embodiment of the present invention one provides；
Fig. 3 is intelligent image editing apparatus schematic diagram provided by Embodiment 2 of the present invention.
【Embodiment】
In order to make the object, technical solutions and advantages of the present invention clearer, below in conjunction with the accompanying drawings with specific embodiment pair
The present invention is described in detail.
In order to select editing template the most matched, it is necessary to pre-establish figure according to information intelligent input by user
As editing template storehouse.Picture editting's template that can be to target image into edlin is preserved in picture editting's template library.It is different
Picture editting's template correspond to different picture editting's strategies, for example, can be to view picture figure corresponding to the editing template of adjustment tone
The tone of picture is adjusted, and entire image can be zoomed in or out corresponding to the editing template of adjustment picture size.
Further, scene Editing Strategy can also be included in picture editting's template, to be used for target image scene
In correspondence some specific objective elements, i.e. situation elements are into edlin, for example, corresponding to situation elements in picture editting's template
The scene Editing Strategy in blue sky can carry out corresponding edit operation to the blue sky in target image, as increased cloud layer in blue sky
Scene Editing Strategy can increase cloud layer on the blue sky of target image.It is understood that the example above is merely for exemplary mesh
, the embodiment of the present invention not limited to this.
For the corresponding description label of each picture editting's template in picture editting's template library, description label is used
In the Editing Strategy for characterizing its corresponding picture editting's template, in order to retrieve matched image according to information input by user
Editing template.For example, corresponding to by the description label of the graying picture editting's template of target image tone can be " tone,
Ash ", can be " amplification, one times " corresponding to target image to be put to the description label of twice picture editting's template, meanwhile, if
The Editing Strategy to special scenes is further included in picture editting's template, then it describes label and equally characterizes picture editting's template
In scene Editing Strategy.
Further, mood can also be added in the description label of picture editting's template and describe label, for example, corresponding
The tone of image can be brightened for picture editting's template of " happy " in mood label.At the same time it can also in picture editting's template
Description label in add music label.It is understood that the example above is merely for exemplary purpose, the embodiment of the present invention
Not limited to this.
Embodiment one
Fig. 1 is the intelligent image edit method flow chart that the embodiment of the present invention one provides, as shown in Figure 1, this method bag
Include
S101, according to the crucial editor's information of acquisition of information input by user.
In method provided in an embodiment of the present invention, information input by user can be：Voice messaging, music information and word
Information.
If input by user is voice messaging, first by acoustic sensor, voice messaging input by user is obtained, it
Afterwards, acquired voice messaging is converted to by corresponding text information by speech recognition technology, using the text information as pass
Key editor's information.Further, for voice messaging input by user, it can also be obtained and used by voice mood identification technology
Emotional information in the voice messaging of family, and edit information using the emotional information as key, i.e., believed by voice input by user
Crucial editor's information that breath obtains, can include the text information obtained after voice messaging is changed, can also include mood and believe
Breath.
For music information, user can be inputted by word or phonetic entry.Wherein, inputted for word, Yong Huke
To input the name or a lyrics of a song by word, afterwards, the name or the lyrics of the song inputted according to user,
The music information storehouse pre-established is retrieved, obtains the name of song or the music of lyric match inputted with user.For language
Sound inputs, and user afterwards, can be identified by the name or a lyrics of one song of phonetic entry by speech recognition technology
Go out the name or the lyrics for the song that user is inputted, and the music information storehouse pre-established according to the information retrieval, obtain matching
Music；Alternatively, user can hum one section of music, afterwards, music rhythm feature is extracted among the music of user's humming, and
The music rhythm feature is matched with the musical features in the music rhythm feature database pre-established, into getting similarity
Highest music.Finally, using according to the music that music information retrieval input by user obtains information is edited as key.Music is revolved
Rule feature identification technique is the prior art, and details are not described herein.
If input by user is text information, the text information that user is inputted is directly acquired, and by the text information
Information is edited as key.
Further, the voice messaging and text information inputted for user, after processing being analyzed it,
Crucial editor's information therein is obtained again.The present invention provides a kind of preferred embodiment：For text information input by user and
The text information that voice messaging input by user identification is obtained afterwards, carries out word segmentation processing to it, carries out semantic analysis afterwards,
Crucial phrase therein is extracted, such as " amplification ", " one times ", " sky ", and preposition therein, auxiliary word etc. are filtered to semantic information
Influence less word, such as " very ", " very ", " ", finally, will after participle and semantic analysis resulting crucial phrase as pass
Key editor's information.Participle and semantic analysis are the prior art, and details are not described herein.
S102, according to it is described it is crucial edit information inquiry picture editting's template library, and obtain picture editting's template.
According to acquired crucial editor's information inquiry picture editting's template library, and one is obtained from picture editting's template library
A or multiple corresponding picture editting's templates.Specifically inquiry mode can be：Label described in query image editing template storehouse
With picture editting's template of acquired crucial editor's information matches, weight is set for picture editting's template in query process
Value, the description label of picture editting's template and the matching degree of acquired crucial editor's information are higher, then picture editting's mould
The weighted value of plate is also higher, afterwards, weighted value is higher than a certain one or more picture editting's templates for presetting numerical value
As picture editting's template with information matches input by user.For the emotional information or root in user speech information
The crucial editor's information obtained according to music information input by user, again may be by above-mentioned inquiry mode, query specification label
In mood label or music label with those key editor information matches picture editting's templates.
S103, using described image editing template editor's target image.
After picture editting's template with information matches input by user is obtained, according to acquired picture editting's template
In picture editting's strategy image is subjected to corresponding editing and processing.
Further, scene Editing Strategy can also be included in picture editting's template, corresponding to this step of the situation
Implementation method, as shown in Fig. 2, can be realized by following sub-step S1031-S1034：
S1031, check in acquired picture editting's template and whether there is scene Editing Strategy, if in the presence of continuing to walk
Rapid S1032；Otherwise, step S1034 is gone to.
S1032, the set according to the situation elements for corresponding to scene Editing Strategy, analyze the target image, detect institute
State the situation elements having in target image.
Based on the set of the situation elements corresponding to scene Editing Strategy, according to the situation elements algorithms library pre-established,
The situation elements being contained in possessed by detection target image in situation elements set., can be with for specific situation elements
Positioning is detected by specific image detection algorithm, to determine to whether there is the situation elements in present image, and really
Image-region existing for the fixed situation elements.Such as be directed to situation elements " face ", can by specific Face datection algorithm come
Positioning is detected, and is directed to situation elements " sun ", positioning can be detected by specific sun detection algorithm.Base
It is the prior art in the image detection algorithm of special scenes element, details are not described herein.It should be noted that if there is multiple
Picture editting's masterplate into edlin, then only needs to carry out once to be directed to the detection process of the situation elements to same situation elements
, it is not necessary to compute repeatedly.
S1033, according to the scene Editing Strategy, the field to corresponding to the scene Editing Strategy in the target image
Scape element carries out editing and processing.
Possessed by detecting target image corresponding to the situation elements situation elements of scene Editing Strategy after, then root
Editing and processing is carried out to corresponding situation elements according to the scene Editing Strategy editor in picture editting's template；If target image
In situation elements not corresponding with scene Editing Strategy, then ignore the scene Editing Strategy.For example, the matched figure obtained
As there is the scene Editing Strategy for corresponding to " sun " in editing template, if having this situation elements of the sun, pin in target image
Corresponding processing operation is carried out to the sun in target image；If there is no this situation elements of the sun in target image, ignore
The scene Editing Strategy.
S1034, picture editting's strategy in described image editing template, Editorial Services is carried out to the target image
Reason.
When above-mentioned S1031-S1034 includes scene Editing Strategy for picture editting's template, to target image into edlin
The method flow of processing, can also be first according to picture editting's strategy to target figure in above method flow, it is necessary to explanation
As carrying out editing and processing, editing and processing is carried out to the correspondence situation elements in target image further according to scene Editing Strategy afterwards.
If in step s 102, matching obtains multiple images editing template, then the plurality of picture editting's template will be applied to compile
Collect the multiple effects obtained after target image and all show user, and can be further one of as finally for user's selection
Handling result.
