CN105630368A - Handwritten content dividing method and device as well as handwritten content editing device - Google Patents

Handwritten content dividing method and device as well as handwritten content editing device Download PDF

Info

Publication number
CN105630368A
CN105630368A CN201410608682.4A CN201410608682A CN105630368A CN 105630368 A CN105630368 A CN 105630368A CN 201410608682 A CN201410608682 A CN 201410608682A CN 105630368 A CN105630368 A CN 105630368A
Authority
CN
China
Prior art keywords
row
stroke
region
character
input area
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201410608682.4A
Other languages
Chinese (zh)
Inventor
张庆久
乐宁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sharp Corp
Original Assignee
Sharp Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sharp Corp filed Critical Sharp Corp
Priority to CN201410608682.4A priority Critical patent/CN105630368A/en
Priority to PCT/CN2015/093658 priority patent/WO2016070785A1/en
Priority to JP2017523206A priority patent/JP2017534121A/en
Publication of CN105630368A publication Critical patent/CN105630368A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/26Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • User Interface Of Digital Computer (AREA)
  • Character Discrimination (AREA)
  • Character Input (AREA)

Abstract

The invention provides a handwritten content dividing method and device. The handwritten content dividing method comprises the steps: sampling a new input stroke, and computing a new stroke region that covers the new input stroke; if the new stroke region intersects with a row input region, the new input stroke is determined to belong to the row corresponding to the row input region, and the row border region and the row input region of the row are re-computed; if the new stroke region doest not intersect with any row input region, the new input stroke is determined to belong to a new row, and the new stroke region is regarded as the row border region of the new row, and the row input region of the new row is computed according to a preset rule, the row input region covers the row border region. Handwritten content can be divided efficiently and accurately according to the handwritten content dividing method and device provided by embodiments of the invention.

Description

Handwritten content division methods and equipment and handwritten content editing equipment
Technical field
The present invention relates to handwritten content editing technique, more particularly, to a kind of handwritten content division methods and equipment and handwritten content editing equipment, it is possible to handwritten content is divided efficient and highly accurately.
Background technology
Along with the development of information technology, electronic equipment (such as, personal digital assistant, HPC, mobile phone) etc. use more and more universal in the life of people. The input mode of these electronic equipments substantially can be divided into input through keyboard and the big class of handwriting input two. Keyboard entry method adopts the keyboard on electronic equipment, and each button in user point beating keyboard inputs. Handwriting input mode receives the input of user by the touch screen of electronic equipment, processes to identify the content that user inputs to input. This needs electronic equipment to have higher identification ability, namely the accuracy rate of its identification is proposed higher requirement.
Chinese patent application 00103458.8 proposes a kind of hand-written character editing system. In the process of word processing, user's pen and editor's gesture carry out word and write and manuscript editing, editor's gesture and handwriting input mode automatically switch, and user need not use other interactive devices beyond pen, there is both of which, i.e. input pattern and the pattern of selection. This patent application also list multiple editting function, for instance duplication, movement, deletion, insertion etc.
Chinese patent application 201210546034.1 proposes a kind of method carrying out touch screen editor on handwriting image, wherein, the track record that user is slided by processor in handwriting area is in buffer, buffer will input a complete width trace image be converted to the image block of a predefined size the trace graphics that comprises trace image, and the sequencing that image block is formed according to image block will be shown on the display interface of touch screen with predefined procedure arrangement; By identifying the handwriting, graph data is edited, as deleted, replicate, shear and/or pasting, under the premise not interrupting user writing thought, and process of in real time graph data being modified.
Chinese patent application 201110161139.0 proposes a kind of handwriting input recognition methods, including: showing the character input area in hand-written district, this character belongs to region and at least includes newly inputted indicating area; Receive the character that user is hand-written; Judge that user writes whether character is positioned at newly inputted indicating area; If so, then confirming that user writes character is newly inputted character. The method is passed through writing the newly inputted character indicating area of Division, and carries out the location determination of such as character and region, can confirm that the input writing district's person's handwriting is intended to, it is achieved fresh character input or identification are revised.
Chinese patent application 200710304418.1 proposes a kind of handwriting processing method on the information processing terminal, including: the original handwriting data of user are gathered by hand input device; The original handwriting collected is carried out conversion process, changes the size of character script, form and display effect by conversion process, and the handwriting data after conversion process are stored in data storage device; Data presentation device will read on stored handwriting data in data storage device the display screen to the information processing terminal.
But, prior art is absent from contribute to user and inputs standard row when carrying out handwriting input and promote the mechanism of row division degree of accuracy. Existing row partition mechanism is only by using content projection to go division, and degree of accuracy is non-normally low. And, most electronic equipments cannot support that multirow inputs. It addition, whether existing electronic equipment intersects according only to adjacent stroke and carries out stroke and divide. It practice, this word division methods is non-normally low for the division degree of accuracy of hand-written character, because the character of handwriting input is usually nonstandard.
Accordingly, it would be desirable to a kind of mechanism that can handwritten content be divided efficient and highly accurately.
Summary of the invention
Present disclosure proposes a kind of handwritten content division methods and equipment and handwritten content editing equipment, it is possible to handwritten content is divided efficient and highly accurately.
According to an aspect of the present invention, it is proposed that a kind of handwritten content division methods, including: newly inputted stroke is sampled, the new stroke region of the stroke that calculating covering is newly inputted. If new stroke region is intersected with a row input area, then newly inputted stroke is divided into the row belonging to corresponding with described row input area, and recalculates row bound region and the row input area of described row. If new stroke region is all non-intersect with any row input area, the stroke then thinking newly inputted belongs to new a line, described new stroke region is regarded the row bound region of this new a line, and the row input area of this new a line is calculated according to pre-defined rule, wherein, described row input area cover row borderline region.
Preferably, described region is rectangle.
Preferably, described pre-defined rule is one below: row input area predetermined ratio bigger than the row bound region of correspondence; Big first predetermined ratio in row bound region corresponding to the aspect ratio of row input area, predetermined greatly the second ratio in row bound region that width ratio is corresponding; Row input area preliminary dimension bigger than the row bound region of correspondence; Big first preliminary dimension in row bound region corresponding to the aspect ratio of row input area, big second preliminary dimension in row bound region that width ratio is corresponding.
Preferably, described method also includes: show described row input area.
Preferably, described method also includes: calculate the height in each row bound region; For highly lower than the row bound region of predetermined threshold, carry out row union operation, be incorporated into other row.
Preferably, row union operation includes: computed altitude is lower than the distance in the row bound region of predetermined threshold with adjacent row bound region; If described distance is less than preset distance, then row corresponding lower than the row bound region of predetermined threshold for height is incorporated to the row that adjacent row bound region is corresponding, and recalculates row bound region and row input area.
Preferably, described method also includes: calculate the height in each row bound region; For highly more than the row bound region of predetermined threshold, carry out row cutting operation, corresponding row is divided into multiple sub-line.
Preferably, row cutting operation includes: for highly more than the row bound region of predetermined threshold, calculate the central point of the stroke covered, if the Y coordinate of the central point of a stroke is positioned at the Y scope of another stroke, then thinking that two strokes belong to identical sub-line, the Y scope of one of them stroke is the Y coordinate scope of this stroke; According to belonging to the stroke of identical sub-line start-stop scope in the Y direction, it is determined that sub-line cut-off rule, corresponding row being divided into each sub-line, and calculates the sub-line borderline region covering each sub-line, wherein Y-direction is the direction vertical with the direction of row.
Preferably, described method also includes: for each row, execution character division operation.
Preferably, described character division operation includes: calculate the central point of each stroke; If the X-coordinate of the central point of a stroke is positioned at the X scope of another stroke, then think that two strokes belong to identical character, otherwise it is assumed that two strokes belong to different characters, wherein, the X scope of one stroke is the X-coordinate scope of this stroke, and wherein X-coordinate is the coordinate on the direction parallel with the direction of row.
Preferably, described character division operation also includes: for the character divided, calculate the width of each character and average character pitch, if the ratio of the spacing of a character and adjacent character and average character pitch is less than a threshold value, and the width sum of itself and adjacent character is less than another threshold value, then described character and described adjacent character are merged into a character.
According to the second aspect of the invention, it is provided that a kind of handwritten content divides equipment, including: input module, it is configured to receive the handwriting input content of user; Processing module, is configured to, when newly inputted stroke, newly inputted stroke be sampled, and calculates the new stroke region covering newly inputted stroke; Memorizer, is configured to storage stroke sampling, represents the parameter of row bound region and row input area; Row judge module, it is configured to the row input area by new stroke region Yu storage compare, if new stroke region is intersected with a row input area, then newly inputted stroke is divided into the row belonging to corresponding with described row input area, and recalculate row bound region and the row input area of described row, the more row bound region of the described row of storage and row input area in new memory; If new stroke region is all non-intersect with any row input area, the stroke then thinking newly inputted belongs to new a line, described new stroke region is regarded the row bound region of this new a line, and the row input area of this new a line is calculated according to pre-defined rule, computed row bound region and row input area are stored in memory, wherein, described row input area cover row borderline region.
According to the third aspect of the invention we, it is provided that a kind of handwritten content editing equipment, including handover module, it is configured between edit pattern and input pattern to switch the state of handwritten content editing equipment; Above-mentioned handwritten content divides equipment, is configured to be under input pattern at handwritten content editing equipment and enables; And editor module, it is configured to be under edit pattern at described handwritten content editing equipment and enables, it is achieved to the editor of the handwriting input content of storage in memorizer, and the handwriting input content after editor is stored in memory.
Unlike the prior art, handwritten content division methods and equipment according to embodiments of the present invention improve, in many aspects, the performance that handwritten content divides, and at least include:
1. contribute to the content of user writing specification more. When user attempts to add new content to target line, it is shown that a row input area, inform the tram of user writing specification stroke. Therefore, the row of handwriting input is more straight, this degree of accuracy also contributing to promote row division.
2. improve row and divide degree of accuracy. After row is carried out initial division, go and merge and row segmentation, further increase row and divide degree of accuracy.
3. improve character and divide degree of accuracy. Use the gap between the central point of adjacent stroke and each candidate characters to optimize character division result, there is good character and divide degree of accuracy.
Accompanying drawing explanation
By the preferred embodiments of the present invention being described below in conjunction with accompanying drawing, the above-mentioned of the present invention and other objects, features and advantages will be made clearly, wherein:
Fig. 1 shows handwritten content according to embodiments of the present invention and divides the schematic block diagram of equipment.
Fig. 2 illustrates an example of row union operation according to embodiments of the present invention.
Fig. 3 illustrates an example of row cutting operation according to embodiments of the present invention.
Fig. 4 illustrates an example of character Partitioning optimization operation according to embodiments of the present invention.
Fig. 5 illustrates the flow chart of handwritten content division methods according to embodiments of the present invention.
Fig. 6 illustrates the block diagram of handwritten content editing equipment according to embodiments of the present invention.
Fig. 7 illustrates that handwritten content division methods according to embodiments of the present invention is adapted to the block diagram of system therein.
Detailed description of the invention
Referring to accompanying drawing, the example embodiment of the present invention is described in detail. In the following description, some specific embodiments only for descriptive purposes, and should not be construed and the present invention has any restriction, and are the example of the present invention. May result in the understanding of the present invention cause obscure time, conventional structure or structure will be omitted.
Fig. 1 shows handwritten content according to embodiments of the present invention and divides the schematic block diagram of equipment 100. This handwritten content divides equipment 100 and includes: input module 110, processing module 120, memorizer 130 and row judge module 140.
Input module 110 is configured to receive the handwriting input content of user. Such as, input module can be the touch screen of electronic equipment, and user utilizes writing pencil or finger directly to input on the touchscreen.
Processing module 120 is configured to, when newly inputted stroke, newly inputted stroke be sampled, and calculates the new stroke region covering newly inputted stroke.
Memorizer 130 is configured to storage stroke sampling, represents the parameter of row bound region and row input area.
Row judge module 140 is configured to compare the row input area in new stroke region Yu storage, if new stroke region is intersected with a row input area, then newly inputted stroke is divided into the row belonging to corresponding with described row input area, and recalculate row bound region and the row input area of described row, the more row bound region of the described row of storage and row input area in new memory; If new stroke region is all non-intersect with any row input area, the stroke then thinking newly inputted belongs to new a line, described new stroke region is regarded the row bound region of this new a line, and the row input area of this new a line is calculated according to pre-defined rule, computed row bound region and row input area are stored in memory, wherein, described row input area cover row borderline region.
According to an embodiment, region can be rectangle. It is of course also possible to use the region of other shapes.
For example it is assumed that the row bound rectangle of row is B. The row input rectangle A corresponding to this row is calculated according to pre-defined rule. A is more than B. Described pre-defined rule can be one below: the big predetermined ratio of row bound rectangle that row input squareness ratio is corresponding; Predetermined greatly the second ratio of row bound rectangle that big first predetermined ratio of row bound rectangle corresponding to aspect ratio of row input rectangle, width ratio are corresponding; The big preliminary dimension of row bound rectangle that row input squareness ratio is corresponding; Big first preliminary dimension of row bound rectangle corresponding to aspect ratio of row input rectangle, big second preliminary dimension of row bound rectangle that width ratio is corresponding. Such as, row input rectangle A is the rectangle than row bound rectangle B big 15%. Again such as, row input rectangle A is than row bound rectangle B rectangle of big 10% on big 15%, height on width. Again such as, row input rectangle A is than row bound rectangle B rectangle of all big 10 pixels on height and width. Again such as, row input rectangle A is than the rectangle of big 5 pixels on row bound rectangle B in height big 10 pixels, width. Of course, it is possible to use other rules to facilitate the calculating of row input rectangle and to improve the degree of accuracy of row division.
Such as, the upper left corner of rectangle B and the coordinate in the lower right corner respectively (x1, and (x4, y4), the upper left corner of A and the coordinate in the lower right corner respectively (x1-l y1), and (x4+r y1-t), y4+b), wherein (l > 0, t > 0, r > 0, b > 0).
The new stroke region of newly inputted stroke is C. If C and A intersects, then this stroke is divided into row corresponding to A. Meanwhile, rectangle A and B is recalculated. If C and A is non-intersect, then it is assumed that this row is first stroke of new a line. At this moment, by the C row bound rectangle B as new a line, and the row input rectangle A of new a line is calculated.
When inputting the first stroke, owing to not having A and B, this stroke is considered as the first stroke of the first row.
Handwritten content shown in Fig. 1 divides equipment 100 and may also include display 150, is display configured to the content of user's handwriting input. Such as, display 150 can also be the touch screen of electronic equipment, shows the content of handwriting input thereon.
According to an embodiment, display 150 is additionally configured to display line input area, informs the tram of user writing specification stroke. So, not only when user's handwriting input, row is divided, additionally aid the row of user writing specification more. Therefore, user can input more straight row, and this contributes to promoting degree of accuracy that row divides then.
Handwritten content shown in Fig. 1 divides equipment 100 and may also include row Partitioning optimization module 160, is configured to calculate the height in each row bound region, for highly lower than the row bound region of predetermined threshold, carries out row union operation, is incorporated into other row; For highly more than the row bound region of predetermined threshold, carry out row cutting operation, corresponding row is divided into multiple sub-line.
As it is shown in figure 1, row Partitioning optimization module 160 can include row merges module 162 and row segmentation module 164. row merges module 162 and is configured to the computed altitude distance lower than the row bound region of predetermined threshold with adjacent row bound region. at this, adjacent refer to adjacent in the Y direction. if described distance is less than preset distance, then row corresponding lower than the row bound region of predetermined threshold for height is incorporated to the row that adjacent row bound region is corresponding, and recalculates row bound region and row input area. row segmentation module 164 is configured to for highly more than the row bound region of predetermined threshold, calculate the central point of the stroke covered, if the Y coordinate of the central point of a stroke is positioned at the Y scope of another stroke, then think that two strokes belong to identical sub-line, the Y scope of one of them stroke is the Y coordinate scope of this stroke, according to belonging to the stroke of identical sub-line start-stop scope in the Y direction, determine sub-line dividing line, corresponding row is divided into each sub-line, and calculate the row bound region covering each sub-line, wherein Y-direction is the direction vertical with the direction of row, the i.e. short transverse of stroke.
Go while entering stroke and after dividing, would be likely to occur some division mistakes. Therefore, it can perform to merge optimize with dividing method or correct row division result. The row dividing mistake is generally of abnormal height. Such as, some strokes are divided into a line mistakenly due to other stroke away from affiliated character, and the height of this row is generally much less than the height of normal row. Again such as, some row are likely to comprise multiple row, the row comprised can be called sub-line. The height of the row comprising multiple sub-line is typically much deeper than the height of normal row. Therefore, row Partitioning optimization module 160, by calculating the height in row bound region, is found out abnormal row, these row is carried out union operation and cutting operation. Little row can be merged into adjacent row. According to an embodiment, row merges module 162 and calculates the distance of little row and adjacent row. If distance is less than a threshold value, then little row can be incorporated to adjacent row. According to another embodiment, it is possible to compare the row bound region of each row, little row is incorporated to the row intersected with its row bound region corresponding to the row bound region reaching predetermined ratio. Such as, little row is overlapping with adjacent row reaches 50%, then the row that this is little can be incorporated to adjacent row. Fig. 2 illustrates an example of row union operation according to embodiments of the present invention. For the row comprising multiple sub-line, row segmentation module 164 is configured to, with the split position that stroke height, line height, row projection properties etc. check whether to there is multiple sub-line and sub-line, and then sub-line is divided into independent row. Fig. 3 illustrates an example of row cutting operation according to embodiments of the present invention. Such as, height in the row bound region covering the first sub-line and the second sub-line