WO2016070785A1 - 手写内容划分方法和设备、以及手写内容编辑设备 - Google Patents

手写内容划分方法和设备、以及手写内容编辑设备 Download PDF

Info

Publication number
WO2016070785A1
WO2016070785A1 PCT/CN2015/093658 CN2015093658W WO2016070785A1 WO 2016070785 A1 WO2016070785 A1 WO 2016070785A1 CN 2015093658 W CN2015093658 W CN 2015093658W WO 2016070785 A1 WO2016070785 A1 WO 2016070785A1
Authority
WO
WIPO (PCT)
Prior art keywords
row
line
stroke
input
area
Prior art date
Application number
PCT/CN2015/093658
Other languages
English (en)
French (fr)
Inventor
张庆久
乐宁
Original Assignee
夏普株式会社
张庆久
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 夏普株式会社, 张庆久 filed Critical 夏普株式会社
Priority to JP2017523206A priority Critical patent/JP2017534121A/ja
Publication of WO2016070785A1 publication Critical patent/WO2016070785A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation 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

Definitions

  • the present invention relates to handwritten content editing technology, and more particularly to a handwritten content dividing method and apparatus and a handwritten content editing apparatus capable of dividing handwritten content efficiently and accurately.
  • the input methods of these electronic devices can be roughly divided into two categories: keyboard input and handwriting input.
  • the keyboard input mode uses a keyboard on the electronic device, and the user clicks each button in the keyboard to input.
  • the handwriting input method receives the user's input through the touch screen of the electronic device, and processes the input to recognize the content input by the user. This requires the electronic device to have a higher recognition ability, that is, a higher requirement for the accuracy of its recognition.
  • Chinese patent application 00103458.8 proposes a handwritten character editing system.
  • the user uses the pen and editing gestures for text writing and document editing, and the editing gesture and handwriting input mode are automatically switched.
  • the user does not need to use other interactive devices other than the pen, and there are two modes, namely, input mode and selection. mode.
  • the patent application also lists various editing functions such as copying, moving, deleting, inserting, and the like.
  • Chinese Patent Application No. 201210546034.1 proposes a method for performing touch screen editing on a handwriting image, wherein the processor records the trajectory of the user sliding in the handwriting area in the buffer memory, and inserts a completed input in the buffer memory. Converting the trajectory image into an image block of a predetermined size and containing the trajectory pattern of the trajectory image, and displaying the image blocks on the display interface of the touch screen in a predetermined order according to the order in which the image blocks are formed; by editing the handwriting graphic data, For example, deleting, copying, cutting, and/or pasting, the graphic data is modified in real time without interrupting the user's writing ideas.
  • Chinese patent application 201110161139.0 proposes a handwriting input recognition method, including: Displaying a character input area in the handwriting area, the character belonging to the area including at least a new input indication area; receiving a character handwritten by the user; determining whether the user write character is located in the new input indication area; if yes, confirming that the user writes the character as a new input character .
  • a handwriting input recognition method including: Displaying a character input area in the handwriting area, the character belonging to the area including at least a new input indication area; receiving a character handwritten by the user; determining whether the user write character is located in the new input indication area; if yes, confirming that the user writes the character as a new input character .
  • Chinese Patent Application No. 200710304418.1 proposes a handwriting processing method for an information processing terminal, comprising: collecting original handwritten handwriting data of a user through a handwriting input device; converting the collected original handwritten handwriting, and changing character handwriting through conversion processing; The size, shape, and display effect are stored, and the converted handwritten handwriting data is stored in the data storage device; the data display device reads the handwriting data stored in the data storage device and displays it on the screen on the information processing terminal.
  • the present disclosure proposes a handwritten content dividing method and apparatus and a handwritten content editing apparatus capable of dividing handwritten content efficiently and with high precision.
  • a handwritten content dividing method comprising: sampling a newly input stroke to calculate a new stroke area covering a newly input stroke. If the new stroke area intersects with one of the line input areas, the newly input stroke is divided into lines corresponding to the line input area, and the line boundary area and the line input area of the line are recalculated. If the new stroke area does not intersect any of the line input areas, the newly input stroke is considered to belong to a new line, the new stroke area is treated as the line boundary area of the new line, and the new line is calculated according to a predetermined rule.
  • a line input area wherein the line input area covers a line boundary area.
  • the area is rectangular.
  • the predetermined rule is one of the following: the row input region is larger than the corresponding row boundary region by a predetermined ratio; the height of the row input region is larger than the corresponding row boundary region by a first predetermined ratio, and the width is larger than the corresponding row boundary region.
  • the second ratio is predetermined; the row input region is larger than the corresponding row boundary region by a predetermined size; the row input region has a height greater than the corresponding row boundary region by a first predetermined size, and the width is greater than the corresponding row boundary region by a second predetermined size.
  • the method further comprises: displaying the line input area.
  • the method further comprises: calculating a height of each row boundary region; performing a row merging operation for the row boundary region whose height is lower than a predetermined threshold, and incorporating it into other rows.
  • the row merging operation comprises: calculating a distance between the row boundary region of the height lower than the predetermined threshold and the adjacent row boundary region; if the distance is less than the predetermined distance, the row corresponding to the row boundary region having the height lower than the predetermined threshold The rows corresponding to the adjacent row boundary regions are merged, and the row boundary region and the row input region are recalculated.
  • the method further comprises: calculating a height of each row boundary region; performing a row segmentation operation on the row boundary region whose height is greater than a predetermined threshold, and dividing the corresponding row into a plurality of sub-rows.
  • the line segmentation operation comprises: calculating a center point of the covered stroke for a line boundary area whose height is greater than a predetermined threshold, and if the Y coordinate of the center point of one stroke is within the Y range of the other stroke, the two strokes are considered Belong to the same sub-row, the Y range of one stroke is the Y coordinate range of the stroke; according to the starting and ending range of the strokes belonging to the same sub-line in the Y direction, the sub-line dividing line is determined, and the corresponding line is divided into sub-rows, And calculating a sub-line boundary area covering each sub-line, wherein the Y direction is a direction perpendicular to the direction of the line.
  • the method further comprises performing a character division operation for each row.
  • the character dividing operation comprises: calculating a center point of each stroke; if the X coordinate of the center point of one stroke is within the X range of another stroke, the two strokes are considered to belong to the same character, otherwise two Strokes belong to different characters, wherein the X range of a stroke is the X coordinate range of the stroke, wherein the X coordinate is the coordinate in the direction parallel to the direction of the line.
  • the character dividing operation further comprises: calculating, for the divided characters, a width and an average character spacing of each character, if a ratio of a character to an adjacent character and an average character spacing is less than a threshold, and If the sum of the widths of adjacent characters is less than another threshold, then The character is combined with the adjacent characters into one character.
  • a handwritten content dividing apparatus comprising: an input module configured to receive a user's handwritten input content; and a processing module configured to perform a newly input stroke when a new stroke is newly input Sampling, calculating a new stroke area covering the newly input stroke; a memory configured to store stroke sampling, parameters representing a line boundary area and a line input area; and a line determination module configured to input the new stroke area and the stored line input area Performing an alignment, if the new stroke area intersects with a line input area, dividing the newly input stroke into a line corresponding to the line input area, and recalculating the line boundary area and the line input area of the line, Updating a row boundary area and a line input area of the row stored in the memory; if the new stroke area does not intersect any of the line input areas, the newly input stroke is considered to belong to a new line, and the new stroke area is regarded as the new a row boundary area of a row, and calculate a row input
  • a handwritten content editing apparatus comprising: a switching module configured to switch a state of a handwritten content editing device between an editing mode and an input mode; the handwritten content dividing device configured to The handwriting content editing device is enabled in the input mode; and the editing module is configured to be enabled when the handwritten content editing device is in the editing mode, to implement editing of the handwritten input content stored in the memory, and input the edited handwriting input The content is stored in memory.
  • the handwritten content dividing method and device improve the performance of handwritten content partitioning in various aspects, including at least:
  • FIG. 1 is a schematic block diagram showing a handwritten content dividing apparatus according to an embodiment of the present invention.
  • FIG. 2 shows an example of a row merging operation according to an embodiment of the present invention.
  • FIG. 3 shows an example of a line splitting operation according to an embodiment of the present invention.
  • FIG. 4 shows an example of a character division optimization operation according to an embodiment of the present invention.
  • FIG. 5 shows a flow chart of a handwritten content dividing method according to an embodiment of the present invention.
  • FIG. 6 shows a block diagram of a handwritten content editing device in accordance with an embodiment of the present invention.
  • FIG. 7 shows a block diagram of a system in which a handwritten content partitioning method can be adapted in accordance with an embodiment of the present invention.
  • FIG. 1 is a schematic block diagram showing a handwritten content dividing apparatus 100 according to an embodiment of the present invention.
  • the handwritten content dividing apparatus 100 includes an input module 110, a processing module 120, a memory 130, and a line judging module 140.
  • the input module 110 is configured to receive handwritten input content of the user.
  • the input module may be a touch screen of an electronic device, and the user inputs directly on the touch screen using a stylus or a finger.
  • the processing module 120 is configured to sample the newly input strokes when new strokes are input, and to calculate a new stroke area that covers the newly input strokes.
  • the memory 130 is configured to store stroke samples, parameters representing row boundary regions and row input regions.
  • the line judging module 140 is configured to compare the new stroke area with the stored line input area, and if the new stroke area intersects with one line input area, divide the newly input stroke into belonging to the line input area. Rows, and recalculating the row boundary region and the row input region of the row, updating the row boundary region and the row input region of the row stored in the memory; if new strokes If the area does not intersect with any of the line input areas, the newly input stroke is considered to belong to a new line, the new stroke area is regarded as the line boundary area of the new line, and the row input area of the new line is calculated according to a predetermined rule.
  • the calculated row boundary region and row input region are stored in a memory, wherein the row input region covers the row boundary region.
  • the area may be rectangular. Of course, other shapes of areas can also be used.
  • the line input rectangle A corresponding to the line is calculated according to a predetermined rule.
  • A is greater than B.
  • the predetermined rule may be one of the following: the row input rectangle is larger than the corresponding row boundary rectangle by a predetermined ratio; the height of the row input rectangle is larger than the corresponding row boundary rectangle by a first predetermined ratio, and the width is larger than the corresponding row boundary rectangle.
  • the second ratio the row input rectangle is larger than the corresponding row boundary rectangle by a predetermined size; the height of the row input rectangle is larger than the corresponding row boundary rectangle by a first predetermined size, and the width is larger than the corresponding row boundary rectangle by a second predetermined size.
  • the line input rectangle A is a rectangle that is 15% larger than the line boundary rectangle B.
  • the row input rectangle A is a rectangle that is 15% larger in width and 10% larger in height than the row boundary rectangle B.
  • the line input rectangle A is a rectangle that is 10 pixels larger in height and width than the line boundary rectangle B.
  • the row input rectangle A is a rectangle having a height of 10 pixels larger than the row boundary rectangle B and a width of 5 pixels.
  • other rules can be used to facilitate the calculation of the input rectangle and improve the accuracy of the line division.
  • the coordinates of the upper left and lower right corners of the rectangle B are (x1, y1) and (x4, y4), respectively, and the coordinates of the upper left and lower right corners of A are (x1-1, y1-t) and (x4+, respectively).
  • r, y4+b where (1>0, t>0, r>0, b>0).
  • the new stroke area of the newly entered stroke is C. If C intersects A, the stroke is divided into rows corresponding to A. At the same time, the rectangles A and B are recalculated. If C does not intersect A, the line is considered to be the first stroke of the new line. At this time, C is treated as a new line of the line boundary rectangle B, and a new line of the line input rectangle A is calculated.
  • the handwritten content dividing apparatus 100 shown in FIG. 1 may further include a display 150 configured to display content input by a user's handwriting.
  • the display 150 can also be a touch screen of an electronic device on which the content of the handwritten input is displayed.
  • display 150 is also configured to display a line input area informing the user to write the correct position of the specification stroke. In this way, not only the line is divided when the user inputs the handwriting, but also helps the user to write a more standardized line. Therefore, the user can input more straight lines, which in turn helps to improve the accuracy of the line division.
  • the handwritten content dividing apparatus 100 shown in FIG. 1 may further include a line dividing optimization module 160 configured to calculate a height of each row boundary area, perform a line merge operation for a line boundary area whose height is lower than a predetermined threshold, and incorporate the same Other rows; for a row boundary region whose height is greater than a predetermined threshold, a row dividing operation is performed to divide the corresponding row into a plurality of sub-rows.
  • a line dividing optimization module 160 configured to calculate a height of each row boundary area, perform a line merge operation for a line boundary area whose height is lower than a predetermined threshold, and incorporate the same Other rows; for a row boundary region whose height is greater than a predetermined threshold, a row dividing operation is performed to divide the corresponding row into a plurality of sub-rows.
  • the row partitioning optimization module 160 can include a row merge module 162 and a row split module 164.
  • the row merge module 162 is configured to calculate a distance of a row boundary region having a height below a predetermined threshold from an adjacent row boundary region.
  • adjacent means adjacent in the Y direction. If the distance is less than the predetermined distance, the row corresponding to the row boundary region whose height is lower than the predetermined threshold is merged into the row corresponding to the adjacent row boundary region, and the row boundary region and the row input region are recalculated.
  • the line segmentation module 164 is configured to calculate a center point of the covered stroke for a line boundary area having a height greater than a predetermined threshold.
  • the two strokes are considered to belong to The same sub-row, wherein the Y range of one stroke is the Y coordinate range of the stroke, and the sub-line division line is determined according to the start and end range of the strokes belonging to the same sub-line in the Y direction, and the corresponding line is divided into sub-rows, and A row boundary region covering each subrow is calculated, wherein the Y direction is a direction perpendicular to the direction of the row, that is, the height direction of the stroke.
  • the lines that divide the error usually have an abnormal height.
  • some strokes are erroneously divided into one line due to other strokes away from the character in question, and the height of the line is typically much smaller than the height of the normal line.
  • some rows may contain multiple rows, and the included rows may be referred to as child rows.
  • the height of a row containing multiple subrows is typically much larger than the height of a normal row. Therefore, the line division optimization module 160 finds the abnormal lines by calculating the height of the line boundary area, and performs a merge operation and a split operation on the lines.
  • row merge module 162 calculates the distance of a small row from an adjacent row. If the distance is less than a threshold, small rows can be merged into adjacent rows.
  • the row boundary regions of the respective rows may be compared, and the small rows are merged into rows corresponding to the row boundary regions where the row boundary regions intersect with a predetermined ratio. For example, a small line with adjacent The rows overlap by up to 50%, and the small rows can be merged into adjacent rows.
  • FIG. 2 shows an example of a row merging operation according to an embodiment of the present invention.
  • the row segmentation module 164 is configured to verify whether there are multiple sub-rows and the segmentation locations of the sub-rows using stroke height, row height, row projection features, etc., and then split the sub-rows into separate rows.
  • FIG. 3 shows an example of a line splitting operation according to an embodiment of the present invention.
  • the height of the line boundary area covering the first sub-row and the second sub-row is higher than the height of the line boundary area including only one line, and the stroke in the first sub-line and the stroke in the second sub-line are on the line At different positions (heights) in the middle, and the height of no stroke is similar to the height of the corresponding line, it can be judged that the line includes a plurality of sub-rows.
  • the row merge module and the row split module may take other existing or future proposed mechanisms for row merge and row splitting to optimize row partitioning results. For example, as shown in FIG.
  • the handwritten content dividing apparatus 100 shown in FIG. 1 may further include a character dividing module 170 configured to perform a character dividing operation for each line, including: calculating a center point of each stroke; if the X coordinate of the center point of one stroke is located at another In the X range of a stroke, the two strokes are considered to belong to the same character, otherwise the two strokes are considered to belong to different characters, wherein the X range of one stroke is the X coordinate range of the stroke. It is assumed that the center point of the stroke S1 is P(x1, y1), and the X range of the other stroke S2 is [x3, x4]. When x1[x3, x4], S1 and S2 are divided into the same character.
  • character division The module 170 is further configured to calculate a width and an average character spacing of each character for the divided characters if the ratio of the spacing of one character to the adjacent character to the average character spacing is less than a threshold and the width of the adjacent character If the sum is less than another threshold, the characters are merged with the adjacent characters into one character.
  • adjacent means adjacent in the X direction.
  • FIG. 4 shows an example of a character division optimization operation according to an embodiment of the present invention. For example, after the initial character division, the "wood”, "again”, and "inch” of the "tree” word are divided into three characters, but the X range of these three characters is significantly smaller than the X range of other characters.
  • FIG. 5 shows a flow diagram of a handwritten content partitioning method 500 in accordance with an embodiment of the present invention.
  • step S501 the newly input stroke is sampled, and a new stroke area covering the newly input stroke is calculated.
  • step S502 the new stroke area and the line input area are compared. If the new stroke area intersects with one of the line input areas, then in step S503, the newly input stroke is divided into lines belonging to the line input area, and the line boundary area and the line input area of the line are recalculated.
  • step S504 the newly input stroke is considered to belong to a new line, the new stroke area is treated as the line boundary area of the new line, and the calculation is performed according to a predetermined rule.
  • the new line of the line input area is not intersect with any of the line input areas.
  • the new stroke area, the line input area, and the line boundary area may each be a rectangle. This is because the shape of the line is a rectangle, and the shape of each area is set to a rectangle to easily distinguish the lines. Of course, other shapes are also suitable.
  • the predetermined rule may be one of the following: the row input region is larger than the corresponding row boundary region by a predetermined ratio; and the row input region has a height greater than the corresponding row boundary region by a first predetermined ratio and a width ratio corresponding to the row The boundary region is larger than the predetermined second ratio; the row input region is larger than the corresponding row boundary region by a predetermined size; the height of the row input region is larger than the corresponding row boundary region by the first predetermined size, and the width is larger than the corresponding row boundary region by the second predetermined size .
  • the line input area is displayed on the device. This helps the user write More standardized lines.
  • a highly abnormal row boundary region is found, and a row merge operation or a row split operation is performed. For example, for a line boundary area whose height is lower than a predetermined threshold, the distance of the line boundary area from the adjacent line boundary area is calculated. If the distance is less than the predetermined distance, the row corresponding to the row boundary region is merged into the row corresponding to the adjacent row boundary region to form a new row, and the row boundary region and the row input region of the new row are renewed. For example, for a line boundary area whose height is greater than a predetermined threshold, the center point of the covered stroke is calculated.
  • the Y range of one of the strokes is the Y coordinate range of the stroke.
  • the sub-line dividing line is determined, the corresponding line is divided into sub-rows, and the sub-line boundary area covering each sub-line is calculated, wherein the Y direction is the direction of the line Vertical direction.
  • the character dividing operation comprises: calculating a center point of each stroke; if the X coordinate of the center point of one stroke is within the X range of the other stroke, then the two strokes are considered to belong to the same character, otherwise two Strokes belong to different characters, wherein the X range of a stroke is the X coordinate range of the stroke, wherein the X coordinate is the coordinate in the direction parallel to the direction of the line.
  • optimization may be performed for the initially divided characters. For example, for the divided characters, the width of each character and the average character spacing are calculated, if the ratio of the spacing of one character to the adjacent character to the average character spacing is less than a threshold, and the sum of the widths of the adjacent characters is less than the other A threshold is used to combine the character with the adjacent character into one character.
  • FIG. 6 shows a block diagram of a handwritten content editing device 600 in accordance with an embodiment of the present invention.
  • the device 600 includes a switching module 610 configured to switch the state of the handwritten content editing device 600 between an edit mode and an input mode; the handwritten content dividing device 620 according to an embodiment of the present invention is configured to be at the input of the handwritten content editing device 600 The mode is enabled; the editing module 630 is configured to enable the handwritten content editing device to be edited in the edit mode, implement editing of the handwritten input content stored in the memory, and store the edited handwritten input content in the memory.
  • the switching module 610 can be a toggle button on the stylus that the user switches between the edit mode and the input mode by pressing the button.
  • the input mode the user writes on the screen of the device 600 through the stylus.
  • the edit mode the user edits the content displayed on the screen of the device 600 through the stylus.
  • editing module It can support operations such as copying, moving, inserting, and deleting content.
  • the display shown in FIG. 6 may be a display of the handwritten content dividing device 620 or a separately provided display.
  • the handwritten content editing device 600 may be a projection device having a tablet. The user can perform handwriting input on a tablet that is both an input device and a display, while displaying input and editing content on the display panel of the projection device.
  • FIG. 7 illustrates a block diagram of a system 700 in which a handwritten content partitioning method may be adapted in accordance with an embodiment of the present invention.
  • the system can be a system with a handwriting tablet.
  • the user can conveniently enter and edit content on the handwriting input pad, and the input or edited content is processed and stored by the processor and memory in the system.
  • a user can edit a greeting card through the system.
  • the user can edit the PPT while presenting the PPT on the handwriting input board.
  • the handwritten content dividing method and apparatus can be applied to an electronic device having a document and an image processing function such as an electronic whiteboard, a tablet PC, a desktop PC with a touch screen, a mobile phone, a PDA, or the like. Users can perform handwriting input and editing on such electronic devices. Electronic devices can divide handwritten content efficiently and with high precision.
  • the computer program product is an embodiment having a computer readable medium encoded with computer program logic, the computer program logic providing related operations when provided on a computing device to provide The above technical solution.
  • the computer program logic When executed on at least one processor of a computing system, the computer program logic causes the processor to perform the operations (methods) described in the embodiments of the present invention.
  • Such an arrangement of the present invention is typically provided as software, code and/or other data structures, or such as one or more, that are arranged or encoded on a computer readable medium such as an optical medium (e.g., CD-ROM), floppy disk, or hard disk.
  • Software or firmware or such a configuration may be installed on the computing device such that one or more processors in the computing device perform the techniques described in this embodiment of the invention.
  • a software process that operates in conjunction with a computing device, such as a group of data communication devices or other entities, may also provide the device in accordance with the present invention.
  • the device according to the invention may also be distributed between multiple software processes on multiple data communication devices, or all software processes running on a small set of dedicated computers, or all software processes running on a single computer.
  • embodiments of the invention may be implemented as software programs, software and hardware on a computer device, or as separate software and/or separate circuits.

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

本发明涉及手写内容划分方法和设备。该方法包括：对新输入的笔画进行采样，计算覆盖新输入的笔画的新笔画区域，如果新笔画区域与一个行输入区域相交，则将新输入的笔画划分为属于与所述行输入区域相对应的行，并重新计算所述行的行边界区域和行输入区域；如果新笔画区域与任意行输入区域都不相交，则认为新输入的笔画属于新的一行，将所述新笔画区域当做该新的一行的行边界区域，并根据预定规则计算该新的一行的行输入区域，其中，所述行输入区域覆盖行边界区域。根据本发明实施例的手写内容划分方法和设备，能够高效且高精确地对手写内容进行划分。

Description

手写内容划分方法和设备、以及手写内容编辑设备 技术领域
本发明涉及手写内容编辑技术，更具体地，涉及一种手写内容划分方法和设备以及手写内容编辑设备，能够高效且高精确地对手写内容进行划分。
背景技术
随着信息技术的发展，电子设备(例如，个人数字助理、手持电脑、手机)等的使用在人们的生活中越来越普及。这些电子设备的输入方式大致可以划分为键盘输入和手写输入两大类。键盘输入方式采用电子设备上的键盘，使用者点击键盘中的各个按键进行输入。手写输入方式通过电子设备的触摸屏接收使用者的输入，对输入进行处理以识别用户输入的内容。这需要电子设备具有较高的识别能力，即对其识别的准确率提出了较高的要求。
中国专利申请00103458.8提出了一种手写字符编辑系统。在字处理的过程中，用户用笔和编辑手势进行文字书写和文稿编辑，编辑手势和手写输入模式自动切换，用户不需要使用笔以外的其他交互设备、存在两种模式，即输入模式和选择模式。该专利申请还列出了多种编辑功能，例如复制、移动、删除、插入等。
中国专利申请201210546034.1提出了一种在笔迹图像上进行触屏编辑的方法，其中，处理器将使用者在手写区域内滑动的轨迹记录于缓冲储存器中，将缓冲储存器中输入完毕的一幅轨迹图像转换为一预定大小并包含轨迹图像的轨迹图形的图像块，并将图像块按照图像块形成的先后顺序以预定顺序排列显示于触屏的显示界面上；通过对笔迹图形数据进行编辑，如删除、复制、剪切和/或粘贴，在不打断使用者写作思路的前提下，实时对图形数据进行修改处理。
中国专利申请201110161139.0提出了一种手写输入识别方法，包括： 显示手写区内的字符输入区域，该字符属于区域至少包括新输入指示区域；接收用户手写的字符；判断用户写入字符是否位于新输入指示区域；若是，则确认用户写入字符为新输入字符。该方法通过对书写区划分新输入字符指示区域，并进行诸如字符和区域的位置判定，即可确认书写区笔迹的输入意图，实现新字符输入或识别修正。
中国专利申请200710304418.1提出了一种用于信息处理终端上的笔迹处理方法，包括：通过手写输入装置采集用户的原始手写笔迹数据；将采集到的原始手写笔迹进行转换处理，通过转换处理改变字符笔迹的大小、形态以及显示效果，并将转换处理后的手写笔迹数据存储在数据存储装置中；数据显示装置将读取数据存储装置中所存储的笔迹数据并显示至信息处理终端上的屏幕上。
但是，现有技术中不存在有助于用户在进行手写输入时输入标准行且提升行划分精确度的机制。现有的行划分机制仅通过使用内容投影来进行行划分，精确度非常低。而且，多数电子设备无法支持多行输入。另外，现有电子设备仅根据相邻笔画是否相交来进行字划分。实际上，这种字划分方法对于手写字符的划分精确度非常低，因为手写输入的字符通常是不规范的。
因此，需要一种能够高效且高精确地对手写内容进行划分的机制。
发明内容
本公开提出了一种手写内容划分方法和设备以及手写内容编辑设备，能够高效且高精确地对手写内容进行划分。
根据本发明的一个方面，提出了一种手写内容划分方法，包括：对新输入的笔画进行采样，计算覆盖新输入的笔画的新笔画区域。如果新笔画区域与一个行输入区域相交，则将新输入的笔画划分为属于与所述行输入区域相对应的行，并重新计算所述行的行边界区域和行输入区域。如果新笔画区域与任意行输入区域都不相交，则认为新输入的笔画属于新的一行，将所述新笔画区域当做该新的一行的行边界区域，并根据预定规则计算该新的一行的行输入区域，其中，所述行输入区域覆盖行边界区域。
优选地，所述区域是矩形。
优选地，所述预定规则是以下之一：行输入区域比对应的行边界区域大预定比例；行输入区域的高度比对应的行边界区域大第一预定比例、宽度比对应的行边界区域大预定第二比例；行输入区域比对应的行边界区域大预定尺寸；行输入区域的高度比对应的行边界区域大第一预定尺寸、宽度比对应的行边界区域大第二预定尺寸。
优选地，所述方法还包括：显示所述行输入区域。
优选地，所述方法还包括：计算各个行边界区域的高度；针对高度低于预定阈值的行边界区域，进行行合并操作，将其并入其他行。
优选地，行合并操作包括：计算高度低于预定阈值的行边界区域与相邻的行边界区域的距离；如果所述距离小于预定距离，则将高度低于预定阈值的行边界区域对应的行并入相邻的行边界区域对应的行，并重新计算行边界区域和行输入区域。
优选地，所述方法还包括：计算各个行边界区域的高度；针对高度大于预定阈值的行边界区域，进行行分割操作，将所对应的行分割为多个子行。
优选地，行分割操作包括：针对高度大于预定阈值的行边界区域，计算所覆盖的笔画的中心点，如果一个笔画的中心点的Y坐标位于另一笔画的Y范围内，则认为两个笔画属于相同的子行，其中一个笔画的Y范围是该笔画的Y坐标范围；根据属于相同子行的笔画在Y方向上的起止范围，确定子行分割线，将对应行分割为各个子行，并计算覆盖各个子行的子行边界区域，其中Y方向是与行的方向垂直的方向。
优选地，所述方法还包括：针对各行，执行字符划分操作。
优选地，所述字符划分操作包括：计算每个笔画的中心点；如果一个笔画的中心点的X坐标位于另一笔画的X范围内，则认为两个笔画属于相同的字符，否则认为两个笔画属于不同的字符，其中，一个笔画的X范围是该笔画的X坐标范围，其中X坐标是与行的方向平行的方向上的坐标。
优选地，所述字符划分操作还包括：针对所划分的字符，计算每个字符的宽度和平均字符间距，如果一个字符与相邻字符的间距与平均字符间距的比例小于一阈值，且其与相邻字符的宽度之和小于另一阈值，则将所 述字符与所述相邻字符合并成一个字符。
根据本发明的第二方面，提供了一种手写内容划分设备，包括：输入模块，被配置为接收用户的手写输入内容；处理模块，被配置为在新输入笔画时，对新输入的笔画进行采样，计算覆盖新输入的笔画的新笔画区域；存储器，被配置为存储笔画采样、表示行边界区域和行输入区域的参数；行判断模块，被配置为将新笔画区域与存储的行输入区域进行比对，如果新笔画区域与一个行输入区域相交，则将新输入的笔画划分为属于与所述行输入区域相对应的行，并重新计算所述行的行边界区域和行输入区域，更新存储器中存储的所述行的行边界区域和行输入区域；如果新笔画区域与任意行输入区域都不相交，则认为新输入的笔画属于新的一行，将所述新笔画区域当做该新的一行的行边界区域，并根据预定规则计算该新的一行的行输入区域，将所计算的行边界区域和行输入区域存储在存储器中，其中，所述行输入区域覆盖行边界区域。
根据本发明的第三方面，提供了一种手写内容编辑设备，包括：切换模块，被配置为在编辑模式和输入模式之间切换手写内容编辑设备的状态；上述手写内容划分设备，被配置为在手写内容编辑设备处于输入模式下启用；以及编辑模块，被配置为在所述手写内容编辑设备处于编辑模式下启用，实现对存储器中存储的手写输入内容的编辑，并将编辑后的手写输入内容存储在存储器中。
与现有技术不同，根据本发明实施例的手写内容划分方法和设备在多个方面改善了手写内容划分的性能，至少包括：
1.有助于用户书写更加规范的内容。当用户试图向目标行添加新的内容时，示出一个行输入区域，告知用户书写规范笔画的正确位置。因此，手写输入的行更加笔直，这也有助于提升行划分的精确度。
2.提升了行划分精确度。在对行进行初始划分之后，进行行合并和行分割，进一步提高了行划分精确度。
3.提升了字符划分精确度。使用相邻笔画的中心点和每个候选字符之间的间隙来优化字符划分结果，具有良好的字符划分精确度。
附图说明
通过下面结合附图说明本发明的优选实施例，将使本发明的上述及其它目的、特征和优点更加清楚，其中：
图1是示出了根据本发明实施例的手写内容划分设备的示意框图。
图2示出了根据本发明实施例的行合并操作的一个示例。
图3示出了根据本发明实施例的行分割操作的一个示例。
图4示出了根据本发明实施例的字符划分优化操作的一个示例。
图5示出了根据本发明实施例的手写内容划分方法的流程图。
图6示出了根据本发明实施例的手写内容编辑设备的框图。
图7示出了根据本发明实施例的手写内容划分方法可以适应于其中的系统的框图。
具体实施方式
以下参照附图，对本发明的示例实施例进行详细描述。在以下描述中，一些具体实施例仅用于描述目的，而不应该理解为对本发明有任何限制，而只是本发明的示例。在可能导致对本发明的理解造成混淆时，将省略常规结构或构造。
图1是示出了根据本发明实施例的手写内容划分设备100的示意框图。该手写内容划分设备100包括：输入模块110、处理模块120、存储器130和行判断模块140。
输入模块110被配置为接收用户的手写输入内容。例如，输入模块可以是电子设备的触摸屏，用户利用手写笔或者手指直接在触摸屏上进行输入。
处理模块120被配置为在新输入笔画时，对新输入的笔画进行采样，计算覆盖新输入的笔画的新笔画区域。
存储器130被配置为存储笔画采样、表示行边界区域和行输入区域的参数。
行判断模块140被配置为将新笔画区域与存储的行输入区域进行比对，如果新笔画区域与一个行输入区域相交，则将新输入的笔画划分为属于与所述行输入区域相对应的行，并重新计算所述行的行边界区域和行输入区域，更新存储器中存储的所述行的行边界区域和行输入区域；如果新笔画 区域与任意行输入区域都不相交，则认为新输入的笔画属于新的一行，将所述新笔画区域当做该新的一行的行边界区域，并根据预定规则计算该新的一行的行输入区域，将所计算的行边界区域和行输入区域存储在存储器中，其中，所述行输入区域覆盖行边界区域。
根据一个实施例，区域可以是矩形。当然，也可以使用其他形状的区域。
例如，假定行的行边界矩形是B。根据预定规则计算对应于该行的行输入矩形A。A大于B。所述预定规则可以是以下之一：行输入矩形比对应的行边界矩形大预定比例；行输入矩形的高度比对应的行边界矩形大第一预定比例、宽度比对应的行边界矩形大预定第二比例；行输入矩形比对应的行边界矩形大预定尺寸；行输入矩形的高度比对应的行边界矩形大第一预定尺寸、宽度比对应的行边界矩形大第二预定尺寸。例如，行输入矩形A是比行边界矩形B大15％的矩形。再例如，行输入矩形A是比行边界矩形B在宽度上大15％、高度上大10％的矩形。再例如，行输入矩形A是比行边界矩形B在高度和宽度上均大10个像素的矩形。再例如，行输入矩形A是比行边界矩形B在高度上大10个像素、宽度上大5个像素的矩形。当然，可以使用其他规则来方便行输入矩形的计算并提高行划分的精确度。
例如，矩形B的左上角和右下角的坐标分别为(x1，y1)和(x4，y4)，A的左上角和右下角的坐标分别为(x1-1，y1-t)和(x4+r，y4+b)，其中(1＞0，t＞0，r＞0，b＞0)。
新输入的笔画的新笔画区域是C。如果C与A相交，则将该笔画划分为A对应的行。同时，重新计算矩形A和B。如果C与A不相交，则认为该行是新的一行的第一笔画。这时，将C当做新的一行的行边界矩形B，并计算新的一行的行输入矩形A。
在输入第一笔画时，由于没有A和B，该笔画被认为是第一行的第一笔画。
图1所示的手写内容划分设备100还可包括显示器150，被配置为显示用户手写输入的内容。例如，显示器150也可以是电子设备的触摸屏，在其上显示手写输入的内容。
根据一个实施例，显示器150还被配置为显示行输入区域，告知用户书写规范笔画的正确位置。这样，不仅在用户手写输入时对行进行划分，还有助于用户书写更加规范的行。因此，用户可以输入更加笔直的行，而这继而又有助于提升行划分的精确度。
图1所示的手写内容划分设备100还可包括行划分优化模块160，被配置为计算各个行边界区域的高度，针对高度低于预定阈值的行边界区域，进行行合并操作，将其并入其他行；针对高度大于预定阈值的行边界区域，进行行分割操作，将所对应的行分割为多个子行。
如图1所示，行划分优化模块160可以包括行合并模块162和行分割模块164。行合并模块162被配置为计算高度低于预定阈值的行边界区域与相邻的行边界区域的距离。在此，相邻是指在Y方向上相邻。如果所述距离小于预定距离，则将高度低于预定阈值的行边界区域对应的行并入相邻的行边界区域对应的行，并重新计算行边界区域和行输入区域。行分割模块164被配置为针对高度大于预定阈值的行边界区域，计算所覆盖的笔画的中心点，如果一个笔画的中心点的Y坐标位于另一笔画的Y范围内，则认为两个笔画属于相同的子行，其中一个笔画的Y范围是该笔画的Y坐标范围，根据属于相同子行的笔画在Y方向上的起止范围，确定子行划分线，将对应行分割为各个子行，并计算覆盖各个子行的行边界区域，其中Y方向是与行的方向垂直的方向，即笔画的高度方向。
在输入笔画的同时进行行划分之后可能存在一些划分错误。因此，可以执行合并和分割方法来优化或纠正行划分结果。划分错误的行通常具有异常的高度。例如，一些笔画由于远离所属字符的其它笔画而被错误地划分为一行，该行的高度通常远小于正常的行的高度。再例如，一些行可能包含多个行，可将所包含的行称为子行。包含多个子行的行的高度通常远大于正常的行的高度。因此，行划分优化模块160通过计算行边界区域的高度，找出异常的行，对这些行进行合并操作和分割操作。可以将小的行合并到相邻的行。根据一个实施例，行合并模块162计算小的行与相邻的行的距离。如果距离小于一个阈值，则可以将小的行并入相邻的行。根据另一个实施例，可以比较各个行的行边界区域，将小的行并入与其行边界区域相交达预定比例的行边界区域所对应的行。例如，一个小的行与相邻 的行重叠达50％，则可以将该小的行并入相邻的行。图2示出了根据本发明实施例的行合并操作的一个示例。针对包含多个子行的行，行分割模块164被配置为利用笔画高度、行高度、行投影特征等来检验是否存在多个子行以及子行的分割位置，然后将子行分割为单独的行。图3示出了根据本发明实施例的行分割操作的一个示例。例如，在覆盖第一子行和第二子行的行边界区域的高度高于仅包含一行的行边界区域的高度，而且第一子行中的笔画与第二子行中的笔画在该行中的不同位置(高度)处，且没有笔画的高度与对应行的高度近似，因此可以判断该行包括多个子行。计算所覆盖的笔画的中心点。根据各个笔画的中心点与其他笔画的Y范围，确定存在两个字行，即第一子行“这是第一行”和第二子行“这是第二行”。根据第一子行中的笔画在Y方向上的起止范围以及第二子行中的笔画在Y方向上的起止范围，确定子行划分线，即如图3所示的第一子行的下边界线和第二子行的上边界线，将对应的行分割为两个子行。从图3中可以看出，分割出的两个子行相交。然后计算覆盖各个子行的子行边界区域，作为新的行边界区域。当然，行合并模块和行分割模块可以采取其他现有的或者将来提出的机制来进行行合并和行分割，以优化行划分结果。例如，如图3所示，检索分割前行内所有笔画，把具有最大高度的笔画找出来。此笔画为第一个“这”的偏旁。假定其高度与整个行的高度的比例为0.4，小于指定阈值，而且其下边界距离该行(行边界区域)的下边界的距离与该行(行边界区域)的高度的比例也小于阈值，这说明存在多个子行。根据该笔画的下边界的位置，将该行分为两个子行。然后计算两个子行在Y方向上的中心线，针对笔画的Y坐标同时落入两个子行的笔画，根据它们距离两个中心线的距离来判断最终属于哪一子行。这样可以实现行的分割。
图1所示的手写内容划分设备100还可包括字符划分模块170，被配置为针对各行，执行字符划分操作，包括：计算每个笔画的中心点；如果一个笔画的中心点的X坐标位于另一笔画的X范围内，则认为两个笔画属于相同的字符，否则认为两个笔画属于不同的字符，其中，一个笔画的X范围是该笔画的X坐标范围。假定笔画S1的中心点是P(x1，y1)，另一笔画S2的X范围[x3，x4]。当x1[x3，x4]时，将S1和S2划分到相同的字符。否则，将S1和S2划分到不同的字符。根据一个实施例，字符划分 模块170还被配置为针对所划分的字符，计算每个字符的宽度和平均字符间距，如果一个字符与相邻字符的间距与平均字符间距的比例小于一阈值，且其与相邻字符的宽度之和小于另一阈值，则将所述字符与所述相邻字符合并成一个字符。在此，相邻是指在X方向上相邻。图4示出了根据本发明实施例的字符划分优化操作的一个示例。例如，在初始字符划分之后，“树”字的“木”、“又”、“寸”被划分为三个字符，但是这三个字符的X范围明显小于其他字符的X范围。因此，将这三个字符合并成一个字符。类似地，可以进行优化得到“森”和“林”字。例如，计算平均字符间距和平均字符宽度。假定平均字符间隔为15。找出与相邻字符的间距小于阈值5的字符。如“林”的两个“木”之间的间距小于阈值，而且两个“木”组成字符的宽度为19，小于阈值25。因此可将这两个字符合并成一个字符“林”。“森”字可用同样的优化方法合并得到。当然，根据本发明实施例的字符划分模块也可以采取其他现有的或将来提出的机制来划分字符。
图5示出了根据本发明实施例的手写内容划分方法500的流程图。首先，在步骤S501，对新输入的笔画进行采样，计算覆盖新输入的笔画的新笔画区域。接下来，在步骤S502，比较新笔画区域和行输入区域。如果新笔画区域与一个行输入区域相交，则在步骤S503，将新输入的笔画划分为属于与所述行输入区域相对应的行，并重新计算所述行的行边界区域和行输入区域。如果新笔画区域与任意行输入区域都不相交，则在步骤S504，认为新输入的笔画属于新的一行，将所述新笔画区域当做该新的一行的行边界区域，并根据预定规则计算该新的一行的行输入区域。
根据一个实施例，新笔画区域、行输入区域和行边界区域均可以为矩形。这是因为行的形状为矩形，将各个区域的形状设置为矩形可以方便地区分各个行。当然，其他形状也是适用的。
根据一个实施例，所述预定规则可以是以下之一：行输入区域比对应的行边界区域大预定比例；行输入区域的高度比对应的行边界区域大第一预定比例、宽度比对应的行边界区域大预定第二比例；行输入区域比对应的行边界区域大预定尺寸；行输入区域的高度比对应的行边界区域大第一预定尺寸、宽度比对应的行边界区域大第二预定尺寸。
根据一个实施例，在设备上显示行输入区域。这样，有助于用户书写 更加规范的行。
根据一个实施例，找到高度异常的行边界区域，进行行合并操作或行分割操作。例如，针对高度低于预定阈值的行边界区域，计算行边界区域与相邻的行边界区域的距离。如果所述距离小于预定距离，则将该行边界区域对应的行并入相邻的行边界区域对应的行，形成一个新的行，并重新新的行的计算行边界区域和行输入区域。例如，针对高度大于预定阈值的行边界区域，计算所覆盖的笔画的中心点，如果一个笔画的中心点的Y坐标位于另一笔画的Y范围内，则认为两个笔画属于相同的子行，其中一个笔画的Y范围是该笔画的Y坐标范围。根据属于相同子行的笔画在Y方向上的起止范围，确定子行分割线，将对应行分割为各个子行，并计算覆盖各个子行的子行边界区域，其中Y方向是与行的方向垂直的方向。
为了实现完整的书写内容识别，针对各行，执行字符划分操作。根据一个实施例，字符划分操作包括：计算每个笔画的中心点；如果一个笔画的中心点的X坐标位于另一笔画的X范围内，则认为两个笔画属于相同的字符，否则认为两个笔画属于不同的字符，其中，一个笔画的X范围是该笔画的X坐标范围，其中X坐标是与行的方向平行的方向上的坐标。
根据一个实施例，可以针对初始划分的字符执行优化。例如，针对所划分的字符，计算每个字符的宽度和平均字符间距，如果一个字符与相邻字符的间距与平均字符间距的比例小于一阈值，且其与相邻字符的宽度之和小于另一阈值，则将所述字符与所述相邻字符合并成一个字符。
图6示出了根据本发明实施例的手写内容编辑设备600的框图。设备600包括切换模块610，被配置为在编辑模式和输入模式之间切换手写内容编辑设备600的状态；根据本发明实施例的手写内容划分设备620，被配置为在手写内容编辑设备600处于输入模式下启用；编辑模块630，被配置为在所述手写内容编辑设备处于编辑模式下启用，实现对存储器中存储的手写输入内容的编辑，并将编辑后的手写输入内容存储在存储器中。
根据一个实施例，切换模块610可以是手写笔上的一个切换按钮，用户通过按下该按钮来在编辑模式和输入模式之间切换设备600。在输入模式下，用户通过手写笔在设备600的屏幕上进行书写。在编辑模式下，用户通过手写笔对设备600的屏幕上显示的内容进行编辑。例如，编辑模块 可以支持对内容的复制、移动、插入、删除等操作。
图6所示的显示器可以是手写内容划分设备620的显示器，或者是另外设置的显示器。例如，手写内容编辑设备600可以是具有手写板的投影设备。用户可以在同时作为输入设备和显示器的手写板上进行手写输入，同时在投影设备的显示板上显示输入和编辑的内容。
图7示出了根据本发明实施例的手写内容划分方法可以适应于其中的系统700的框图。该系统可以是具有手写输入板的系统。用户可以方便地在手写输入板上输入和编辑内容，由系统中的处理器和存储器对输入或编辑的内容进行处理和存储。例如，用户可以通过该系统来编辑贺卡。再例如，用户可以在手写输入板上呈现PPT的同时对PPT进行编辑。
根据本发明实施例的手写内容划分方法和设备可以应用于电子白板、平板PC、具有触摸屏的桌面PC、移动电话、PDA等具有文档和图像处理功能的电子设备中。用户可以在这种电子设备上进行手写输入和编辑。电子设备可以高效且高精确地对手写内容进行划分。
这里所公开的本发明实施例的其他设置包括执行在先概述的方法实施例的步骤和操作的软件程序。更具体地，计算机程序产品是如下的一种实施例：具有计算机可读介质，计算机可读介质上编码有计算机程序逻辑，当在计算设备上执行时，计算机程序逻辑提供相关的操作，从而提供上述技术方案。当在计算系统的至少一个处理器上执行时，计算机程序逻辑使得处理器执行本发明实施例所述的操作(方法)。本发明的这种设置典型地提供为设置或编码在例如光介质(例如CD-ROM)、软盘或硬盘等的计算机可读介质上的软件、代码和/或其他数据结构、或者诸如一个或多个ROM或RAM或PROM芯片上的固件或微代码的其他介质、或专用集成电路(ASIC)、或一个或多个模块中的可下载的软件图像、共享数据库等。软件或固件或这种配置可安装在计算设备上，以使得计算设备中的一个或多个处理器执行本发明实施例所述的技术。结合诸如一组数据通信设备或其他实体中的计算设备进行操作的软件过程也可以提供根据本发明的设备。根据本发明的设备也可以分布在多个数据通信设备上的多个软件过程、或者在一组小型专用计算机上运行的所有软件过程、或者单个计算机上运行的所有软件过程之间。
应该理解，严格地讲，本发明的实施例可以实现为计算机设备上的软件程序、软件和硬件、或者单独的软件和/或单独的电路。
应当注意的是，在以上的描述中，仅以示例的方式，示出了本发明的技术方案，但并不意味着本发明局限于上述步骤和单元结构。在可能的情形下，可以根据需要对步骤和单元结构进行调整和取舍。因此，某些步骤和单元并非实施本发明的总体发明思想所必需的元素。因此，本发明所必需的技术特征仅受限于能够实现本发明的总体发明思想的最低要求，而不受以上具体实例的限制。
至此已经结合优选实施例对本发明进行了描述。应该理解，本领域技术人员在不脱离本发明的精神和范围的情况下，可以进行各种其它的改变、替换和添加。因此，本发明的范围不局限于上述特定实施例，而应由所附权利要求所限定。

Claims (21)

  1. 一种手写内容划分方法，包括：
    对新输入的笔画进行采样，计算覆盖新输入的笔画的新笔画区域；
    如果新笔画区域与一个行输入区域相交，则将新输入的笔画划分为属于与所述行输入区域相对应的行，并重新计算所述行的行边界区域和行输入区域；
    如果新笔画区域与任意行输入区域都不相交，则认为新输入的笔画属于新的一行，将所述新笔画区域当做该新的一行的行边界区域，并根据预定规则计算该新的一行的行输入区域，其中，所述行输入区域覆盖行边界区域。
  2. 根据权利要求1的划分方法，其中，所述区域是矩形。
  3. 根据权利要求2的划分方法，其中，所述预定规则是以下之一：行输入区域比对应的行边界区域大预定比例；行输入区域的高度比对应的行边界区域大第一预定比例、宽度比对应的行边界区域大预定第二比例；行输入区域比对应的行边界区域大预定尺寸；行输入区域的高度比对应的行边界区域大第一预定尺寸、宽度比对应的行边界区域大第二预定尺寸。
  4. 根据权利要求3所述的划分方法，还包括：
    显示所述行输入区域。
  5. 根据权利要求1所述的划分方法，还包括：
    计算各个行边界区域的高度；
    针对高度低于预定阈值的行边界区域，进行行合并操作，将其并入其他行。
  6. 根据权利要求5所述的划分方法，其中行合并操作包括：
    计算高度低于预定阈值的行边界区域与相邻的行边界区域的距离；
    如果所述距离小于预定距离，则将高度低于预定阈值的行边界区域对应的行并入相邻的行边界区域对应的行，并重新计算行边界区域和行输入区域。
  7. 根据权利要求1所述的划分方法，还包括：
    计算各个行边界区域的高度；
    针对高度大于预定阈值的行边界区域，进行行分割操作，将所对应的行分割为多个子行。
  8. 根据权利要求7所述的划分方法，其中行分割操作包括：
    针对高度大于预定阈值的行边界区域，计算所覆盖的笔画的中心点，
    如果一个笔画的中心点的Y坐标位于另一笔画的Y范围内，则认为两个笔画属于相同的子行，其中，笔画的Y范围是该笔画的Y坐标范围；
    根据属于相同子行的笔画在Y方向上的起止范围，确定子行分割线，将对应行分割为各个子行，并计算覆盖各个子行的子行边界区域，其中，Y方向是与行的方向垂直的方向。
  9. 根据权利要求1所述的划分方法，还包括：
    针对各行，执行字符划分操作。
  10. 根据权利要求9所述的划分方法，其中，所述字符划分操作包括：
    计算每个笔画的中心点；
    如果一个笔画的中心点的X坐标位于另一笔画的X范围内，则认为两个笔画属于相同的字符，否则认为两个笔画属于不同的字符，
    其中，笔画的X范围是该笔画的X坐标范围，其中，X坐标是与行的方向平行的方向上的坐标。
  11. 根据权利要求10所述的划分方法，还包括：
    针对所划分的字符，计算每个字符的宽度和平均字符间距，如果一个字符与相邻字符的间距与平均字符间距的比例小于一阈值，且其与相邻字符的宽度之和小于另一阈值，则将所述字符与所述相邻字符合并成一个字符。
  12. 一种手写内容划分设备，包括：
    输入模块，被配置为接收用户的手写输入内容；
    处理模块，被配置为在新输入笔画时，对新输入的笔画进行采样，计算覆盖新输入的笔画的新笔画区域；
    存储器，被配置为存储笔画采样、表示行边界区域和行输入区域的参数；
    行判断模块，被配置为将新笔画区域与存储的行输入区域进行比对，如果新笔画区域与一个行输入区域相交，则将新输入的笔画划分为属于与 所述行输入区域相对应的行，并重新计算所述行的行边界区域和行输入区域，更新存储器中存储的所述行的行边界区域和行输入区域；如果新笔画区域与任意行输入区域都不相交，则认为新输入的笔画属于新的一行，将所述新笔画区域当做该新的一行的行边界区域，并根据预定规则计算该新的一行的行输入区域，将所计算的行边界区域和行输入区域存储在存储器中，其中，所述行输入区域覆盖行边界区域。
  13. 根据权利要求12的手写内容划分设备，其中，所述区域是矩形。
  14. 根据权利要求13的手写内容划分设备，其中，所述预定规则是以下之一：行输入区域比对应的行边界区域大预定比例；行输入区域的高度比对应的行边界区域大第一预定比例、宽度比对应的行边界区域大预定第二比例；行输入区域比对应的行边界区域大预定尺寸；行输入区域的高度比对应的行边界区域大第一预定尺寸、宽度比对应的行边界区域大第二预定尺寸。
  15. 根据权利要求14所述的手写内容划分设备，还包括：
    显示器，被配置为显示所述行输入区域。
  16. 根据权利要求12所述的手写内容划分设备，还包括：
    行划分优化模块，被配置为计算各个行边界区域的高度，针对高度低于预定阈值的行边界区域，进行行合并操作，将其并入其他行；针对高度大于预定阈值的行边界区域，进行行分割操作，将所对应的行分割为多个子行。
  17. 根据权利要求16所述的手写内容划分设备，其中，行划分优化模块包括行合并模块，被配置为：
    计算高度低于预定阈值的行边界区域与相邻的行边界区域的距离；
    如果所述距离小于预定距离，则将高度低于预定阈值的行边界区域对应的行并入相邻的行边界区域对应的行，并重新计算行边界区域和行输入区域。
  18. 根据权利要求16所述的手写内容划分设备，其中，行划分优化模块包括行分割模块，被配置为：
    针对高度大于预定阈值的行边界区域，计算所覆盖的笔画的中心点；
    如果一个笔画的中心点的Y坐标位于另一笔画的Y范围内，则认为 两个笔画属于相同的子行，其中，笔画的Y范围是该笔画的Y坐标范围；
    根据属于相同子行的笔画在Y方向上的起止范围，确定子行分割线，将对应行分割为各个子行，并计算覆盖各个子行的行边界区域，其中，Y方向是与行的方向垂直的方向。
  19. 根据权利要求12所述的手写内容划分设备，还包括：
    字符划分模块，被配置为针对各行，执行字符划分操作，包括：
    计算每个笔画的中心点；
    如果一个笔画的中心点的X坐标位于另一笔画的X范围内，则认为两个笔画属于相同的字符，否则认为两个笔画属于不同的字符，其中，笔画的X范围是该笔画的X坐标范围，其中，X坐标是与行的方向平行的方向上的坐标。
  20. 根据权利要求19所述的手写内容划分设备，字符划分模块还被配置为：
    针对所划分的字符，计算每个字符的宽度和平均字符间距，如果一个字符与相邻字符的间距与平均字符间距的比例小于一阈值，且其与相邻字符的宽度之和小于另一阈值，则将所述字符与所述相邻字符合并成一个字符。
  21. 一种手写内容编辑设备，包括：
    切换模块，被配置为在编辑模式和输入模式之间切换手写内容编辑设备的状态；
    根据权利要求12-20之一所述的手写内容划分设备，被配置为在手写内容编辑设备处于输入模式下启用；以及
    编辑模块，被配置为在所述手写内容编辑设备处于编辑模式下启用，实现对存储器中存储的手写输入内容的编辑，并将编辑后的手写输入内容存储在存储器中。
PCT/CN2015/093658 2014-11-03 2015-11-03 手写内容划分方法和设备、以及手写内容编辑设备 WO2016070785A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2017523206A JP2017534121A (ja) 2014-11-03 2015-11-03 手書き内容区分方法およびその機器、並びに手書き内容編集機器

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410608682.4A CN105630368A (zh) 2014-11-03 2014-11-03 手写内容划分方法和设备、以及手写内容编辑设备
CN201410608682.4 2014-11-03

Publications (1)

Publication Number Publication Date
WO2016070785A1 true WO2016070785A1 (zh) 2016-05-12

Family

ID=55908580

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/093658 WO2016070785A1 (zh) 2014-11-03 2015-11-03 手写内容划分方法和设备、以及手写内容编辑设备

Country Status (3)

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

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111722727A (zh) * 2019-03-22 2020-09-29 北京搜狗科技发展有限公司 应用于手写输入的模型训练方法、手写输入方法及装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110135426B (zh) * 2018-02-09 2021-04-30 北京世纪好未来教育科技有限公司 样本标注方法及计算机存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1127044A (zh) * 1993-05-20 1996-07-17 阿哈软件公司 手写、印刷和作图的电子表示的分组和处理方法与装置
CN101350068A (zh) * 2007-07-17 2009-01-21 佳能株式会社 显示控制设备和显示控制方法
CN102214043A (zh) * 2011-06-15 2011-10-12 安徽科大讯飞信息科技股份有限公司 一种手写输入识别方法及系统
KR101144675B1 (ko) * 2010-04-09 2012-05-24 주식회사 디오텍 개선된 연속 필기 입력 장치 및 그 방법
CN103268166A (zh) * 2013-05-17 2013-08-28 济南道能仁晟软件技术有限公司 手写输入设备的原笔迹信息采集和显示方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10307675A (ja) * 1997-05-01 1998-11-17 Hitachi Ltd 手書き文字認識方法及び装置
JP5481988B2 (ja) * 2009-07-21 2014-04-23 株式会社リコー 画像処理装置、制御方法及びプログラム

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1127044A (zh) * 1993-05-20 1996-07-17 阿哈软件公司 手写、印刷和作图的电子表示的分组和处理方法与装置
CN101350068A (zh) * 2007-07-17 2009-01-21 佳能株式会社 显示控制设备和显示控制方法
KR101144675B1 (ko) * 2010-04-09 2012-05-24 주식회사 디오텍 개선된 연속 필기 입력 장치 및 그 방법
CN102214043A (zh) * 2011-06-15 2011-10-12 安徽科大讯飞信息科技股份有限公司 一种手写输入识别方法及系统
CN103268166A (zh) * 2013-05-17 2013-08-28 济南道能仁晟软件技术有限公司 手写输入设备的原笔迹信息采集和显示方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111722727A (zh) * 2019-03-22 2020-09-29 北京搜狗科技发展有限公司 应用于手写输入的模型训练方法、手写输入方法及装置
CN111722727B (zh) * 2019-03-22 2024-01-12 北京搜狗科技发展有限公司 应用于手写输入的模型训练方法、手写输入方法及装置

Also Published As

Publication number Publication date
JP2017534121A (ja) 2017-11-16
CN105630368A (zh) 2016-06-01

Similar Documents

Publication Publication Date Title
KR102197501B1 (ko) 고정 서식 문서에서의 동아시아 레이아웃 특징들의 검출 및 재구성
CN103718149B (zh) 一种文本的处理方法及触屏设备
US20140013215A1 (en) Paragraph Alignment Detection and Region-Based Section Reconstruction
US8713464B2 (en) System and method for text input with a multi-touch screen
JP6430197B2 (ja) 電子機器および方法
US20140160032A1 (en) Swipe Stroke Input and Continuous Handwriting
KR102075433B1 (ko) 필기 입력 장치 및 그 제어 방법
TW201447611A (zh) 用於編輯墨文本資料的裝置及方法
JP6054547B2 (ja) 手書き文書情報を処理するための電子機器および方法
WO2016070785A1 (zh) 手写内容划分方法和设备、以及手写内容编辑设备
WO2024140094A1 (zh) 数字文档的段落确定方法、装置、电子设备及存储介质
JP2018067298A (ja) 手書き内容編集装置および手書き内容編集方法
JP6667452B2 (ja) テキスト情報の入力方法及び装置
WO2016192664A1 (zh) 手写表识别方法和设备
CN114663902B (zh) 文档图像处理方法、装置、设备和介质
JP2000099223A (ja) 手書き文字入力インタフェースを有するデータ処理装置及び記録媒体
US20140289619A1 (en) Information display device
TW201823966A (zh) 電子設備的輸入方法及系統
US9804777B1 (en) Gesture-based text selection
TWI598748B (zh) 電子設備及字元校正方法
CN111078028A (zh) 输入方法、相关设备及可读存储介质
JP5666011B1 (ja) 方法及び電子機器
US20100123719A1 (en) Method and tool for creating irregular-shaped tables
US20240143160A1 (en) Electronic whiteboard system and operation method thereof
US20230060045A1 (en) Systems and methods for grid-aligned inking

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15857029

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017523206

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15857029

Country of ref document: EP

Kind code of ref document: A1