Further, the multiple images editing template obtained can also be merged to one comprehensive picture editting of generation
Template, during merging, picture editting's template based on the highest picture editting's template of weighted value, afterwards by weighted value by height
It is extremely low, the Editing Strategy in remaining picture editting's template is added in base image editing template successively, in merging process
In, picture editting's strategy for that can produce conflict, then select the image in the wherein highest picture editting's template of weighted value to compile
Collect strategy to be added in base image editing template, the scene Editing Strategy of corresponding same situation elements, then equally give up it
Scene Editing Strategy in the relatively low picture editting's template of middle weighted value, for example, having pattern colour in base image editing template
The bright picture editting's strategy of modulation, and have in the secondary high picture editting's template of weighted value by the dimmed picture editting's plan of picture tone
Slightly, then just give up by picture editting's strategy that picture tone is dimmed in the high picture editting's template of weighted value time, and only retain
Picture editting's strategy that picture tone brightens in base image editing template.It is understood that the example above is merely for showing
The purpose of example, the embodiment of the present invention not limited to this.
In the following, the method provided the embodiments of the present invention one of further illustrating is described.
For example, the voice messaging is converted to word by user by speech recognition first by phonetic entry " here comes the rain "
Information, afterwards, can segment the text information and semantic analysis is handled, extract keyword therein " rainy ", with
The keyword goes to query image editing template storehouse as crucial editor's information, gets in a description label and contains " rainy "
With picture editting's template of the key information match, picture editting's strategy in picture editting's template can increase in the target image
Add rainy figure layer, weaken light, the scene Editing Strategy corresponding to the sun is further included in picture editting's template, if target image
In have this situation elements of the sun, then can add cloud layer corresponding to the scene Editing Strategy of the sun and block the sun.
In another example user can be sighed by phonetic entry, afterwards, the mood for learning user by voice mood identification is believed
Cease for " losing ", information inquiry picture editting's template library is edited using the emotional information as key, get a description label
In the mood label containing " losing " picture editting's template, picture editting's strategy in picture editting's template can be image
Tone is graying.
In another example user can hum a piece of music by phonetic entry, such as《About in the winter time》, afterwards, user is groaned
The music sung carries out affective property identification of music, and the music for learning user's humming is《About in the winter time》, using " about in the winter time " as
Key editor's information inquiry picture editting's template library, gets in a description label containing music label " about in the winter time "
Picture editting's template, picture editting's strategy in picture editting's template can snow figure layer in increase in the target image.
It is understood that the example above is merely for exemplary purpose, the embodiment of the present invention not limited to this.
Embodiment two
Fig. 3 is intelligent image editing apparatus schematic diagram provided by Embodiment 2 of the present invention, as shown in figure 3, the device bag
Include：Pretreatment unit 10, query unit 20, edit cell 30.
Pretreatment unit 10, for according to the crucial editor's information of acquisition of information input by user.
In device provided in an embodiment of the present invention, information input by user can be：Voice messaging, music information and word
Information.
If input by user is voice messaging, pretreatment unit 10 by acoustic sensor, obtains user's input first
Voice messaging, afterwards, acquired voice messaging is converted to by corresponding text information by speech recognition technology, by this article
Word information is as crucial editor's information.Further, can also lead to for voice messaging input by user, pretreatment unit 10
The emotional information in voice mood identification technology acquisition user speech information is crossed, and using the emotional information as key editor's letter
Breath, i.e., the crucial editor's information obtained by voice messaging input by user, can include what is obtained after voice messaging is changed
Text information, can also include emotional information.
For music information, user can be inputted by word or phonetic entry.Wherein, inputted for word, Yong Huke
To input the name or a lyrics of a song, afterwards, the song that pretreatment unit 10 is inputted according to user by word
Name or the lyrics, retrieve the music information storehouse pre-established, obtain the name or lyric match of the song inputted with user
Music.For phonetic entry, user can pass through the name or a lyrics of one song of phonetic entry, afterwards, pretreatment
Unit 10 identifies the name or the lyrics of song that user inputted by speech recognition technology, and advance according to the information retrieval
The music information storehouse of foundation, obtains matched music；Alternatively, user can hum one section of music, afterwards, pretreatment unit 10 from
Music rhythm feature is extracted among the music of user's humming, and by the music rhythm feature and the musical features storehouse pre-established
Music rhythm feature matched, into getting the highest music of similarity.Finally, pretreatment unit 10 will be defeated according to user
The music that the music information retrieval entered obtains is as crucial editor's information.Music rhythm feature identification technique is the prior art,
This is repeated no more.
If input by user is text information, pretreatment unit 10 directly acquires the text information that user is inputted, and
Information is edited using the text information as key.
Further, the voice messaging and text information inputted for user, pretreatment unit 10 can to its into
After row analyzing and processing, then obtain crucial editor's information therein.The present invention provides a kind of preferred embodiment：Inputted for user
Text information and the text information that obtains afterwards of voice messaging input by user identification, pretreatment unit 10 divide it
Word processing, carries out semantic analysis, extracts crucial phrase therein afterwards, such as " amplification ", " one times ", " sky ", and filters it
In preposition, auxiliary word etc. less word is influenced on semantic information, such as " very ", " very ", " ", finally, will participle and semanteme
Resulting crucial phrase is as crucial editor's information after analysis.Participle and semantic analysis are the prior art, and details are not described herein.
Query unit 20, for editing information inquiry described image editing template storehouse according to the key, with the description
Label is matched, and obtains the corresponding picture editting's template of description label that matching obtains.
Query unit 20 is according to acquired crucial editor's information inquiry picture editting's template library, and from picture editting's template
One or more corresponding picture editting's templates are obtained in storehouse.Specifically inquiry mode can be：Query image editing template storehouse
Described in label and picture editting's template of acquired crucial editor's information matches, be picture editting's template in query process
Weighted value is set, the description label of picture editting's template and the matching degree of acquired crucial editor's information are higher, then the figure
As the weighted value of editing template is also higher, afterwards, weighted value is higher than a certain one or more images for presetting numerical value
Editing template is as picture editting's template with information matches input by user.Believe for the mood in user speech information
Breath or the crucial editor's information obtained according to music information input by user, query unit 20 again may be by above-mentioned inquiry
Mode, mood label or music label and picture editting's templates of those keys editor's information matches in query specification label.
Edit cell 30, for picture editting's strategy editor's target image in described image editing template.
After picture editting's template with information matches input by user is obtained, edit cell 30 is according to acquired figure
As image is carried out corresponding editing and processing by picture editting's strategy in editing template.
Further, scene Editing Strategy can also be included in picture editting's template, corresponding to the situation edit cell
30 can perform operations described below S2031-S2032 to complete the editor to image：
S2031, check in acquired picture editting's template and whether there is scene Editing Strategy, if in the presence of continuing to walk
Rapid S2032；Otherwise, step S2034 is gone to.
S2032, the set according to the situation elements for corresponding to scene Editing Strategy, analyze the target image, detect institute
State the situation elements having in target image.
Based on the set of the situation elements corresponding to scene Editing Strategy, according to the situation elements algorithms library pre-established,
The situation elements being contained in possessed by detection target image in situation elements set., can be with for specific situation elements
Positioning is detected by specific image detection algorithm, to determine to whether there is the situation elements in present image, and really
Image-region existing for the fixed situation elements.Such as be directed to situation elements " face ", can by specific Face datection algorithm come
Positioning is detected, and is directed to situation elements " sun ", positioning can be detected by specific sun detection algorithm.Base
It is the prior art in the image detection algorithm of special scenes element, details are not described herein.It should be noted that if there is multiple
Picture editting's masterplate into edlin, then only needs to carry out once to be directed to the detection process of the situation elements to same situation elements
, it is not necessary to compute repeatedly.
S2033, according to the scene Editing Strategy, the field to corresponding to the scene Editing Strategy in the target image
Scape element carries out editing and processing.
After the possessed situation elements corresponding to scene Editing Strategy in detecting target image, then compiled according to image
The scene Editing Strategy editor collected in template carries out editing and processing to corresponding situation elements；If in target image not with
The corresponding situation elements of scene Editing Strategy, then ignore the scene Editing Strategy.For example, the matched picture editting's mould obtained
There is the scene Editing Strategy corresponding to " sun " in plate, if there are this situation elements of the sun in target image, for target figure
The sun as in carries out corresponding processing operation；If there is no this situation elements of the sun in target image, ignore scene volume
Collect strategy.