is higher than the height in the row bound region only comprising a line, and diverse location (highly) place that the stroke in the first sub-line and the stroke in the second sub-line are in this row, and do not have the height of stroke to be similar to the height of corresponding row, therefore may determine that this row includes multiple sub-line. Calculate the central point of the stroke covered. The Y scope of the central point according to each stroke and other strokes, it is determined that there are two word row, i.e. the first sub-line " this is the first row " and the second sub-line " this is the second row ". According to the start-stop scope in the Y direction of the stroke in the start-stop scope in the Y direction of the stroke in the first sub-line and the second sub-line, determine sub-line dividing line, the i.e. upper border line of the following boundary line of the first sub-line as shown in Figure 3 and the second sub-line, is divided into two sub-line by corresponding row. From figure 3, it can be seen that be partitioned into two sub-line intersect. Then the sub-line borderline region covering each sub-line is calculated, as new row bound region. Certainly, row merges module and row segmentation module can take other existing or the mechanism that proposes in the future goes merging and row segmentation, to optimize row division result. Such as, interior all strokes as it is shown on figure 3, retrieval segmentation moves ahead, the stroke with maximum height is found out. This stroke is the radical of first " this ". Assuming that its height is 0.4 with the ratio of the height of whole row, less than specifying threshold value, and its lower boundary apart from the ratio of the height of distance and this row (row bound region) of the lower boundary of this row (row bound region) again smaller than threshold value, there is multiple sub-line in this explanation. The position of the lower boundary according to this stroke, is divided into two sub-line by this row.Then calculate two sub-line centrage in the Y direction, fall into the stroke of two sub-line for the Y coordinate of stroke, according to the distance of they two centrages of distance judges finally belong to which sub-line simultaneously. So can realize the segmentation of row.
Handwritten content shown in Fig. 1 divides equipment 100 and may also include character and divide module 170, is configured to for each row, execution character division operation, including: calculate the central point of each stroke; If the X-coordinate of the central point of a stroke is positioned at the X scope of another stroke, then it is assumed that two strokes belong to identical character, otherwise it is assumed that two strokes belong to different characters, wherein, the X scope of a stroke is the X-coordinate scope of this stroke. Assuming that stroke S1Central point be P (x1, y1), another stroke S2X scope [x3, x4]. WhenTime, by S1And S2It is divided into identical character. Otherwise, by S1And S2It is divided into different characters. According to an embodiment, character divides module 170 and is additionally configured to for the character divided, calculate the width of each character and average character pitch, if the ratio of the spacing of a character and adjacent character and average character pitch is less than a threshold value, and the width sum of itself and adjacent character is less than another threshold value, then described character and described adjacent character are merged into a character. At this, adjacent refer to adjacent in the X direction. Fig. 4 illustrates an example of character Partitioning optimization operation according to embodiments of the present invention. Such as, after original character divides, " wood ", " again " of " tree " word, " very little " are divided into three characters, but the X scope of these three character is significantly less than the X scope of other characters. Therefore, these three character is merged into a character. " gloomy " and " woods " word is obtained it is likewise possible to be optimized. Such as, average character pitch and average character duration are calculated. Assuming that average character pitch is 15. Find out the character being smaller than threshold value 5 with adjacent character. Such as the threshold value that is smaller than between two " wood " of " woods ", and the width of two " wood " composition character is 19, less than threshold value 25. Therefore the two character can be merged into a character " woods ". " gloomy " word can obtain with same optimization method merging. Certainly, the mechanism that character division module can also take other existing or future proposes according to embodiments of the present invention is to divide character.
Fig. 5 illustrates the flow chart of handwritten content division methods 500 according to embodiments of the present invention. First, in step S501, newly inputted stroke is sampled, calculate the new stroke region covering newly inputted stroke. It follows that in step S502, relatively new stroke region and row input area. If new stroke region is intersected with a row input area, then in step S503, newly inputted stroke is divided into the row belonging to corresponding with described row input area, and recalculates row bound region and the row input area of described row. If new stroke region is all non-intersect with any row input area, then in step S504, the stroke thinking newly inputted belongs to new a line, described new stroke region is regarded the row bound region of this new a line, and calculates the row input area of this new a line according to pre-defined rule.
According to an embodiment, new stroke region, row input area and row bound region can be all rectangle. This is because that goes is shaped as rectangle, regional is shaped to rectangle and can distinguish each row easily. Certainly, other shapes are also applicable.
According to an embodiment, described pre-defined rule can be one below: row input area predetermined ratio bigger than the row bound region of correspondence; Big first predetermined ratio in row bound region corresponding to the aspect ratio of row input area, predetermined greatly the second ratio in row bound region that width ratio is corresponding; Row input area preliminary dimension bigger than the row bound region of correspondence; Big first preliminary dimension in row bound region corresponding to the aspect ratio of row input area, big second preliminary dimension in row bound region that width ratio is corresponding.
According to an embodiment, display line input area on equipment. So, the row of user writing specification more is contributed to.
According to an embodiment, find the row bound region of Height Anomalies, carry out row union operation or row cutting operation. Such as, for highly lower than the row bound region of predetermined threshold, the distance in row bound region and adjacent row bound region is calculated. If described distance is less than preset distance, then row corresponding for this row bound region is incorporated to the row that adjacent row bound region is corresponding, forms a new row, and the calculating row bound region of again new row and row input area. Such as, for highly more than the row bound region of predetermined threshold, calculate the central point of the stroke covered, if the Y coordinate of the central point of a stroke is positioned at the Y scope of another stroke, then thinking that two strokes belong to identical sub-line, the Y scope of one of them stroke is the Y coordinate scope of this stroke. According to belonging to the stroke of identical sub-line start-stop scope in the Y direction, it is determined that sub-line cut-off rule, corresponding row being divided into each sub-line, and calculates the sub-line borderline region covering each sub-line, wherein Y-direction is the direction vertical with the direction of row.
In order to realize complete written contents identification, for each row, execution character division operation. According to an embodiment, character division operation includes: calculate the central point of each stroke; If the X-coordinate of the central point of a stroke is positioned at the X scope of another stroke, then think that two strokes belong to identical character, otherwise it is assumed that two strokes belong to different characters, wherein, the X scope of one stroke is the X-coordinate scope of this stroke, and wherein X-coordinate is the coordinate on the direction parallel with the direction of row.
According to an embodiment, it is possible to the character for initial division performs optimization. Such as, for the character divided, calculate the width of each character and average character pitch, if the ratio of the spacing of a character and adjacent character and average character pitch is less than a threshold value, and the width sum of itself and adjacent character is less than another threshold value, then described character and described adjacent character are merged into a character.
Fig. 6 illustrates the block diagram of handwritten content editing equipment 600 according to embodiments of the present invention. Equipment 600 includes handover module 610, be configured between edit pattern and input pattern switch handwritten content editing equipment 600 state; Handwritten content according to embodiments of the present invention divides equipment 620, is configured to be under input pattern at handwritten content editing equipment 600 and enables; Editor module 630, is configured to be under edit pattern at described handwritten content editing equipment and enables, it is achieved to the editor of the handwriting input content of storage in memorizer, and the handwriting input content after editor stored in memory.
According to an embodiment, handover module 610 can be a switching push button on writing pencil, and user carrys out switching equipment 600 between edit pattern and input pattern by pressing this button. In input mode, user is write on the screen of equipment 600 by writing pencil. Under edit pattern, user is edited by the content of display on the writing pencil screen to equipment 600. Such as, editor module can support the operations such as the duplication to content, movement, insertion, deletion.
Display shown in Fig. 6 can be the display that handwritten content divides equipment 620, or the display additionally arranged. Such as, handwritten content editing equipment 600 can be the projector equipment with handwriting pad. User at the same time as carrying out handwriting input on the handwriting pad of input equipment and display, can show the content of input and editor simultaneously on the display panel of projector equipment.
Fig. 7 illustrates that handwritten content division methods according to embodiments of the present invention is adapted to the block diagram of system 700 therein. This system can be the system with writing input board. User can input and content of edit easily on writing input board, by the processor in system and memorizer, the content of input or editor is processed and is stored. Such as, user can edit greeting card by this system. Again such as, PPT can be edited while presenting PPT on writing input board by user.
Handwritten content division methods and equipment according to embodiments of the present invention can apply to electronic whiteboard, flat board PC, have the desktop PC of touch screen, mobile phone, PDA etc. has in the electronic equipment of document and image processing function. User can carry out handwriting input and editor on this electronic equipment. Handwritten content can be divided by electronic equipment efficient and highly accurately.
Other of the embodiment of the present invention disclosed herein arrange the software program of the steps and operations of embodiment of the method including performing formerly summarizing. More specifically, computer program is following a kind of embodiment: have computer-readable medium, and on computer-readable medium, coding has computer program logic, when performing on the computing device, computer program logic provides relevant operation, thus providing technique scheme. When performing at least one processor of computing system, computer program logic makes processor perform the operation (method) described in the embodiment of the present invention. This set of the present invention is typically provided as the Downloadable software image in other media or special IC (ASIC) or one or more module arranging or encoding the firmware on the software on the computer-readable medium of such as light medium (such as CD-ROM), floppy disk or hard disk etc., code and/or other data structures or such as one or more ROM or RAM or PROM chip or microcode, shared data bank etc. Software or firmware or this configuration may be installed on computing equipment, so that the one or more processors in computing equipment perform the technology described in the embodiment of the present invention. The software process being operated in conjunction with the computing equipment in such as one group of data communications equipment or other entities can also provide the equipment according to the present invention. Equipment according to the present invention can also be distributed on the multiple software process in multiple data communications equipment or all software process run on one group of small, dedicated computer or single computer between all software process run.
It should be understood that strictly say, embodiments of the invention can be implemented as the software program on computer equipment, software and hardware or independent software and/or independent circuit.
It should be noted that, in the above description, only in an illustrative manner, it is shown that technical scheme, but be not intended that the invention be limited to above-mentioned steps and cellular construction. In a likely scenario, it is possible to as required step and cellular construction be adjusted and accept or reject. Therefore, element necessary to the overall invention thought of some step and the unit not enforcement present invention. Therefore, technical characteristic essential to the invention is limited solely by the minimum requirements of the overall invention thought being capable of the present invention, and not by the restriction of above instantiation.
So far already in connection with preferred embodiment, invention has been described. It should be understood that those skilled in the art are without departing from the spirit and scope of the present invention, it is possible to carry out other change, replacement and interpolation various. Therefore, the scope of the present invention is not limited to above-mentioned specific embodiment, and should be defined by the appended claims.

Claims (21)

1. a handwritten content division methods, including:
Newly inputted stroke is sampled, calculates the new stroke region covering newly inputted stroke;
If new stroke region is intersected with a row input area, then newly inputted stroke is divided into the row belonging to corresponding with described row input area, and recalculates row bound region and the row input area of described row;
If new stroke region is all non-intersect with any row input area, the stroke then thinking newly inputted belongs to new a line, described new stroke region is regarded the row bound region of this new a line, and the row input area of this new a line is calculated according to pre-defined rule, wherein, described row input area cover row borderline region.
2. division methods according to claim 1, wherein, described region is rectangle.
3. division methods according to claim 2, wherein, described pre-defined rule is one below: row input area predetermined ratio bigger than the row bound region of correspondence; Big first predetermined ratio in row bound region corresponding to the aspect ratio of row input area, predetermined greatly the second ratio in row bound region that width ratio is corresponding; Row input area preliminary dimension bigger than the row bound region of correspondence; Big first preliminary dimension in row bound region corresponding to the aspect ratio of row input area, big second preliminary dimension in row bound region that width ratio is corresponding.
4. division methods according to claim 3, also includes:
Show described row input area.
5. division methods according to claim 1, also includes:
Calculate the height in each row bound region;
For highly lower than the row bound region of predetermined threshold, carry out row union operation, be incorporated into other row.
6. division methods according to claim 5, wherein row union operation includes:
Computed altitude is lower than the distance in the row bound region of predetermined threshold with adjacent row bound region;
If described distance is less than preset distance, then row corresponding lower than the row bound region of predetermined threshold for height is incorporated to the row that adjacent row bound region is corresponding, and recalculates row bound region and row input area.
7. division methods according to claim 1, also includes:
Calculate the height in each row bound region;
For highly more than the row bound region of predetermined threshold, carry out row cutting operation, corresponding row is divided into multiple sub-line.
8. division methods according to claim 7, wherein row cutting operation includes:
For highly more than the row bound region of predetermined threshold, calculate the central point of the stroke covered,
If the Y coordinate of the central point of a stroke is positioned at the Y scope of another stroke, then it is assumed that two strokes belong to identical sub-line, and wherein, the Y scope of stroke is the Y coordinate scope of this stroke;
According to belonging to the stroke of identical sub-line start-stop scope in the Y direction, it is determined that sub-line cut-off rule, corresponding row being divided into each sub-line, and calculates the sub-line borderline region covering each sub-line, wherein, Y-direction is the direction vertical with the direction of row.
9. division methods according to claim 1, also includes:
For each row, execution character division operation.
10. division methods according to claim 9, wherein, described character division operation includes:
Calculate the central point of each stroke;
If the X-coordinate of the central point of a stroke is positioned at the X scope of another stroke, then it is assumed that two strokes belong to identical character, otherwise it is assumed that two strokes belong to different characters,
Wherein, the X scope of stroke is the X-coordinate scope of this stroke, and wherein, X-coordinate is the coordinate on the direction parallel with the direction of row.
11. division methods according to claim 10, also include:
For the character divided, calculate the width of each character and average character pitch, if the ratio of the spacing of a character and adjacent character and average character pitch is less than a threshold value, and the width sum of itself and adjacent character is less than another threshold value, then described character and described adjacent character are merged into a character.
12. handwritten content divides an equipment, including:
Input module, is configured to receive the handwriting input content of user;
Processing module, is configured to, when newly inputted stroke, newly inputted stroke be sampled, and calculates the new stroke region covering newly inputted stroke;
Memorizer, is configured to storage stroke sampling, represents the parameter of row bound region and row input area;
Row judge module, it is configured to the row input area by new stroke region Yu storage compare, if new stroke region is intersected with a row input area, then newly inputted stroke is divided into the row belonging to corresponding with described row input area, and recalculate row bound region and the row input area of described row, the more row bound region of the described row of storage and row input area in new memory; If new stroke region is all non-intersect with any row input area, the stroke then thinking newly inputted belongs to new a line, described new stroke region is regarded the row bound region of this new a line, and the row input area of this new a line is calculated according to pre-defined rule, computed row bound region and row input area are stored in memory, wherein, described row input area cover row borderline region.
13. handwritten content according to claim 12 divides equipment, wherein, described region is rectangle.
14. handwritten content according to claim 13 divides equipment, wherein, described pre-defined rule is one below: row input area predetermined ratio bigger than the row bound region of correspondence; Big first predetermined ratio in row bound region corresponding to the aspect ratio of row input area, predetermined greatly the second ratio in row bound region that width ratio is corresponding; Row input area preliminary dimension bigger than the row bound region of correspondence; Big first preliminary dimension in row bound region corresponding to the aspect ratio of row input area, big second preliminary dimension in row bound region that width ratio is corresponding.
15. handwritten content according to claim 14 divides equipment, also include:
Display, is display configured to described row input area.
16. handwritten content according to claim 12 divides equipment, also include:
Row Partitioning optimization module, is configured to calculate the height in each row bound region, for highly lower than the row bound region of predetermined threshold, carries out row union operation, is incorporated into other row; Divide equipment for height more than the handwritten content of predetermined threshold, carry out row cutting operation, corresponding row is divided into multiple sub-line.
17. handwritten content according to claim 16 divides equipment, wherein, row Partitioning optimization module includes row and merges module, is configured to:
Computed altitude is lower than the distance in the row bound region of predetermined threshold with adjacent row bound region;
If described distance is less than preset distance, then row corresponding lower than the row bound region of predetermined threshold for height is incorporated to the row that adjacent row bound region is corresponding, and recalculates row bound region and row input area.
18. handwritten content according to claim 16 divides equipment, wherein, row Partitioning optimization module includes row segmentation module, is configured to:
For highly more than the row bound region of predetermined threshold, calculate the central point of the stroke covered;
If the Y coordinate of the central point of a stroke is positioned at the Y scope of another stroke, then it is assumed that two strokes belong to identical sub-line, and wherein, the Y scope of stroke is the Y coordinate scope of this stroke;
According to belonging to the stroke of identical sub-line start-stop scope in the Y direction, it is determined that sub-line cut-off rule, corresponding row being divided into each sub-line, and calculates the row bound region covering each sub-line, wherein, Y-direction is the direction vertical with the direction of row.
19. handwritten content according to claim 12 divides equipment, also include:
Character divides module, is configured to for each row, execution character division operation, including:
Calculate the central point of each stroke;
If the X-coordinate of the central point of a stroke is positioned at the X scope of another stroke, then think that two strokes belong to identical character, otherwise it is assumed that two strokes belong to different characters, wherein, the X scope of stroke is the X-coordinate scope of this stroke, wherein, X-coordinate is the coordinate on the direction parallel with the direction of row.
20. handwritten content according to claim 19 divides equipment, character divides module and is also configured to
For the character divided, calculate the width of each character and average character pitch, if the ratio of the spacing of a character and adjacent character and average character pitch is less than a threshold value, and the width sum of itself and adjacent character is less than another threshold value, then described character and described adjacent character are merged into a character.
21. a handwritten content editing equipment, including:
Handover module, be configured between edit pattern and input pattern switch handwritten content editing equipment state;
Divide equipment according to the handwritten content one of claim 12-20 Suo Shu, be configured to be under input pattern at handwritten content editing equipment and enable; And
Editor module, is configured to be under edit pattern at described handwritten content editing equipment and enables, it is achieved to the editor of the handwriting input content of storage in memorizer, and the handwriting input content after editor stored in memory.
CN201410608682.4A 2014-11-03 2014-11-03 Handwritten content dividing method and device as well as handwritten content editing device Pending CN105630368A (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201410608682.4A CN105630368A (en) 2014-11-03 2014-11-03 Handwritten content dividing method and device as well as handwritten content editing device
PCT/CN2015/093658 WO2016070785A1 (en) 2014-11-03 2015-11-03 Handwritten content dividing method, device and handwritten content editing device
JP2017523206A JP2017534121A (en) 2014-11-03 2015-11-03 Handwritten content classification method and device, and handwritten content editing device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410608682.4A CN105630368A (en) 2014-11-03 2014-11-03 Handwritten content dividing method and device as well as handwritten content editing device

Publications (1)

Publication Number Publication Date
CN105630368A true CN105630368A (en) 2016-06-01

Family

ID=55908580

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410608682.4A Pending CN105630368A (en) 2014-11-03 2014-11-03 Handwritten content dividing method and device as well as handwritten content editing device

Country Status (3)

Country Link
JP (1) JP2017534121A (en)
CN (1) CN105630368A (en)
WO (1) WO2016070785A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110135426A (en) * 2018-02-09 2019-08-16 北京世纪好未来教育科技有限公司 Sample mask method and computer storage medium

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111722727B (en) * 2019-03-22 2024-01-12 北京搜狗科技发展有限公司 Model training method applied to handwriting input, handwriting input method and device

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5517578A (en) * 1993-05-20 1996-05-14 Aha! Software Corporation Method and apparatus for grouping and manipulating electronic representations of handwriting, printing and drawings
JPH10307675A (en) * 1997-05-01 1998-11-17 Hitachi Ltd Method and device for recognizing handwritten character
JP5123588B2 (en) * 2007-07-17 2013-01-23 キヤノン株式会社 Display control apparatus and display control method
JP5481988B2 (en) * 2009-07-21 2014-04-23 株式会社リコー Image processing apparatus, control method, and program
KR101144675B1 (en) * 2010-04-09 2012-05-24 주식회사 디오텍 Improved continuous hand writing input device and method thereof
CN102214043B (en) * 2011-06-15 2013-12-11 安徽科大讯飞信息科技股份有限公司 Handwriting input identifying method and system
CN103268166B (en) * 2013-05-17 2015-12-09 山东明海软件技术有限公司 The former handwriting information collection of handwriting input device and display packing

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110135426A (en) * 2018-02-09 2019-08-16 北京世纪好未来教育科技有限公司 Sample mask method and computer storage medium

Also Published As

Publication number Publication date
WO2016070785A1 (en) 2016-05-12
JP2017534121A (en) 2017-11-16

Similar Documents

Publication Publication Date Title
JP5248696B1 (en) Electronic device, handwritten document creation method, and handwritten document creation program
US8600164B2 (en) Method and tool for recognizing a hand-drawn table
US9715623B2 (en) Reduced document stroke storage
JP6430197B2 (en) Electronic apparatus and method
CN105511792A (en) In-position hand input method and system for form
US20140297276A1 (en) Editing apparatus, editing method, and computer program product
WO2014163981A2 (en) Detection and reconstruction of east asian layout features in a fixed format document
JP2016071819A (en) Electronic apparatus and method
US20140129931A1 (en) Electronic apparatus and handwritten document processing method
JP5694234B2 (en) Electronic device, handwritten document display method, and display program
US20180046351A1 (en) Controlling display object on display screen
US10359920B2 (en) Object management device, thinking assistance device, object management method, and computer-readable storage medium
JP5306528B1 (en) Electronic device and handwritten document processing method
KR102075433B1 (en) Handwriting input apparatus and control method thereof
US8494278B2 (en) Handwritten character recognition based on frequency variations in characters
JPWO2014147712A1 (en) Information processing apparatus, information processing method, and program
US20130335369A1 (en) Information processing device, information processing method
JP6054547B2 (en) Electronic device and method for processing handwritten document information
US9927971B2 (en) Electronic apparatus, method and storage medium for generating chart object
JP2018067298A (en) Handwritten content editing device and handwritten content editing method
CN105630368A (en) Handwritten content dividing method and device as well as handwritten content editing device
US10127704B2 (en) Information processing device, program, and recording medium
CN114663902B (en) Document image processing method, device, equipment and medium
WO2015107692A1 (en) Electronic device and method for handwriting
JP6201838B2 (en) Information processing apparatus and information processing program

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20160601