CN104951749A - Image content recognition device and image content recognition method - Google Patents
Image content recognition device and image content recognition method Download PDFInfo
- Publication number
- CN104951749A CN104951749A CN201510240225.9A CN201510240225A CN104951749A CN 104951749 A CN104951749 A CN 104951749A CN 201510240225 A CN201510240225 A CN 201510240225A CN 104951749 A CN104951749 A CN 104951749A
- Authority
- CN
- China
- Prior art keywords
- image
- self
- module
- symbol
- defined symbol
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/36—Matching; Classification
- G06V30/387—Matching; Classification using human interaction, e.g. selection of the best displayed recognition candidate
Abstract
The invention discloses an image content recognition device and an image content recognition method. The image content recognition method includes: acquiring custom symbols and setting processing elements corresponding to the custom symbols according to user's instructions; extracting custom symbol features from the custom symbols; acquiring an image with the custom symbols; matching the custom symbol features with the acquired image and matching to obtain the custom symbol features and positions thereof in the image from the image; selecting picture content, corresponding to the custom symbol features, at the positions of the custom symbol features from the image; subjecting the acquired picture content to recognition and corresponding editing processing according to the corresponding processing elements of the custom symbols; displaying an editing processing result. The image content recognition device and the image content recognition method have the advantages that processing efficiency for recognizing local content in the image can be improved, and application range and application scenes for local content recognition can be widened.
Description
Technical field
The application relates to technical field of image processing, particularly relates to a kind of picture material recognition device and method.
Background technology
In daily life, user often needs to take passages the local content in image, such as passage, and a width is drawn, and is indifferent to other element in image.
When user is taken by current camera, captured record be complete picture, if user only need local content, then need after shooting completes, edited captured image by third party's graphics software, amendment, uses very loaded down with trivial details.
A kind of optical character identification (OCR is have also appeared in prior art, Optical Character Recognition) technology, described OCR refers to that electronic equipment (such as scanner or digital camera) checks the character that paper prints, determining its shape by detecting dark, bright pattern, then with character identifying method, shape being translated into the process of computword.
But, the following shortcoming of existing image recognition technology ubiquity:
1) take photograph time, display be complete photograph, when user needs to carry out identifying processing to the local content in image, need by use third party software edit, troublesome poeration, inefficiency.
2) automatic OCR recognition technology is used, such as business card recognition software, can according to certain template, content is added in automatic identification, but its template is fixed, cannot self-defined identification range, therefore its range of application and application scenarios are too little, only be confined to the file (as business card) of certain specific normal form, and it is for the lack of identification of user's handwritten content.
Summary of the invention
In view of this, fundamental purpose of the present invention is to provide a kind of picture material recognition device and method, to improve the treatment effeciency to local content identification in image, expands range of application and the application scenarios of local content identification.
Technical scheme of the present invention is achieved in that
A kind of picture material recognition device, comprising:
Symbol custom block, for being obtained from define symbol, arranges process key element corresponding to described self-defined symbol according to user instruction;
Characteristic extracting module, for extracting self-defined symbolic feature from self-defined symbol;
Image collection module, for obtaining the image with self-defined symbol;
Characteristic matching module, for being mated with obtained image by described self-defined symbolic feature, matches described self-defined symbolic feature and position in the picture thereof from this image;
Contents selection module, for from described image, chooses the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place;
Content recognition editor module, for according to process key element corresponding to described self-defined symbol, carries out identifying and corresponding editing and processing to obtained image content;
Display module, for showing editing and processing result.
In an advantageous embodiment, described symbol custom block comprises with the Arbitrary Term in lower module:
For the module of the hand-written self-defined symbol that the input equipment obtaining smart machine inputs;
For showing the shape of acquiescence, according to the selection instruction of user's input, from the shape of acquiescence, select a module as the self-defined symbol got;
For calling the picture of filming apparatus shooting containing self-defined symbol, from this picture, identify the module of described self-defined symbol.
In an advantageous embodiment, the process key element that described self-defined symbol is corresponding comprises editor's action and object content form; Described content recognition editor module comprises: for carrying out the module of corresponding editing operation to obtained image content according to editor's action corresponding to described self-defined symbol; For the image content after described editing operation being converted into the module of described object content form.
In an advantageous embodiment, described image collection module comprises with the Arbitrary Term in lower module:
There is for calling filming apparatus shooting the image of self-defined symbol, the image of shooting being sent to the module of characteristic matching module;
For obtaining the image file in smart machine file system, this image file is sent to the module of described characteristic matching module;
For intercepting the image of screen of intelligent device display, this image is sent to the module of described characteristic matching module;
Obtaining user to marking operation screen showing image for calling screen interface, the screen picture after labelling being sent to the module of described characteristic matching module.
In an advantageous embodiment, described contents selection module comprises with the Arbitrary Term in lower module:
For when described self-defined symbol is characterized as closed figure, automatically choose the module of the image content in this closed figure;
For when described self-defined symbol is characterized as linear, according to the high and spacing of row of setting, obtain this linear on the module meeting the image content of the high and pitch requirements of described row;
For when described self-defined symbol is characterized as linear, obtain the module of the image content in region between white space above this horizontal line to horizontal line.
A kind of picture material recognition methods, comprising:
Be obtained from define symbol, process key element corresponding to described self-defined symbol is set according to user instruction;
Self-defined symbolic feature is extracted from self-defined symbol;
Obtain the image with self-defined symbol;
Described self-defined symbolic feature is mated with obtained image, from this image, matches described self-defined symbolic feature and position in the picture thereof;
From described image, choose the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place;
According to the process key element that described self-defined symbol is corresponding, carry out identifying and corresponding editing and processing to obtained image content;
Show editing and processing result.
In an advantageous embodiment, being obtained from define symbol described in comprises with the Arbitrary Term under type:
The hand-written self-defined symbol that the input equipment obtaining smart machine inputs;
Show the shape of acquiescence, according to the selection instruction of user's input, from the shape of acquiescence, select one as the self-defined symbol got;
Call the picture of filming apparatus shooting containing self-defined symbol, from this picture, identify described self-defined symbol.
In an advantageous embodiment, the process key element that described self-defined symbol is corresponding comprises editor's action and object content form; The described process key element corresponding according to described self-defined symbol, carry out identifying and corresponding editing and processing to obtained image content, specifically comprise: the editor action corresponding according to described self-defined symbol carries out corresponding editing operation to obtained image content; Image content after described editing operation is converted into described object content form.
In an advantageous embodiment, described acquisition has the image of self-defined symbol, specifically comprises with the Arbitrary Term under type:
Call the image that filming apparatus shooting has self-defined symbol, using the image of shooting as the image got;
Obtain the image file in smart machine file system, using this image file as the image got;
Intercept the image of screen of intelligent device display, using this image as the image got;
Call screen interface and obtain user to marking operation screen showing image, using the screen picture after labelling as the image got.
In an advantageous embodiment, described from described image, choose the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place, specifically comprise with the Arbitrary Term under type:
When described self-defined symbol is characterized as closed figure, automatically choose the image content in this closed figure;
When described self-defined symbol is characterized as linear, according to the high and spacing of row of setting, obtain this linear on the image content meeting the high and pitch requirements of described row;
When described self-defined symbol is characterized as linear, obtain the image content in region between white space above this horizontal line to horizontal line.
Compared with prior art, picture material recognition device of the present invention and method can be obtained from define symbol and arrange corresponding process key element, as long as have self-defined symbol in image, the most common scene is user's hand-written described self-defined symbol on image, to mark local content, then can automatically identify this self-defined symbol and position thereof, choose the image content corresponding with described self-defined symbolic feature and the local content at described self-defined symbol feature locations place, and according to process key element corresponding to described self-defined symbol, carry out identifying and corresponding editing and processing to obtained image content, show editing and processing result.Therefore user need not carry out editing and processing to choose local content utilizing numerous and diverse third party software to image, improves the treatment effeciency to local content identification in image.Simultaneously, what the present invention adopted is self-defined symbol, and be that the image content corresponding to this self-defined symbol carries out identifying processing, therefore a certain specific normal form template is not limited to, as long as set corresponding process key element, arbitrarily can be useful in corresponding image local content recognition field, therefore expand range of application and the application scenarios of local content identification.
Accompanying drawing explanation
Fig. 1 is the one composition schematic diagram of picture material recognition device of the present invention;
Fig. 2 a is the interface schematic diagram of a kind of self-defined symbol of the present invention and alignment processing key element thereof;
Fig. 2 b is for drawing self-defined symbol described in Fig. 2 a on the original image to mark a kind of schematic diagram of local content;
Fig. 2 c obtains a kind of schematic diagram described in described Fig. 2 b with the image of self-defined symbol for utilizing smart mobile phone camera head;
A kind of schematic diagram of the result that Fig. 2 d is shown by display module through contents selection module, the process of content recognition editor module for image described in Fig. 2 c;
Fig. 3 a is the interface schematic diagram of another self-defined symbol of the present invention and alignment processing key element thereof;
Fig. 3 b draws self-defined symbol described in Fig. 3 a to mark a kind of schematic diagram of local content on the original image that shows at smart mobile phone by writing pencil mode;
A kind of schematic diagram of the result that Fig. 3 c is shown by display module through contents selection module, the process of content recognition editor module for image described in Fig. 3 b;
Fig. 4 is a kind of schematic flow sheet of picture material recognition methods of the present invention.
Embodiment
Below in conjunction with drawings and the specific embodiments, the present invention is further described in more detail.
Fig. 1 is the one composition schematic diagram of picture material recognition device of the present invention, see Fig. 1, this device mainly comprises: symbol custom block 101, characteristic extracting module 102, image collection module 103, characteristic matching module 104, contents selection module 105, content recognition editor module 106, display module 107.
Described symbol custom block 101, for being obtained from define symbol, arranges process key element corresponding to described self-defined symbol according to user instruction.
Concrete, the mode that described symbol custom block 101 is obtained from define symbol can have multiple, and the corresponding corresponding acquisition module of difference, such as can comprise the Arbitrary Term in the module of following (11) ~ (13):
(11) module of the hand-written self-defined symbol inputted for the input equipment obtaining smart machine.Described self-defined symbol can be such as self-defining shape: as triangle, square, circular, oval; Also can be self-defining linear: as wave, broken line, straight line, bilinear, two wave; Also can be different symbols: as bracket, bracket, braces etc.The input equipment of described smart machine can be such as the touch keypad, hand-written screen etc. of mobile phone.User can input hand-written symbol by smart machine ends such as mobile phones: such as carry out hand-written symbol input at smart machines such as mobile phones, such as draw lines, the hand-written symbol such as shape and color, closed curve 200 as shown in Figure 2 a, and the horizontal line 300 shown in Fig. 3 a.
(12) for showing the shape of acquiescence, such as triangle, wave, dual slope etc., according to the selection instruction of user's input, select a module as the self-defined symbol got from the shape of acquiescence.
(13) for calling the picture of filming apparatus shooting containing self-defined symbol, from this picture, the module of described self-defined symbol is identified.This mode is a kind of typical hand-written symbol definition, such as: user takes pictures in content required, hand-written self-defined symbol is drawn by writing pencil, then the filming apparatus of mobile phone is utilized to take the photograph containing hand-written self-defined symbol, self-defined symbol in this photograph is identified, obtains this self-defined symbol.
The process key element that described self-defined symbol is corresponding comprises editor's action and object content form; Different self-defined meeting corresponding can arrange different process key elements.As illustrated in figures, such as described editor's action can be such as shear, preserve, editor's action such as translation; Described object content form refers to which kind of content-form image content is changed into, can be such as such as word, picture, exercise question of filling a vacancy, select exercise question, judge exercise question, calculation question etc.
Described characteristic extracting module 102 for extracting self-defined symbolic feature from self-defined symbol.Such as, for the self-defined symbol got, carry out feature extraction, described feature is shape, color, the thickness of such as lines, length, regular shape, amplitude, quantity etc.For different self-defined symbols, can pre-set and store different feature extraction modes, the concrete Feature Extraction Technology for each self-defined symbol can adopt existing Feature Extraction Technology scheme.
Described image collection module 103 is for obtaining the image with self-defined symbol.The obtain manner that described image collection module 103 is obtained from the image of define symbol also can have multiple, also can distinguish corresponding different modules, such as, can comprise the Arbitrary Term in the module of following (31) ~ (34):
(31) there is for calling filming apparatus (camera as smart mobile phone) shooting the image of self-defined symbol, the image of shooting being sent to the module of characteristic matching module.
Before this, user is needed to draw in required content of taking pictures and the same or similar pattern of self-defined symbol, as shown in Figure 2 b, user irises out a multiple-choice question with closed meander line on a paper, and the present invention afterwards can to take pictures the paper image obtaining and have this closed meander line with mobile phone camera; Or, for the content in required content of taking pictures, if having printed by forms such as printers the content comprising this self-defined symbol before, then without the need to manually being drawn by user again.
(32) for obtaining the image file in smart machine file system, this image file is sent to the module of described characteristic matching module.If preserve the image file containing self-defined symbol shown in Fig. 2 b in the file system of such as smart machine, then directly read this image file, can take.
(33) for intercepting the image of screen of intelligent device display, this image is sent to the module of described characteristic matching module.This module refers to and also by shooting push button, as long as show the image with all self-defined symbols as shown in Figure 2 b on user mobile phone screen, then can not take, but intercept this screen picture, pass to described characteristic matching module.
(34) obtaining user to marking operation screen showing image for calling screen interface, the screen picture after labelling being sent to the module of described characteristic matching module.This scene refers to, user mobile phone have taken a pictures, as a paper, but paper does not mark self-defined symbol, then user also directly can open this picture on mobile phone, utilize the input equipments such as writing pencil to mark on this picture and come from define symbol, the screen picture (as shown in Figure 2 b image) after labelling just can be sent to the module of described characteristic matching module by the present invention afterwards.
Described characteristic matching module 104, for being mated with obtained image by described self-defined symbolic feature, matches described self-defined symbolic feature and position in the picture thereof from this image.Namely to find the operating area of subsequent operation, wherein specifically can adopt fuzzy matching technology, the content that intelligent fuzzy coupling is corresponding.
Described contents selection module 105, for from described image, chooses the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place.Concrete mode of choosing also corresponding self-defined symbol, also corresponding corresponding module, such as described contents selection module 105 comprises with the Arbitrary Term in lower module (51) ~ (53):
(51) for when described self-defined symbol is characterized as closed figure, such as closed curve as shown in Figure 2, chooses the module of the image content in this closed figure automatically.
(52) for when described self-defined symbol is characterized as linear, such as horizontal line as shown in Figure 3, the high and spacing according to the row of setting, obtain this linear on the module meeting the image content of the high and pitch requirements of described row.User can manually or the row of Lookup protocol acquiescence be high and spacing size herein.
(53) for when described self-defined symbol is characterized as linear, the module of the image content in region between white space above this horizontal line to horizontal line is obtained.
More specifically, before contents selection module carries out contents selection, operating area can also be carried out convergent-divergent to occupy on the prescribed percentage of screen (such as 90%) to make described self-defined symbol area, thus make follow-up selection operation more accurate.
More specifically, if when described self-defined symbol is characterized as linear, then after can also carrying out the image procossing such as fuzzy, binaryzation to image further, then obtain the image content in region between white space above this horizontal line to horizontal line.
Described content recognition editor module 106, for according to process key element corresponding to described self-defined symbol, carries out identifying and corresponding editing and processing to obtained image content.
Described display module 107 is for showing editing and processing result.For editing and processing result, can also further operate according to user instruction and manage, the operations such as such as typesetting.
The process key element that the concrete process operation of described content recognition editor module 106 is corresponding to described self-defined symbol is relevant.The process key element that described self-defined symbol is corresponding comprises editor's action and object content form; Different self-defined meeting corresponding can arrange different process key elements.As illustrated in figures, such as described editor's action can be such as shear, preserve, editor's action such as translation; Described object content form refers to which kind of content-form image content is changed into, can be such as such as word, picture, exercise question of filling a vacancy, select exercise question, judge exercise question, calculation question etc.
Described content recognition editor module 106 specifically can comprise: (61) are for carrying out the module of corresponding editing operation to obtained image content according to editor's action corresponding to described self-defined symbol; (62) for the image content after described editing operation being converted into the module of described object content form.
Such as: as shown in Figure 2 a, the editor's action in the process key element that user is arranged is " preservation ", and object content form is " multiple-choice question ", and needs in described content recognition editor module to preserve processing logic corresponding to " multiple-choice question ".This processing logic is differentiation, and the corresponding corresponding processing logic of different process key elements, this processing logic can pre-set storage well.According to this processing logic, when described content recognition editor module carries out secondary editor to the image content obtained, identify and be somebody's turn to do specific features corresponding to " multiple-choice question ", the specific features parameter of being somebody's turn to do " multiple-choice question " can pre-set in a program, can directly call these characteristic parameters herein, whether characteristic parameter such as has option a, b, c etc., or option one, 2,3 etc.Such as shown in Fig. 2, these eigenwerts of option A, B, C can have been identified, extract image content on these eigenwerts as stem, extract the option of the image content after these eigenwerts as correspondence, then corresponding typesetting process is carried out, the first row is placed on by the stem part 201 above option, using the content of option A, B, C as three row below option 202 is placed on, thus a composition multiple-choice question.And according to the requirement of editor's action " preservation ", demonstrating " preservation " button, user clicks " preservation " button then can save as a file by the multiple-choice question shown in Fig. 2 d.
Again such as, as shown in Figure 3 a, the editor's action in the process key element that user is arranged is " preservation ", and object content form is " topic of filling a vacancy ", and needs in described content recognition editor module to preserve processing logic corresponding to " topic of filling a vacancy ".This processing logic is differentiation, and the corresponding corresponding processing logic of different process key elements, this processing logic can pre-set storage well.According to this processing logic, when described content recognition editor module carries out secondary editor to the image content obtained, identify and be somebody's turn to do specific features corresponding to " topic of filling a vacancy ", the specific features parameter of being somebody's turn to do " topic of filling a vacancy " can pre-set in a program, can directly call these characteristic parameters herein, whether characteristic parameter such as has horizontal line, bracket etc., such as shown in Fig. 3 b, this eigenwert of horizontal line can have been identified, then show corresponding topic of filling a vacancy, and content corresponding for option is carried out editing and composing, namely before the image content 301 extracted before horizontal line is placed on, after image content 302 on horizontal line part is placed on, thus a composition topic of filling a vacancy.And according to the requirement of editor's action " preservation ", demonstrating " preservation " button, user clicks " preservation " button then can save as a file by the topic of filling a vacancy shown in Fig. 3 c.In embodiment herein, also further the content in filling a vacancy can be carried out Text region, change into editable computer character and be placed in content 302.
Same reason, again such as, as shown in Figure 2 a, if the editor's action in the process key element that user is arranged is " shearing ", object content form is " word ", then described content recognition editor module needs the image content to obtaining to carry out secondary editor, image content is converted into the word of computing machine codified, put into the shear plate of smart machine, this section of word, in follow-up " sticky note " operation that directly can utilize smart machine, is glued note in any document by user.If the editor's action in the process key element that user is arranged is " translation ", object content form is " word ", then described content recognition editor module needs the image content of acquisition to translate into target language as English, then this section of English is put into the shear plate of smart machine, this section of English letter, in follow-up " sticky note " operation that directly can utilize smart machine, is glued note in any document by user.If the object content form in the process key element that user is arranged is " picture ", then described content recognition editor module needs image content described in extracting directly, retains the picture format of this image content.
Corresponding with above-mentioned picture material recognition device, the invention also discloses a kind of picture material recognition methods.Fig. 4 is a kind of process flow diagram of picture material recognition methods of the present invention.See Fig. 4, the method comprises:
Step 401: be obtained from define symbol feature, arranges process key element corresponding to described self-defined symbol according to user instruction.Specifically can any number of by following 3 kinds of modes, be obtained from define symbol feature:
411) the hand-written self-defined symbol that the input equipment obtaining smart machine inputs.Such as carry out hand-written symbol input at smart machines such as mobile phones, such as, draw lines, the hand-written symbol such as shape and color, closed curve 200 as shown in Figure 2 a, and the horizontal line 300 shown in Fig. 3 a.
412) show the shape of acquiescence, such as triangle, wave, dual slope etc., according to the selection instruction of user's input, from the shape of acquiescence, select one as the self-defined symbol got.Such as system can show the shape of acquiescence, such as triangle, wave, dual slope etc., and user can select from the shape of system default.
413) call the picture of filming apparatus shooting containing self-defined symbol, from this picture, identify described self-defined symbol.Such as: a) user takes pictures in content required, draws hand-written symbol by writing pencil, the b) photograph of shooting containing hand-written symbol, therefrom identify described self-defined symbol, concrete recognition method can with reference to prior art.
Described process key element corresponding to described self-defined symbol is set according to user instruction, specifically can shown in reference diagram 2a and Fig. 3 a, can self-defined different operational processes key element to different self-defined symbolic features.The process key element that described self-defined symbol is corresponding comprises editor's action and object content form, and wherein: 1) for the self-defined symbol identified, user can set different editor's actions, such as: shear, preserves, translation etc.2) select corresponding object content form, such as: word, picture, exercise question of filling a vacancy, selects exercise question, judges exercise question, calculation question etc.
Step 402: extract self-defined symbolic feature in self-defined symbol.Such as, for the self-defined symbol got, carry out feature extraction, described feature is shape, color, the thickness of such as lines, length, regular shape, amplitude, quantity etc.For different self-defined symbols, can pre-set and store different feature extraction modes, the concrete Feature Extraction Technology for each self-defined symbol can adopt existing Feature Extraction Technology scheme.
Step 403: obtain the image with self-defined symbol.Concrete obtain manner also can have multiple, and what such as can comprise in following (431) ~ (434) is any number of:
(431) image that filming apparatus (camera as smart mobile phone) shooting has self-defined symbol is called, as the image obtained.
Before this, user is needed to draw in required content of taking pictures and the same or similar pattern of self-defined symbol, as shown in Figure 2 b, user irises out a multiple-choice question with closed meander line on a paper, and the present invention afterwards can to take pictures the paper image obtaining and have this closed meander line with mobile phone camera; Or, for the content in required content of taking pictures, if having printed by forms such as printers the content comprising this self-defined symbol before, then without the need to manually being drawn by user again.
(432) image file in smart machine file system is obtained, as the image obtained.If preserve the image file containing self-defined symbol shown in Fig. 2 b in the file system of such as smart machine, then directly read this image file, can take.
(433) image of screen of intelligent device display is intercepted, as the image obtained.The manner refers to and also by shooting push button, as long as show the image with all self-defined symbols as shown in Figure 2 b on user mobile phone screen, then can not take, but intercept this screen picture, pass to described characteristic matching module.
(434) call screen interface and obtain user to marking operation screen showing image, using the screen picture after labelling as the image obtained.This scene refers to, user mobile phone have taken a pictures, as a paper, but paper does not mark self-defined symbol, then user also directly can open this picture on mobile phone, utilize the input equipments such as writing pencil to mark on this picture and come from define symbol, the screen picture (as shown in Figure 2 b image) after labelling just can be sent to the module of described characteristic matching module by the present invention afterwards.
Step 404: mated with obtained image by described self-defined symbolic feature, matches described self-defined symbolic feature and position in the picture thereof from this image.Namely to find the operating area of subsequent operation, wherein specifically can adopt fuzzy matching technology, the content that intelligent fuzzy coupling is corresponding.
Step 405: from described image, chooses the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place.Concrete mode of choosing also corresponding self-defined symbol, what such as comprise in following several mode is any number of:
(451) when described self-defined symbol is characterized as closed figure, such as closed curve as shown in Figure 2, chooses the image content in this closed figure automatically.
(452) when described self-defined symbol is characterized as linear, such as horizontal line as shown in Figure 3, the high and spacing according to the row of setting, obtain this linear on the image content meeting the high and pitch requirements of described row.User can manually or the row of Lookup protocol acquiescence be high and spacing size herein.
(453) when described self-defined symbol is characterized as linear, the module of the image content in region between white space above this horizontal line to horizontal line is obtained.
More specifically, before contents selection module carries out contents selection, operating area can also be carried out convergent-divergent to occupy on the prescribed percentage of screen (such as 90%) to make described self-defined symbol area, thus make follow-up selection operation more accurate.
More specifically, if when described self-defined symbol is characterized as linear, then after can also carrying out the image procossing such as fuzzy, binaryzation to image further, then obtain the image content in region between white space above this horizontal line to horizontal line.
Step 406: according to the process key element that described self-defined symbol is corresponding, carries out identifying and corresponding editing and processing to obtained image content, i.e. secondary editor.
Step 407: show editing and processing result.For editing and processing result, can also further operate according to user instruction and manage, the operations such as such as typesetting.
The process key element that the concrete process operation of described step 406 is corresponding to described self-defined symbol is relevant.The process key element that described self-defined symbol is corresponding comprises editor's action and object content form; Different self-defined meeting corresponding can arrange different process key elements.As illustrated in figures, such as described editor's action can be such as shear, preserve, editor's action such as translation; Described object content form refers to which kind of content-form image content is changed into, can be such as such as word, picture, exercise question of filling a vacancy, select exercise question, judge exercise question, calculation question etc.Described 406 specifically can comprise: (461) carry out corresponding editing operation according to editor's action corresponding to described self-defined symbol to obtained image content; Such as shear, preserve, translation; Convert specific topic type to, such as, when selecting exercise question, automatically content is cut into several pieces of contents, and generates option, for exercise question of filling a vacancy, can input frame be generated, direct input content.
(462) image content after described editing operation is converted into described object content form, such as, is converted into the topic type of word, picture or correspondence.
Preserve the scene of the multiple-choice question on paper below for user, the present invention will be further described, can see Fig. 2 a ~ 2d.Comprise:
Step a1: User Defined symbol, namely user is by writing pencil, screen is drawn the closed figure of red wave, as shown in Figure 2 a.
Step a2: the process key element that user selects this figure corresponding, as shown in Figure 2 a, user selects corresponding editing operation for " shearing ", and corresponding object content form is " multiple-choice question ".
Step a3: at the upper corresponding position graphing of reference object (as paper etc.).Such as shown in Fig. 2 b, user needs the regional area preserved on examination question paper, draws the wave closed, as the region 203 of Fig. 2 b with red pen.
Step a4: shooting photograph, such as user takes shown in Fig. 2 b with the content on the paper of red wave person's handwriting.
Step a5: through the process of described device of the present invention, changes into the form of multiple-choice question, shows with the region 203 that the waveform closed marks by user.As shown in Figure 2 d, only show the content on the paper of the correspondence that described red wave marks, according to the form of multiple-choice question, carry out storage editor etc. to content, user can directly use next time.
Preserve the scene of the topic of filling a vacancy on paper below again for user, the present invention will be further described, can see Fig. 3 a ~ 3c.Comprise::
Step b1: User Defined symbol.Namely user is by writing pencil, screen is drawn red horizontal line, as shown in Figure 3 a.
Step b2: select the process key element that this figure is corresponding, comprising selecting corresponding editing operation to be " shearing "; Select corresponding object content form for " topic of filling a vacancy ".
Step b3: shooting image or open image file and mark.Such as user is with camera shooting image or open the red brush of image file user on screen at iso-surface patch horizontal line under topic of filling a vacancy.
Step b4: through the process of described device of the present invention, the form of the topic that becomes to fill a vacancy with the regioinvertions that red horizontal line marks by user, shows.As shown in Figure 3 c, only show the content on the paper of the correspondence that described red horizontal line marks, according to the form of topic of filling a vacancy, carry out storage editor etc. to content, user can directly use next time.
Utilize the present invention, following beneficial effect can also be produced:
(1) user can self-defined symbol content, and described content comprises, lines, pattern, color, symbol.And the process key element corresponding to different self-defined symbol, such as: cut out, preserve, translation, converts multiple-choice question to, topic of filling a vacancy, True-False, the special efficacy etc. of font.Extensibility is strong, and the field of use is very wide, need not be confined to the normal form image that certain is fixing.
(2) the present invention can choose the region content of local automatically, directedly can take the local content identified in photograph, instead of full content.Extracts and the cutting of local can be carried out to the image content in photograph.
(3) the present invention to the content of the local chosen, can carry out secondary editor, according to self-defining symbol content, performs corresponding process key element, generates corresponding object content form, very convenient operation.
Be placed in concrete application scenarios, invention particularly provides a kind of novel shooting recording mode, when user is when taking, reference object or screen directly draw specific hand-written symbol, after device of the present invention recognizes these special symbols, automatically can process and record content corresponding to this symbol.Such as draw particular color wave, system can identify the content on this wave in certain limit, and can carry out Classification Management to content.In this way, can effective record notes content.Make the content of shooting more accurate, and pre-service can be carried out to content.
In addition, each functional module in each embodiment of device of the present invention can be integrated in a processing unit, also can be that the independent physics of modules exists, also can two or more module integrations in a unit.Above-mentioned integrated unit both can adopt the form of hardware to realize, and the form of SFU software functional unit also can be adopted to realize.The functional module of described each embodiment can be positioned at a terminal or network node, or also can be distributed on multiple terminal or network node.
In addition, each embodiment of the present invention can be realized by the data processor performed as computing machine by data processing equipment.Obviously, data processor constitutes the present invention.In addition, program is read out storage medium or memory device (as hard disk and or internal memory) the middle execution by program being installed or copied to data processing equipment by direct by the data processor be usually stored in a storage medium.Therefore, such storage medium also constitutes the present invention.Storage medium can use the recording mode of any type, such as paper storage medium (as paper tape etc.), magnetic storage medium (as floppy disk, hard disk, flash memory etc.), optical storage media (as CD-ROM etc.), magnetic-optical storage medium (as MO etc.) etc.
Therefore the invention also discloses a kind of storage medium, wherein store data processor, this data processor is for performing any one embodiment of said method of the present invention.
In addition, method step of the present invention is except realizing with data processor, can also be realized by hardware, such as, can be realized by logic gate, switch, special IC (ASIC), programmable logic controller (PLC) and embedding microcontroller etc.Therefore this hardware that can realize the method for the invention also can form the present invention.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (10)
1. a picture material recognition device, is characterized in that, comprising:
Symbol custom block, for being obtained from define symbol, arranges process key element corresponding to described self-defined symbol according to user instruction;
Characteristic extracting module, for extracting self-defined symbolic feature from self-defined symbol;
Image collection module, for obtaining the image with self-defined symbol;
Characteristic matching module, for being mated with obtained image by described self-defined symbolic feature, matches described self-defined symbolic feature and position in the picture thereof from this image;
Contents selection module, for from described image, chooses the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place;
Content recognition editor module, for according to process key element corresponding to described self-defined symbol, carries out identifying and corresponding editing and processing to obtained image content;
Display module, for showing editing and processing result.
2. device according to claim 1, is characterized in that, described symbol custom block comprises with the Arbitrary Term in lower module:
For the module of the hand-written self-defined symbol that the input equipment obtaining smart machine inputs;
For showing the shape of acquiescence, according to the selection instruction of user's input, from the shape of acquiescence, select a module as the self-defined symbol got;
For calling the picture of filming apparatus shooting containing self-defined symbol, from this picture, identify the module of described self-defined symbol.
3. device according to claim 1, is characterized in that,
The process key element that described self-defined symbol is corresponding comprises editor's action and object content form;
Described content recognition editor module comprises:
For carrying out the module of corresponding editing operation to obtained image content according to editor's action corresponding to described self-defined symbol;
For the image content after described editing operation being converted into the module of described object content form.
4. device according to claim 1, is characterized in that, described image collection module comprises with the Arbitrary Term in lower module:
There is for calling filming apparatus shooting the image of self-defined symbol, the image of shooting being sent to the module of characteristic matching module;
For obtaining the image file in smart machine file system, this image file is sent to the module of described characteristic matching module;
For intercepting the image of screen of intelligent device display, this image is sent to the module of described characteristic matching module;
Obtaining user to marking operation screen showing image for calling screen interface, the screen picture after labelling being sent to the module of described characteristic matching module.
5. device according to claim 1, is characterized in that, described contents selection module comprises with the Arbitrary Term in lower module:
For when described self-defined symbol is characterized as closed figure, automatically choose the module of the image content in this closed figure;
For when described self-defined symbol is characterized as linear, according to the high and spacing of row of setting, obtain this linear on the module meeting the image content of the high and pitch requirements of described row;
For when described self-defined symbol is characterized as linear, obtain the module of the image content in region between white space above this horizontal line to horizontal line.
6. a picture material recognition methods, is characterized in that, comprising:
Be obtained from define symbol, process key element corresponding to described self-defined symbol is set according to user instruction;
Self-defined symbolic feature is extracted from self-defined symbol;
Obtain the image with self-defined symbol;
Described self-defined symbolic feature is mated with obtained image, from this image, matches described self-defined symbolic feature and position in the picture thereof;
From described image, choose the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place;
According to the process key element that described self-defined symbol is corresponding, carry out identifying and corresponding editing and processing to obtained image content;
Show editing and processing result.
7. method according to claim 6, is characterized in that, described in be obtained from define symbol and comprise with the Arbitrary Term under type:
The hand-written self-defined symbol that the input equipment obtaining smart machine inputs;
Show the shape of acquiescence, according to the selection instruction of user's input, from the shape of acquiescence, select one as the self-defined symbol got;
Call the picture of filming apparatus shooting containing self-defined symbol, from this picture, identify described self-defined symbol.
8. method according to claim 6, is characterized in that,
The process key element that described self-defined symbol is corresponding comprises editor's action and object content form;
The described process key element corresponding according to described self-defined symbol, carry out identifying and corresponding editing and processing to obtained image content, specifically comprise: the editor action corresponding according to described self-defined symbol carries out corresponding editing operation to obtained image content; Image content after described editing operation is converted into described object content form.
9. method according to claim 6, is characterized in that, described acquisition has the image of self-defined symbol, specifically comprises with the Arbitrary Term under type:
Call the image that filming apparatus shooting has self-defined symbol, using the image of shooting as the image got;
Obtain the image file in smart machine file system, using this image file as the image got;
Intercept the image of screen of intelligent device display, using this image as the image got;
Call screen interface and obtain user to marking operation screen showing image, using the screen picture after labelling as the image got.
10. method according to claim 6, is characterized in that, described from described image, chooses the image content corresponding with described self-defined symbolic feature at described self-defined symbol feature locations place, specifically comprises with the Arbitrary Term under type:
When described self-defined symbol is characterized as closed figure, automatically choose the image content in this closed figure;
When described self-defined symbol is characterized as linear, according to the high and spacing of row of setting, obtain this linear on the image content meeting the high and pitch requirements of described row;
When described self-defined symbol is characterized as linear, obtain the image content in region between white space above this horizontal line to horizontal line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510240225.9A CN104951749B (en) | 2015-05-12 | 2015-05-12 | Picture material identification device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510240225.9A CN104951749B (en) | 2015-05-12 | 2015-05-12 | Picture material identification device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104951749A true CN104951749A (en) | 2015-09-30 |
| CN104951749B CN104951749B (en) | 2018-07-20 |
Family
ID=54166391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510240225.9A Active CN104951749B (en) | 2015-05-12 | 2015-05-12 | Picture material identification device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104951749B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105787480A (en) * | 2016-02-26 | 2016-07-20 | 广东小天才科技有限公司 | Test question shooting method and device |
| CN105825721A (en) * | 2016-03-16 | 2016-08-03 | 广东小天才科技有限公司 | Questioning method and device via shooting and intelligent equipment |
| CN106446884A (en) * | 2016-09-19 | 2017-02-22 | 广东小天才科技有限公司 | Method and device for rapidly capturing images |
| CN111344735A (en) * | 2017-09-13 | 2020-06-26 | 深圳传音通讯有限公司 | Picture editing method, mobile terminal and readable storage medium |
| WO2020258523A1 (en) * | 2019-06-25 | 2020-12-30 | 浙江飙速教育科技有限公司 | Exercise acquiring method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101312482A (en) * | 2007-05-22 | 2008-11-26 | 夏普株式会社 | Image output system and image processing apparatus |
| CN101626448A (en) * | 2008-07-10 | 2010-01-13 | 富士施乐株式会社 | Image processing apparatus and image processing method |
| CN202548962U (en) * | 2012-05-10 | 2012-11-21 | 吴方 | Recycling device based on image recognition device |
| CN103247037A (en) * | 2012-02-10 | 2013-08-14 | 联想(北京)有限公司 | Image processing method, device and electronic device |
| US20130308825A1 (en) * | 2011-01-17 | 2013-11-21 | Panasonic Corporation | Captured image recognition device, captured image recognition system, and captured image recognition method |
-
2015
- 2015-05-12 CN CN201510240225.9A patent/CN104951749B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101312482A (en) * | 2007-05-22 | 2008-11-26 | 夏普株式会社 | Image output system and image processing apparatus |
| CN101626448A (en) * | 2008-07-10 | 2010-01-13 | 富士施乐株式会社 | Image processing apparatus and image processing method |
| US20130308825A1 (en) * | 2011-01-17 | 2013-11-21 | Panasonic Corporation | Captured image recognition device, captured image recognition system, and captured image recognition method |
| CN103247037A (en) * | 2012-02-10 | 2013-08-14 | 联想(北京)有限公司 | Image processing method, device and electronic device |
| CN202548962U (en) * | 2012-05-10 | 2012-11-21 | 吴方 | Recycling device based on image recognition device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105787480A (en) * | 2016-02-26 | 2016-07-20 | 广东小天才科技有限公司 | Test question shooting method and device |
| CN105825721A (en) * | 2016-03-16 | 2016-08-03 | 广东小天才科技有限公司 | Questioning method and device via shooting and intelligent equipment |
| CN106446884A (en) * | 2016-09-19 | 2017-02-22 | 广东小天才科技有限公司 | Method and device for rapidly capturing images |
| CN111344735A (en) * | 2017-09-13 | 2020-06-26 | 深圳传音通讯有限公司 | Picture editing method, mobile terminal and readable storage medium |
| CN111344735B (en) * | 2017-09-13 | 2023-08-08 | 深圳传音通讯有限公司 | Picture editing method, mobile terminal and readable storage medium |
| WO2020258523A1 (en) * | 2019-06-25 | 2020-12-30 | 浙江飙速教育科技有限公司 | Exercise acquiring method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104951749B (en) | 2018-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9081759B2 (en) | Image processing apparatus, image processing system and image processing method | |
| CN103218595B (en) | The recognition methods of a kind of terminal and Quick Response Code | |
| US10353997B1 (en) | Freeform annotation transcription | |
| CN104951749A (en) | Image content recognition device and image content recognition method | |
| CN103020619B (en) | A kind of method of handwritten entries in automatic segmentation electronization notebook | |
| US20140270536A1 (en) | Systems and methods for classifying objects in digital images captured using mobile devices | |
| US6351559B1 (en) | User-enclosed region extraction from scanned document images | |
| US20170220858A1 (en) | Optical recognition of tables | |
| CN103034856B (en) | The method of character area and device in positioning image | |
| CN112669515B (en) | Bill image recognition method and device, electronic equipment and storage medium | |
| CN109657669B (en) | Intelligent electronic seal extraction method based on image processing | |
| Hung et al. | Implementing an android application for automatic vietnamese business card recognition | |
| CN108564079A (en) | A kind of portable character recognition device and method | |
| CN111695518B (en) | Method and device for labeling structured document information and electronic equipment | |
| KR20150091948A (en) | A system for recognizing a font and providing its information and the method thereof | |
| WO2015032308A1 (en) | Image recognition method and user terminal | |
| JP7379876B2 (en) | Character recognition device, document file generation method, document file generation program | |
| WO2014086266A1 (en) | Professional notebook convenient for electronization and method for displaying electronic thumbnail thereof | |
| WO2023051384A1 (en) | Display method, information sending method, and electronic device | |
| Bhaskar et al. | Implementing optical character recognition on the android operating system for business cards | |
| JP7318289B2 (en) | Information processing device and program | |
| CN113936187A (en) | Text image synthesis method and device, storage medium and electronic equipment | |
| Rao et al. | MTESSERACT: An Application for Form Recognition in Courier Services | |
| US20190377941A1 (en) | Character recognition apparatus and character recognition method | |
| WO2008081666A1 (en) | Document reader apparatus |
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 |