S2034, picture editting's strategy in described image editing template, Editorial Services is carried out to the target image
Reason.
When above-mentioned S2031-S2034 includes scene Editing Strategy for picture editting's template, performed by edit cell 30
The operation of editing and processing is carried out to target image, it is necessary to which explanation, edit cell 30 can also be first according to picture editting's strategies
Editing and processing is carried out to target image, the correspondence situation elements in target image are compiled further according to scene Editing Strategy afterwards
Collect processing.
If the matching of query unit 20 obtains multiple images editing template, the plurality of picture editting's edit model mesh will be applied
The multiple effects obtained after logo image all show user, and can be further one of as final place for user's selection
Manage result.
Further, the multiple images editing template obtained can also be merged to one comprehensive picture editting of generation
Template, during merging, picture editting's template based on the highest picture editting's template of weighted value, afterwards by weighted value by height
It is extremely low, the Editing Strategy in remaining picture editting's template is added in base image editing template successively, in merging process
In, picture editting's strategy for that can produce conflict, then select the image in the wherein highest picture editting's template of weighted value to compile
Collect strategy to be added in base image editing template, the scene Editing Strategy of corresponding same situation elements, then equally give up it
Scene Editing Strategy in the relatively low picture editting's template of middle weighted value, for example, having pattern colour in base image editing template
The bright picture editting's strategy of modulation, and have in the secondary high picture editting's template of weighted value by the dimmed picture editting's plan of picture tone
Slightly, then just give up by picture editting's strategy that picture tone is dimmed in the high picture editting's template of weighted value time, and only retain
Picture editting's strategy that picture tone brightens in base image editing template.It is understood that the example above is merely for showing
The purpose of example, the embodiment of the present invention not limited to this.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
God and any modification, equivalent substitution, improvement and etc. within principle, done, should be included within the scope of protection of the invention.
Claims (6)
1. a kind of intelligence image edit method, it is characterised in that picture editting's template-setup in picture editting's template library has
Label and picture editting's strategy are described, this method includes：
S101, according to the crucial editor's information of acquisition of information input by user；
S102, according to it is described it is crucial edit information inquiry described image editing template storehouse, matched with each description label, and obtain
Take the corresponding picture editting's template of the description label that matching obtains；
S103, picture editting's strategy editor's target image in described image editing template；Wherein,
The information input by user is music information,
The key editor's information that obtains specifically includes：The music information storehouse pre-established according to the music information retrieval or sound
Happy melody characteristics storehouse, obtains matched musical features, and edits information using acquired matched musical features as key.
2. according to the method described in claim 1, it is characterized in that, the description label that matching obtains is obtained in the step S102
Corresponding picture editting's template specifically includes：
Weighted value is higher than threshold value set in advance one is obtained from the corresponding picture editting's template of description label that matching obtains
A or multiple images editing template, label is weighed with the crucial higher picture editting's template of information matches degree of editing described in it
Weight values are higher.
3. according to the method described in claim 1, it is characterized in that, scene volume is still further comprised in described image editing template
Strategy is collected, the step S103 further comprises：
S601, check and whether there is scene Editing Strategy in described image editing template, if in the presence of continuing step S602；It is no
Then, step S604 is gone to；
S602, the set according to the situation elements for corresponding to the scene Editing Strategy, analyze the target image, described in detection
The situation elements having in target image；
S603, according to the scene Editing Strategy, the scene member to corresponding to the scene Editing Strategy in the target image
Element carries out editing and processing；
S604, picture editting's strategy in described image editing template, editing and processing is carried out to the target image.
4. a kind of intelligence image editing apparatus, it is characterised in that pre-establish picture editting's template library, described image editor's mould
Picture editting's template in plate storehouse has corresponding description label, and described image editing template includes picture editting's strategy, the dress
Put including：
Pretreatment unit, for according to the crucial editor's information of acquisition of information input by user；
Query unit, for according to crucial editor's information inquiry described image editing template storehouse, being carried out with each description label
Matching, and obtain the corresponding picture editting's template of description label that matching obtains；
Edit cell, for picture editting's strategy editor's target image in described image editing template；Wherein,
The information input by user is music information,
The pretreatment unit specifically performs when obtaining crucial editor's information：The sound pre-established according to the music information retrieval
Happy information bank or music rhythm feature database, obtains matched musical features, and using acquired matched musical features as pass
Key editor's information.
5. device according to claim 4, it is characterised in that the query unit obtains the description label pair that matching obtains
Specifically performed during the picture editting's template answered：
Weighted value is higher than threshold value set in advance one is obtained from the corresponding picture editting's template of description label that matching obtains
A or multiple images editing template, label is weighed with the crucial higher picture editting's template of information matches degree of editing described in it
Weight values are higher.
6. device according to claim 4, it is characterised in that scene volume is still further comprised in described image editing template
Strategy is collected, the edit cell specifically performs：
S1201, check and whether there is scene Editing Strategy in described image editing template, if in the presence of continuing step S1202；
Otherwise, step S1204 is gone to；
S1202, the set according to the situation elements for corresponding to the scene Editing Strategy, analyze the target image, detect institute
State the situation elements having in target image；
S1203, according to the scene Editing Strategy, the scene member to corresponding to the scene Editing Strategy in the target image
Element carries out editing and processing；
S1204, picture editting's strategy in described image editing template, editing and processing is carried out to the target image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210428871.4A CN102982572B (en) | 2012-10-31 | 2012-10-31 | A kind of intelligence image edit method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210428871.4A CN102982572B (en) | 2012-10-31 | 2012-10-31 | A kind of intelligence image edit method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102982572A CN102982572A (en) | 2013-03-20 |
| CN102982572B true CN102982572B (en) | 2018-05-01 |
Family
ID=47856509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210428871.4A Active CN102982572B (en) | 2012-10-31 | 2012-10-31 | A kind of intelligence image edit method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102982572B (en) |
Families Citing this family (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103677566A (en) * | 2013-11-27 | 2014-03-26 | 北京百纳威尔科技有限公司 | Picture editing method and picture editing device |
| CN103605975B (en) * | 2013-11-28 | 2018-10-19 | 小米科技有限责任公司 | A kind of method, apparatus and terminal device of image procossing |
| CN103793931A (en) * | 2014-02-18 | 2014-05-14 | 厦门美图网科技有限公司 | Method for fusing words and pictures |
| CN105279751B (en) * | 2014-07-17 | 2019-09-17 | 腾讯科技（深圳）有限公司 | A kind of method and apparatus handled for picture |
| CN106034210A (en) * | 2015-03-09 | 2016-10-19 | 华为技术有限公司 | Image editing method and image editing device |
| CN104735468B (en) * | 2015-04-03 | 2018-08-31 | 北京威扬科技有限公司 | A kind of method and system that image is synthesized to new video based on semantic analysis |
| CN104866308A (en) * | 2015-05-18 | 2015-08-26 | 百度在线网络技术（北京）有限公司 | Scenario image generation method and apparatus |
| CN105302786B (en) * | 2015-11-10 | 2019-05-24 | 百度在线网络技术（北京）有限公司 | The edit methods and device of data |
| CN105979035B (en) * | 2016-06-28 | 2019-08-27 | Oppo广东移动通信有限公司 | A kind of augmented reality AR image processing method, device and intelligent terminal |
| CN106156310A (en) * | 2016-06-30 | 2016-11-23 | 努比亚技术有限公司 | A kind of picture processing apparatus and method |
| CN106780663B (en) * | 2016-11-22 | 2020-02-21 | 维沃移动通信有限公司 | Template frame character processing method and device |
| CN107392982A (en) * | 2017-07-27 | 2017-11-24 | 深圳章鱼信息科技有限公司 | Photographing On-line method, apparatus and system |
| CN110147190B (en) * | 2018-06-29 | 2024-03-08 | 腾讯科技（深圳）有限公司 | Image processing method and electronic terminal |
| CN108965770B (en) * | 2018-08-30 | 2021-12-07 | Oppo广东移动通信有限公司 | Image processing template generation method and device, storage medium and mobile terminal |
| CN109308178A (en) * | 2018-08-31 | 2019-02-05 | 维沃移动通信有限公司 | A kind of voice drafting method and its terminal device |
| CN109445757B (en) * | 2018-09-21 | 2022-07-29 | 深圳变设龙信息科技有限公司 | New design drawing generation method and device and terminal equipment |
| CN111429551A (en) * | 2020-03-20 | 2020-07-17 | 北京达佳互联信息技术有限公司 | Image editing method, device, electronic equipment and storage medium |
| CN113643411A (en) * | 2020-04-27 | 2021-11-12 | 北京达佳互联信息技术有限公司 | Image special effect adding method and device, electronic equipment and storage medium |
| CN111741331B (en) * | 2020-08-07 | 2020-12-22 | 北京美摄网络科技有限公司 | Video clip processing method, device, storage medium and equipment |
| CN112950735A (en) * | 2021-03-04 | 2021-06-11 | 爱昕科技(广州)有限公司 | Method for generating image by combining sketch and voice instruction, computer readable storage medium and display device |
| CN115129924A (en) * | 2021-03-26 | 2022-09-30 | 北京达佳互联信息技术有限公司 | Content searching method and device based on voice, electronic equipment and storage medium |
| CN113467735A (en) * | 2021-06-16 | 2021-10-01 | 荣耀终端有限公司 | Image adjusting method, electronic device and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004178163A (en) * | 2002-11-26 | 2004-06-24 | Matsushita Electric Ind Co Ltd | Image processing method and device |
| WO2005022464A1 (en) * | 2003-08-29 | 2005-03-10 | Koninklijke Philips Electronics N.V. | A method, a device and a computer program arranged to develop and execute an executable template of an image processing protocol |
| CN102422286A (en) * | 2009-03-11 | 2012-04-18 | 香港浸会大学 | Automatic and semi-automatic image classification, annotation and tagging through the use of image acquisition parameters and metadata |
-
2012
- 2012-10-31 CN CN201210428871.4A patent/CN102982572B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN102982572A (en) | 2013-03-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102982572B (en) | A kind of intelligence image edit method and device | |
| Murthy et al. | Content-based music information retrieval (cb-mir) and its applications toward the music industry: A review | |
| CN108806668A (en) | A kind of audio and video various dimensions mark and model optimization method | |
| CN107169430A (en) | Reading environment audio strengthening system and method based on image procossing semantic analysis | |
| Dighe et al. | Swara Histogram Based Structural Analysis And Identification Of Indian Classical Ragas. | |
| CN107393554A (en) | In a kind of sound scene classification merge class between standard deviation feature extracting method | |
| CN109829482A (en) | Song training data processing method, device and computer readable storage medium | |
| CN113420556B (en) | Emotion recognition method, device, equipment and storage medium based on multi-mode signals | |
| Benetos et al. | Characterisation of acoustic scenes using a temporally-constrained shift-invariant model | |
| CN115422947A (en) | Ancient poetry assignment method and system based on deep learning | |
| CN116542817B (en) | Intelligent digital lawyer consultation method and system | |
| CN110059179A (en) | A kind of song text name entity recognition method based on deep learning | |
| CN106446280A (en) | Song data processing method and device | |
| KR100869643B1 (en) | Mp3-based popular song summarization installation and method using music structures, storage medium storing program for realizing the method | |
| KR20170091888A (en) | Method and system for automatically tagging themes suited for songs | |
| CN116129868A (en) | Method and system for generating structured photo | |
| Nagavi et al. | Content based audio retrieval with MFCC feature extraction, clustering and sort-merge techniques | |
| CN101370216A (en) | Emotional processing and playing method for mobile phone audio files | |
| Gupta | Deep audio embeddings and attention based music emotion recognition | |
| Jitendra et al. | An ensemble model of CNN with Bi-LSTM for automatic singer identification | |
| CN113268628A (en) | Music emotion recognition method based on modularized weighted fusion neural network | |
| Kaimal et al. | Introduction to Identification of Raga in Carnatic Music and its Corresponding Hindustani Music | |
| US10489450B1 (en) | Selecting soundtracks | |
| Xu et al. | The Aalto system based on fine-tuned audioset features for dcase2018 task2-general purpose audio tagging | |
| CN114756734B (en) | Music piece subsection emotion marking system and method based on machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |