CN101673408B - Method and device for embedding character information in shape recognition result - Google Patents
Method and device for embedding character information in shape recognition result Download PDFInfo
- Publication number
- CN101673408B CN101673408B CN2008101197914A CN200810119791A CN101673408B CN 101673408 B CN101673408 B CN 101673408B CN 2008101197914 A CN2008101197914 A CN 2008101197914A CN 200810119791 A CN200810119791 A CN 200810119791A CN 101673408 B CN101673408 B CN 101673408B
- Authority
- CN
- China
- Prior art keywords
- input
- stroke
- word message
- literal
- shape
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Character Discrimination (AREA)
Abstract
The invention discloses a method and a device for embedding character information in a shape recognition result, which are used for solving the problem that the recognition accuracy of the prior method needs to be improved. The method comprises the following steps: acquiring stroke information input in a handwriting way in a stroke input mode, and acquiring the character information input in a character input mode; performing shape stroke recognition on the acquired stoke information to obtain one or more shape stroke recognition results; and determining position relationships between the character information and the shape stroke recognition results according to the input position of the character information, and embedding the character information into the shape stroke recognition results according to the position relationships. When a user utilizes handwriting input equipment, and needs to input mixed information comprising graphics, formats, characters and the like, due to no need of classifying shape strokes and character strokes, the method avoids the conditions of character stroke loss and the like caused by classification, and improves the recognition accuracy and the input efficiency of the character information in a handwriting input process.
Description
Technical field
The present invention relates to the information discriminating technology field, particularly relate to a kind of method and device that in shape recognition result, embeds Word message.
Background technology
In handwriting input identification field, can relate to the mixing identification of multiple handwriting information such as figure, form, text usually.For example; User's handwriting input a form; And in said form handwriting input literal, the purpose of recognition system is all to identify the form stroke and the strokes of characters of user's handwriting input, converting computing machine to can recognition data information; And the Word message and the form data that identify carried out integration processing, Word message is embedded form show.
How above-mentioned multiple handwriting information is mixed identification and integration processing; Usual way is: all stroke information of gathering user's input; The calling classification algorithm is classified to all strokes according to stroke feature then; Be divided into different identification types; Generally can be divided into strokes of characters and non-legible stroke (non-legible stroke is designated hereinafter simply as the shape stroke), call different recognizers according to the kind of stroke again and discern respectively, obtaining can be by the shape stroke recognition result of computer Recognition and strokes of characters recognition result; At last, the strokes of characters recognition result being embedded the output of shape stroke recognition result shows.Wherein, for the identification of strokes of characters and non-legible stroke, multiple recognizer is arranged all separately.
Based on above-mentioned identification and integration processing method; When the user utilizes handwriting input device; When needing to import different shapes such as comprising figure, form and Word message; Recognition system utilizes said method just can identify the mixing stroke of user's input, and showing in the Word message embedding different shape wherein.But; When the strokes of characters characteristic was more similar with the shape stroke feature, sorting algorithm classified as this strokes of characters the part of shape stroke probably, thereby causes losing of strokes of characters; The system of influence can not correctly discern literal and figure, and then influences the integration processing of literal and figure.
For example; The shape stroke of user's handwriting input has constituted one close-shaped (for example form), the user in this is close-shaped again handwriting input strokes of characters, suppose that this strokes of characters is bigger; And intersect with the shape stroke or lean on very near; When system carries out the branch time-like to all strokes, may cause classification inaccurate, thereby influence the recognition accuracy of system.
In sum; When the user utilizes handwriting input device; When needing to import mixed informations such as comprising figure, form, literal; How to come out the information Recognition of user's input effectively, and embed Word message wherein in the different shape, existing handwriting input identification and integration method are still waiting to improve on the recognition accuracy of integral body.
Summary of the invention
Technical matters to be solved by this invention provides a kind of method and device that in shape recognition result, embeds Word message, to solve the problem that existing method has much room for improvement on recognition accuracy.
For solving the problems of the technologies described above,, the invention discloses following technical scheme according to specific embodiment provided by the invention:
A kind of method that in shape recognition result, embeds Word message comprises:
Obtain the stroke information of handwriting input under the stroke input pattern, and obtain the Word message of input under the literal input pattern;
The stroke information of obtaining is carried out the identification of shape stroke, obtain one or more shape stroke recognition results;
The position of confirming Word message and said shape stroke recognition result according to the input position of Word message concerns, and according to said position relation Word message is embedded shape stroke recognition result.
Wherein, the said Word message that obtains input under the literal input pattern specifically comprises: dynamic creation literal input frame under the literal input pattern; Be received in the Word message of said literal input frame input, said Word message is through calling various input method inputs, and said Word message comprises one or more characters.
Wherein, the said Word message that obtains input under the literal input pattern specifically comprises: dynamic creation stroke input frame under the literal input pattern; Be received in the strokes of characters information of said stroke input frame handwriting input; Said strokes of characters information is carried out strokes of characters identification, obtain the strokes of characters recognition result as the Word message that obtains, said Word message comprises one or more characters.
Wherein, said dynamic creation literal input frame/stroke input frame specifically comprises: according to the action message that mouse is pressed, drags, lifted, and dynamic creation literal input frame/stroke input frame.
Preferably, said literal input frame/stroke input frame is through outer a bit the hiding arbitrarily of click input frame, but the Word message that reservation has been imported.
Preferably, said obtaining after the Word message of under the literal input pattern, importing, also comprise: write down the positional information of said Word message and said literal input frame/upper left angle point of stroke input frame, this position will be as the input position of said Word message.
Preferably; Said input position according to Word message is confirmed the position relation of Word message and said shape stroke recognition result; And Word message is embedded shape stroke recognition result according to said position relation, specifically comprise: said shape stroke recognition result comprises close-shaped stroke and/or connecting line stroke; If the input position of Word message is positioned at the inside of close-shaped stroke, then the inside text attribute information of said Word message as said close-shaped stroke embedded; If the input position of Word message is positioned near the upper and lower, left and right side of connecting line stroke, then the text attribute information of said Word message as said connecting line stroke next door embedded; If the input position of Word message does not meet above-mentioned two kinds of situation, then with said Word message as independent Word message.
Preferably, said input position according to Word message is confirmed also to comprise after the shape stroke recognition result under this literal information: with said Word message and affiliated output of shape stroke recognition result or demonstration thereof.
The present invention also provides a kind of device that in shape recognition result, embeds Word message, comprising:
The stroke Input Control Element is used to obtain the stroke information of handwriting input under the stroke input pattern;
The literal Input Control Element is used to obtain the Word message of input under the literal input pattern;
The shape recognition unit is used for the stroke information that the stroke Input Control Element that obtains obtains is carried out the identification of shape stroke, obtains one or more shape stroke recognition results;
The data integration unit is used for confirming that according to the input position of Word message the position of Word message and said shape stroke recognition result concerns, and according to said position relation Word message is embedded shape stroke recognition result.
Wherein, said literal Input Control Element specifically comprises: first input frame is created subelement, is used for dynamic creation literal input frame under the literal input pattern; First literal receives subelement, is used to be received in the Word message of said literal input frame input, and said Word message is through calling various input method inputs, and said Word message comprises one or more characters.
Wherein, said literal Input Control Element specifically comprises: second input frame is created subelement, is used for dynamic creation stroke input frame under the literal input pattern; Second literal receives subelement; Be used to be received in the strokes of characters information of said stroke input frame handwriting input; And calling word stroke recognition unit carries out strokes of characters identification to said strokes of characters information; Receive the strokes of characters recognition result as the Word message that obtains from the strokes of characters recognition unit, said Word message comprises one or more characters.
Wherein, said first input frame is created the action message that subelement is pressed, drags, lifted according to mouse, dynamic creation literal input frame; Said second input frame is created the action message that subelement is pressed, drags, lifted according to mouse, dynamic creation stroke input frame.
Preferably, said literal input frame/stroke input frame is through outer a bit the hiding arbitrarily of click input frame, but the Word message that reservation has been imported.
Preferably, said device also comprises: record cell, be used to write down the positional information of said Word message and said literal input frame/upper left angle point of stroke input frame, and this position will be as the input position of said Word message.
Preferably, said data integration unit specifically comprises: travel through subelement, be used for each Word message and the input position thereof of traversal record; Judgment sub-unit is used for judging as follows according to the input position of each Word message: said shape stroke recognition result comprises close-shaped stroke and/or connecting line stroke; If the input position of Word message is positioned at the inside of close-shaped stroke, then the inside text attribute information of said Word message as said close-shaped stroke embedded; If the input position of Word message is positioned near the upper and lower, left and right side of connecting line stroke, then the text attribute information of said Word message as said connecting line stroke next door embedded; If the input position of Word message does not meet above-mentioned two kinds of situation, then with said Word message as only Word message.
Preferably, said device also comprises: output/display unit is used for said Word message and affiliated output of shape stroke recognition result or demonstration thereof.
According to specific embodiment provided by the invention, the present invention has following technique effect:
At first, the present invention proposes a kind of new solution thinking: entire identification process need not classified to shape stroke and strokes of characters, discerns and directly the stroke of user's handwriting input under the stroke input pattern is used as the shape stroke; When under the literal input pattern, the input information that gets access to is a Word message.Then, confirm the shape stroke recognition result under this literal information, just can this literal information be embedded the shape stroke recognition result under it exactly according to the input position of Word message.
When the user utilizes handwriting input device; When needing to import mixed informations such as comprising figure, form, literal; Said method is not owing to need classify to shape stroke and strokes of characters; Thereby strokes of characters that having avoided classification to cause such as loses at situation, has improved the recognition accuracy and the input efficiency of handwriting input process Chinese words information.Simultaneously, the accuracy rate and the input efficiency of the identification of shape stroke have been improved indirectly.
Secondly, the Word message that recognition system receives can be the Word message that calls various input method inputs, also can be the recognition result of handwriting stroke.And the input equipment of Word message can be not limited to keyboard, handwriting pad etc.Therefore, in the handwriting input field, the present invention is more flexible for the input method that the user provides.
Description of drawings
Fig. 1 is the embodiment of the invention one said a kind of method flow diagram that in shape recognition result, embeds Word message;
Fig. 2 is the embodiment of the invention two said a kind of method flow diagrams that in shape recognition result, embed Word message;
Fig. 3 is a stroke input synoptic diagram in the embodiment of the invention two;
Fig. 4 is the embodiment of the invention two a Chinese words input synoptic diagram;
Fig. 5 is the synoptic diagram of hidden text input frame in the embodiment of the invention two;
Fig. 6 is before discerning in the embodiment of the invention two and the contrast synoptic diagram after the identification;
Fig. 7 is all kinds of shape information classification synoptic diagram relevant with process flow diagram in the embodiment of the invention two;
Fig. 8 is the said a kind of structure drawing of device that in shape recognition result, embeds Word message of the embodiment of the invention;
Fig. 9 is the said a kind of structure drawing of device that in shape recognition result, embeds Word message of the preferred embodiment of the present invention.
Embodiment
For make above-mentioned purpose of the present invention, feature and advantage can be more obviously understandable, below in conjunction with accompanying drawing and embodiment the present invention done further detailed explanation.
Embodiment one:
With reference to Fig. 1, be the embodiment of the invention one said a kind of method flow diagram that in shape recognition result, embeds Word message.When the user utilizes handwriting input device, when input comprises mixed information such as figure, form, literal, discern processing as follows:
S101 obtains the stroke information of handwriting input under the stroke input pattern, and obtains the Word message of input under the literal input pattern; Wherein, said stroke input pattern and literal input pattern have the characteristic of mutual switching;
Present embodiment provides two kinds of input patterns, and a kind of is the stroke input pattern, and a kind of is the literal input pattern, and these two kinds of patterns can be switched each other, and switching way has multiple.For example; System default is the stroke input pattern; When just having switched to the literal input pattern through clicking after modes such as interface control, menu selection, double-click start the literal input pattern, the stroke input pattern of the system default that after closing the literal input pattern, just automaticallyed switch back; Perhaps, start a kind of pattern after, switch each other through shortcut.Present embodiment does not limit concrete switching mode at this, and present embodiment includes but not limited to this two kinds of input patterns.
When the user started stroke input pattern input information, what system obtained was the stroke information of user's handwriting input; When the user started literal input pattern input information, what system obtained was Word message.Wherein, said Word message comprises character or character string informations (character string is made up of a plurality of characters) such as literal and the punctuation mark of various language, and promptly each input possibly be a character or character string.Because it is uncertain that the user imports the sequencing of stroke information or Word message,,, then obtain this stroke information earlier if promptly the user imports stroke earlier under the stroke input pattern so the present invention does not limit the priority step of obtaining; If input characters under the literal input pattern then obtains this literal information earlier earlier.And the user possibly carry out repeatedly stroke input or literal input, and system will correspondingly repeatedly obtain operation.
Preferably, present embodiment provides two kinds of literal input patterns commonly used, and a kind of is to call the pattern that various input methods are imported, and a kind of is the pattern that can carry out the stroke input.Under first kind of pattern, the information that gets access to is the information that computing machine can Direct Recognition; Under second kind of pattern; What at first receive is stroke information; Because said stroke information is imported,, convert strokes of characters into Word message that computing machine can be distinguished so system can direct calling word stroke recognizer discern said stroke information under the literal input pattern.But no matter through which kind of pattern input, what system finally got access to all is the Word message that computing machine can be distinguished.Wherein, existing strokes of characters recognizer has multiple, and present embodiment adopts wherein a kind of can identification, is not described in detail in this.
S102 carries out the identification of shape stroke to said stroke information, obtains one or more shape stroke recognition results;
System directly is used as the shape stroke to all stroke information that the user repeatedly obtains under the stroke input pattern; Calling shape stroke recognizer discerns; And no longer earlier stroke is carried out Classification and Identification as prior art calling classification algorithm, because the Word message in the present embodiment is imported under the literal input pattern.
The basic process of identification shape stroke is: because the original handwriting of user's input may comprise a plurality of input shapes, therefore need the shape stroke of input be divided into groups, so that each group stroke can constitute a possible shape; Then, extract some characteristic informations of every group of stroke, and analyze further and contrast, judge which classification is the shape that every group of stroke constitutes belong to all kinds of shape informations that internal system is stored in advance.Therefore, shape stroke recognition result possibly be a circle, or a rectangle, or the connecting line of a band arrow, or the like.And one time identifying possibly comprise a plurality of shape stroke recognition results.
Because it is multiple that existing shape stroke recognizer has, present embodiment adopts wherein a kind of can identification, so be not described in detail in this.
S103 confirms the shape stroke recognition result under this literal information according to the input position of Word message.Promptly confirm the position relation of Word message and said shape stroke recognition result, and Word message is embedded shape stroke recognition result according to said position relation according to the input position of Word message.
After obtaining shape stroke recognition result and Word message that computing machine can distinguish; These data messages are integrated; The Word message (being character or character string) of input is at every turn embedded suitable shape stroke recognition result; Revert to the position relation of user when importing, just can obtain the result of user's needs.
Because the shape stroke of user's input and Word message all have certain position relation,,, just can confirm which shape stroke recognition result is this literal information belong to through comparing with the position of shape stroke according to the input position of Word message.For example, literal is positioned at rectangle, or is positioned at the right of line segment, or the like.
Above-mentioned entire identification process need not classified to shape stroke and strokes of characters, discerns and directly the stroke of user's handwriting input is used as the shape stroke; When under the literal input pattern, the input information that gets access to is a Word message.Then, confirm the shape stroke recognition result under this literal information according to the input position of Word message.When the user utilizes handwriting input device; When needing to import mixed informations such as comprising figure, form, literal; Said recognition methods is not owing to need classify to shape stroke and strokes of characters; Thereby strokes of characters that having avoided classification to cause such as loses at situation, has improved the recognition accuracy and the input efficiency of handwriting input process Chinese words information.Simultaneously, the accuracy rate and the input efficiency of shape recognition have been improved indirectly.
And the Word message that recognition system receives can be the Word message that calls various input method inputs, also can be the recognition result of handwriting stroke.And the input equipment of Word message can be not limited to keyboard, handwriting pad etc.Therefore, in the handwriting input field, the present invention is more flexible for the input method that the user provides.
Embodiment two:
Based on the content of the foregoing description one, present embodiment will be elaborated through instantiation.
With reference to Fig. 2, be the embodiment of the invention two said a kind of method flow diagrams that in shape recognition result, embed Word message.Be that example describes with input characters behind elder generation's input shape stroke below.
S201 under the stroke input state, obtains the stroke information of user's handwriting input in drawing area;
With reference to shown in Figure 3, be stroke input synoptic diagram.The upper right corner is a control regions among the figure, and a uppermost control is represented the stroke input, and when this control was clicked, expression can be at drawing area handwriting input stroke, and the input state of this moment is equivalent to the stroke input pattern described in the embodiment one.
If the stroke input state that system default is clicked for this control, after then the user opens drawing area, just can the handwriting input stroke.As shown in the figure, what the user imported is two rectangles and the connecting line with downward arrow.After the user has imported, the stroke information of system acquisition user's input, and note.
S202 switches to the literal input state, and the optional position in the drawing area that mouse is pressed dynamically generates a literal input frame and is used for importing and editing character;
With reference to shown in Figure 4, be literal input synoptic diagram." A " control of control regions is represented the literal input among the figure, and said through clicking " A " control has just switched to the literal input state from the stroke input state, and said literal input state is equivalent to the literal input pattern described in the embodiment one.Certainly, switching mode is not limited to said control mode, also can switch through menu selection or other modes.
After clicking " A " control, press, drag and lift mouse, just can dynamically generate a literal input frame, and in input frame, focus is set so that user inputs character in the optional position of drawing area.The size of this input frame can dragging and change with mouse.This literal input frame can receive the input information that various input methods send over, and its input equipment can be not limited to keyboard, handwriting pad etc.If handwriting pad, its input stroke also is to convert character string information to through hand-writing input method to send to the literal input frame, if keyboard can send character information through the input method relevant with keyboard and give the literal input frame.Therefore with respect to existing recognition methods, it is flexible more, accurate in the literal input.As shown in Figure 4, the user has imported " input data " four words.
Certainly, but the action message of pressing, drag, lifting according to user's mouse stroke input frame of dynamic creation also, and this input frame can receive the strokes of characters of user's handwriting input.System can calling word stroke recognizer discern the strokes of characters information that receives, and is converted into discernible character of computing machine or character string information.
S203, user's mouse is clicked the positional information pressed and input in drawing area respective symbols string as one group of data recording to system;
In the present embodiment, said positional information is meant the positional information of the upper left angle point of said literal input frame, and this position will be as the input position of character string information in the input frame, and certain said positional information can also have other setting meanss.
S204, whether judges need continue input stroke or character string in other positions in drawing area, continues the input stroke if desired, then carries out S201, jumps to S204 then; Continue input of character string if desired, then carry out S202 and S203, and continue S204 and judge,, carry out S205 up to the click recognition control;
Preferably, present embodiment also provides the hidden function of literal input frame.With reference to shown in Figure 5, be the synoptic diagram of hidden text input frame.The user behind input " input data " several words, after the character string information input of confirming this place is accomplished, click the literal input frame outer more arbitrarily, system will hide this literal input frame, but the information of input will be presented at this place with the form of character string.
In addition, among Fig. 5, the user again below rectangle in imported " output result " four words, simultaneously the literal input frame is hidden.System will write down two groups of character string informations and input position information thereof, i.e. dynamic creation several text input frames, will write down several groups of data.
S205, after the user confirms that all strokes and Word message are all imported completion, the click recognition control, system begins all information of user's input are discerned;
S206, system carry out the identification of shape stroke to the stroke information of handwriting input earlier, and each character string information that will write down is then integrated with each shape stroke recognition result, promptly confirms according to the positional information of each character string which shape stroke recognition result it belongs to;
Said integration process is following:
1) each character string and the positional information thereof of traversal S203 record;
2) judge according to the positional information of each character string which shape stroke recognition result it belongs to;
3) according to the result who judges, with the word attribute of each character string as its correspondingly-shaped stroke recognition result.
Based on said integration process since the classification of shape stroke recognition result is had a variety of, can't be exhaustive, so present embodiment will specify above-mentioned step 2 with all kinds of the be shaped as examples relevant with process flow diagram).
In the present embodiment, all kinds of shape informations relevant with process flow diagram mainly are summed up as two types: close-shaped and connecting line, and wherein close-shaped can also further being subdivided into justified and polygon, and it is classified more specifically can be with reference to shown in Figure 7.
According to the classification of shape stroke recognition result shown in Figure 7, can analyze and have following several kinds of situation between all kinds of shapes and the literal:
Situation 1: literal can be positioned at close-shaped inside;
Situation 2: literal can be positioned at the upper and lower, left and right side of connecting line (or band arrow connecting line);
Situation 3: literal can be positioned at any place;
And close-shaped a common characteristic is arranged when drawing is exactly that the person's handwriting that constitutes them always is made up of one group of end to end stroke group; The drafting of connecting line always is made up of some corresponding connecting line stroke groups, so also has above-mentioned three kinds of corresponding relations between the position of each character string of described each shape stroke recognition result of present embodiment and record.
Correspondingly; Present embodiment is when " judging according to the positional information of each character string which shape stroke recognition result it belongs to "; Mainly be the ownership that the upper left angle point that utilizes the user to import the literal input frame of this character string is confirmed each character string, concrete decision procedure is following:
If this point is positioned at the inside of close-shaped stroke, then it is recorded as the inside text attribute information of close-shaped recognition result;
If this point is positioned near the upper and lower, left and right side of connecting line (or band arrow connecting line) stroke, then it is recorded as the text attribute information on connecting line (or band arrow connecting line) recognition result next door;
If this point does not meet above-mentioned situation, then can be with this corresponding characters string as an independent Word message.
S207 is with shape recognition result and as the output of the Word message attribute of this shape recognition result or show to screen.
With reference to shown in Figure 6, be before the identification and the contrast synoptic diagram after the identification.The 4th of control regions the control (a triangle control) is an identification control among the figure, the startup of expression identifying.With reference to right partly figure, after the user confirms not have input information again, to click and should discern control, the net result of identification can be with reference to shown in half figure of a left side.
To the explanation of said method embodiment, the present invention also provides a kind of device embodiment that in shape recognition result, embeds Word message.With reference to Fig. 8, be the described structure drawing of device of embodiment.Said device mainly comprises:
Stroke Input Control Element U81 is used to obtain the stroke information of handwriting input under the stroke input pattern;
Literal Input Control Element U82 is used to obtain the Word message of input under the literal input pattern;
Shape recognition unit U83 is used for the stroke information that the stroke Input Control Element that obtains obtains is carried out the identification of shape stroke, obtains one or more shape stroke recognition results;
Data integration unit U84 is used for confirming the shape stroke recognition result under this literal information according to the input position of Word message.Promptly confirm the position relation of Word message and said shape stroke recognition result, and Word message is embedded shape stroke recognition result according to said position relation according to the input position of Word message.
Wherein, a kind of implementation structure of said literal Input Control Element U82 is that literal Input Control Element U82 comprises:
First input frame is created subelement, is used for dynamic creation literal input frame under the literal input pattern;
First literal receives subelement, is used to be received in the Word message of said literal input frame input, and said Word message is through calling various input method inputs, and said Word message comprises one or more characters.
Wherein, said first input frame is created the action message that subelement is pressed, drags, lifted according to mouse, dynamic creation literal input frame, and said literal input frame is through outer a bit the hiding arbitrarily of click input frame, but the Word message that reservation has been imported.
The another kind of implementation structure of said literal Input Control Element U82 is that literal Input Control Element U82 comprises:
Second input frame is created subelement, is used for dynamic creation stroke input frame under the literal input pattern;
Second literal receives subelement; Be used to be received in the strokes of characters information of said stroke input frame handwriting input; And calling word stroke recognition unit carries out strokes of characters identification to said strokes of characters information; Receive the strokes of characters recognition result as the Word message that obtains from the strokes of characters recognition unit, said Word message comprises one or more characters.
Wherein, said second input frame is created the action message that subelement is pressed, drags, lifted according to mouse, dynamic creation stroke input frame, and said stroke input frame is through outer a bit the hiding arbitrarily of click input frame, but the Word message that reservation has been imported.
Certainly, said literal Input Control Element U82 specifically can also have other implementation structures, enumerates no longer one by one at this.
Wherein, with all kinds of the be shaped as examples relevant with process flow diagram, a kind of implementation structure of said data integration unit U84 is that data integration unit U84 comprises:
Travel through subelement, be used for each Word message and the input position thereof of traversal record;
Judgment sub-unit is used for judging as follows according to the input position of each Word message:
Said shape stroke recognition result comprises close-shaped stroke and connecting line stroke;
If the input position of Word message is positioned at the inside of close-shaped stroke, then with the inside text attribute information of said Word message as said close-shaped stroke;
If the input position of Word message is positioned at the upper and lower, left and right side of connecting line stroke, then with the text attribute information of said Word message as said connecting line stroke next door;
If the input position of Word message does not meet above-mentioned two kinds of situation, then with said Word message as independent Word message.
When the user utilizes handwriting input device; When needing to import mixed informations such as comprising figure, form, literal; Said apparatus is not owing to need classify to shape stroke and strokes of characters; Thereby strokes of characters that having avoided classification to cause such as loses at situation, has improved the recognition accuracy and the input efficiency of handwriting input process Chinese words information.Simultaneously, the accuracy rate and the input efficiency of the identification of shape stroke have been improved indirectly.
Secondly, the Word message that said device receives can be the Word message that calls various input method inputs, also can be the recognition result of handwriting stroke.And the input equipment of Word message can be not limited to keyboard, handwriting pad etc.Therefore, in the handwriting input field, said device is more flexible for the input method that the user provides.
On the basis of device embodiment shown in Figure 8, another device embodiment provided by the invention can also comprise:
Record cell is used to write down the positional information of said Word message and said literal input frame/upper left angle point of stroke input frame, and this position will be used for Word message is embedded the integration processing of shape stroke recognition result as the input position of said Word message.
On the basis of device embodiment shown in Figure 8, another device embodiment provided by the invention can also comprise:
Output/display unit is used for said Word message and affiliated output of shape stroke recognition result or demonstration thereof.
The data handling procedure of the present invention one preferred embodiment embodiment is described through Fig. 9 below, and the device in the present embodiment comprises stroke Input Control Element U91, literal Input Control Element U92, shape recognition unit U93, data integration unit U94, record cell U95, output/display unit U96.The data handling procedure of this device is following:
Open the stroke input pattern through clicking modes such as control, under this pattern, stroke Input Control Element U91 gathers the stroke information of user in the drawing area input.Open the literal input pattern through clicking modes such as control, under this pattern, literal Input Control Element U92 clicks the optional position of pressing according to user's mouse and dynamically generates a literal input frame in drawing area, be used for receiving the character string of user's input; After the user confirms that the character string input of this place finishes, hide through clicking any one outer this article word input frame of naming a person for a particular job of literal input frame, but the character string in this literal input frame will continue demonstration; At this moment, record cell U95 can carry out record as one group of data with positional information (being the upper left angle point of literal input frame) and the character string of in this place's literal input frame, importing that user's mouse is clicked at the beginning in drawing area.If when the user need continue other position input stroke or the character string in drawing area, stroke Input Control Element U91 and literal Input Control Element U92 will repeat above-mentioned processing.
After the input of all stroke and character string finishes in the user confirms drawing area, trigger identification processing procedure through clicking modes such as control.Shape recognition unit U93 can discern processing according to the recognition methods of shape stroke to the stroke information of stroke Input Control Element U91 collection, obtains one or more shape stroke recognition results.Then; Data integration unit U94 can integrate all character string informations and each shape stroke recognition result that record cell U95 has write down one by one; Promptly confirm according to the positional information of each character string which shape stroke recognition result it belongs to, so that its Word message attribute as shape stroke recognition result is exported or demonstration through output/display unit U96.
The part that does not detail in Fig. 8, the device shown in Figure 9 can be considered for length referring to the relevant portion of Fig. 1, method shown in Figure 2, is not described in detail in this.
More than to a kind of method and device that in shape recognition result, embeds Word message provided by the present invention; Carried out detailed introduction; Used concrete example among this paper principle of the present invention and embodiment are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, part all can change on embodiment and range of application.In sum, this description should not be construed as limitation of the present invention.
Claims (14)
1. a method that in shape recognition result, embeds Word message is characterized in that, comprising:
Obtain the stroke information of handwriting input under the stroke input pattern, and obtain the Word message of input under the literal input pattern;
The stroke information of obtaining is carried out the identification of shape stroke, obtain one or more shape stroke recognition results;
The position of confirming Word message and said shape stroke recognition result according to the input position of Word message concerns, and according to said position relation Word message is embedded shape stroke recognition result;
Said input position according to Word message is confirmed the position relation of Word message and said shape stroke recognition result, and concerns that according to said position Word message is embedded shape stroke recognition result specifically to be comprised:
Said shape stroke recognition result comprises close-shaped stroke and/or connecting line stroke;
If the input position of Word message is positioned at the inside of close-shaped stroke, then the inside text attribute information of said Word message as said close-shaped stroke embedded;
If the input position of Word message is positioned near the upper and lower, left and right side of connecting line stroke, then the text attribute information of said Word message as said connecting line stroke next door embedded;
If the input position of Word message does not meet above-mentioned two kinds of situation, then with said Word message as independent Word message.
2. method according to claim 1 is characterized in that, the said Word message that obtains input under the literal input pattern specifically comprises:
Dynamic creation literal input frame under the literal input pattern;
Be received in the Word message of said literal input frame input, said Word message is through calling various input method inputs, and said Word message comprises one or more characters.
3. method according to claim 1 is characterized in that, the said Word message that obtains input under the literal input pattern specifically comprises:
Dynamic creation stroke input frame under the literal input pattern;
Be received in the strokes of characters information of said stroke input frame handwriting input;
Said strokes of characters information is carried out strokes of characters identification, obtain the strokes of characters recognition result as the Word message that obtains, said Word message comprises one or more characters.
4. according to claim 2 or 3 described methods, it is characterized in that said dynamic creation literal input frame/stroke input frame specifically comprises:
According to the action message that mouse is pressed, drags, lifted, dynamic creation literal input frame/stroke input frame.
5. according to claim 2 or 3 described methods, it is characterized in that:
Said literal input frame/stroke input frame is through outer a bit the hiding arbitrarily of click input frame, but the Word message that reservation has been imported.
6. according to claim 2 or 3 described methods, it is characterized in that said obtaining after the Word message of under the literal input pattern, importing also comprises:
Write down the positional information of said Word message and said literal input frame/upper left angle point of stroke input frame, this position will be as the input position of said Word message.
7. method according to claim 1 is characterized in that, said input position according to Word message is confirmed also to comprise after the shape stroke recognition result under this literal information:
With said Word message and affiliated output of shape stroke recognition result or demonstration thereof.
8. a device that in shape recognition result, embeds Word message is characterized in that, comprising:
The stroke Input Control Element is used to obtain the stroke information of handwriting input under the stroke input pattern;
The literal Input Control Element is used to obtain the Word message of input under the literal input pattern;
The shape recognition unit is used for the stroke information that the stroke Input Control Element that obtains obtains is carried out the identification of shape stroke, obtains one or more shape stroke recognition results;
The data integration unit is used for confirming that according to the input position of Word message the position of Word message and said shape stroke recognition result concerns, and according to said position relation Word message is embedded shape stroke recognition result;
Said data integration unit specifically comprises:
Travel through subelement, be used for each Word message and the input position thereof of traversal record;
Judgment sub-unit is used for judging as follows according to the input position of each Word message:
Said shape stroke recognition result comprises close-shaped stroke and/or connecting line stroke;
If the input position of Word message is positioned at the inside of close-shaped stroke, then the inside text attribute information of said Word message as said close-shaped stroke embedded;
If the input position of Word message is positioned near the upper and lower, left and right side of connecting line stroke, then the text attribute information of said Word message as said connecting line stroke next door embedded;
If the input position of Word message does not meet above-mentioned two kinds of situation, then with said Word message as independent Word message.
9. device according to claim 8 is characterized in that, said literal Input Control Element specifically comprises:
First input frame is created subelement, is used for dynamic creation literal input frame under the literal input pattern;
First literal receives subelement, is used to be received in the Word message of said literal input frame input, and said Word message is through calling various input method inputs, and said Word message comprises one or more characters.
10. device according to claim 8 is characterized in that, said literal Input Control Element specifically comprises:
Second input frame is created subelement, is used for dynamic creation stroke input frame under the literal input pattern;
Second literal receives subelement; Be used to be received in the strokes of characters information of said stroke input frame handwriting input; And calling word stroke recognition unit carries out strokes of characters identification to said strokes of characters information; Receive the strokes of characters recognition result as the Word message that obtains from the strokes of characters recognition unit, said Word message comprises one or more characters.
11., it is characterized in that according to claim 9 or 10 described devices:
Said first input frame is created the action message that subelement is pressed, drags, lifted according to mouse, dynamic creation literal input frame;
Said second input frame is created the action message that subelement is pressed, drags, lifted according to mouse, dynamic creation stroke input frame.
12., it is characterized in that according to claim 9 or 10 described devices:
Said literal input frame/stroke input frame is through outer a bit the hiding arbitrarily of click input frame, but the Word message that reservation has been imported.
13., it is characterized in that said device also comprises according to claim 9 or 10 described devices:
Record cell is used to write down the positional information of said Word message and said literal input frame/upper left angle point of stroke input frame, and this position will be as the input position of said Word message.
14. device according to claim 8 is characterized in that, said device also comprises:
Output/display unit is used for said Word message and affiliated output of shape stroke recognition result or demonstration thereof.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101197914A CN101673408B (en) | 2008-09-10 | 2008-09-10 | Method and device for embedding character information in shape recognition result |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101197914A CN101673408B (en) | 2008-09-10 | 2008-09-10 | Method and device for embedding character information in shape recognition result |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101673408A CN101673408A (en) | 2010-03-17 |
| CN101673408B true CN101673408B (en) | 2012-02-22 |
Family
ID=42020617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008101197914A Expired - Fee Related CN101673408B (en) | 2008-09-10 | 2008-09-10 | Method and device for embedding character information in shape recognition result |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101673408B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102455869B (en) * | 2011-09-29 | 2014-10-22 | 北京壹人壹本信息科技有限公司 | Method and device for editing characters by using gestures |
| JP2014127188A (en) * | 2012-12-27 | 2014-07-07 | Toshiba Corp | Shaping device and method |
| JP2014235452A (en) * | 2013-05-30 | 2014-12-15 | 株式会社東芝 | Shaping device |
| CN104714666B (en) * | 2013-12-12 | 2017-09-05 | 鸿合科技有限公司 | A kind of smart pen and its stroke recognition processing method |
| CN116185209A (en) * | 2014-08-11 | 2023-05-30 | 张锐 | Processing, data splitting and merging and coding and decoding processing method for handwriting input characters |
| CN105607804A (en) * | 2015-12-18 | 2016-05-25 | 深圳市金立通信设备有限公司 | Information display method and terminal |
| CN105930140A (en) * | 2016-04-01 | 2016-09-07 | 乐视控股（北京）有限公司 | Input mode setting method and apparatus |
| CN107273032A (en) * | 2017-06-28 | 2017-10-20 | 广州视源电子科技股份有限公司 | Information typesetting method, device and equipment and computer storage medium |
-
2008
- 2008-09-10 CN CN2008101197914A patent/CN101673408B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN101673408A (en) | 2010-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101673408B (en) | Method and device for embedding character information in shape recognition result | |
| US10191889B2 (en) | Systems, apparatuses and methods for generating a user interface by performing computer vision and optical character recognition on a graphical representation | |
| JP4637455B2 (en) | User interface utilization method and product including computer usable media | |
| JP6431119B2 (en) | System and method for input assist control by sliding operation in portable terminal equipment | |
| CN104298982B (en) | A kind of character recognition method and device | |
| CN101308610B (en) | Calligraphy exercising method and apparatus based on mobile terminal | |
| US8566752B2 (en) | Persistent selection marks | |
| CN114641753A (en) | Composite data generation and Building Information Model (BIM) element extraction from floor plan drawings using machine learning | |
| KR100359961B1 (en) | Handwriting information processing system with character segmentation user interface | |
| US9323456B2 (en) | Multi-character continuous handwriting input method | |
| CN101587489B (en) | Method and system for exhibiting specific page resource for browser | |
| US20020141643A1 (en) | Method for creating and operating control systems | |
| CN106484266A (en) | A kind of text handling method and device | |
| CN102903136B (en) | A kind of handwriting electronization method and system | |
| CN106164932B (en) | The method and apparatus of music symbol for identification | |
| CN105511792A (en) | In-position hand input method and system for form | |
| CN102156865A (en) | Handwritten text line character segmentation method and identification method | |
| CN102135838A (en) | Method and system for partitioned input of handwritten character string | |
| CN101685497B (en) | Method and device for processing hand-written information | |
| CN103226445A (en) | Handwriting input method, system and terminal | |
| CN102419645A (en) | Method and device for inputting Chinese characters in combined mode and palm terminal | |
| CN101581981A (en) | Method and system for directly forming Chinese text by writing Chinese characters on a piece of common paper | |
| CN103019397B (en) | A kind of line of text input method of multilingual mixing and device | |
| WO2013047980A1 (en) | Method for inputting figures into an electronic document in a direct input manner and computer readable recording medium recording a program for inputting figures into an electronic document in direct input manner | |
| CN107103312A (en) | A kind of image processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120222 Termination date: 20200910 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |