CN106293185A - Hand-written table recognition methods and equipment - Google Patents
Hand-written table recognition methods and equipment Download PDFInfo
- Publication number
- CN106293185A CN106293185A CN201510305645.0A CN201510305645A CN106293185A CN 106293185 A CN106293185 A CN 106293185A CN 201510305645 A CN201510305645 A CN 201510305645A CN 106293185 A CN106293185 A CN 106293185A
- Authority
- CN
- China
- Prior art keywords
- point
- form line
- hand
- ruling
- written
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
- G06F3/041—Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to hand-written table recognition methods and equipment.The method includes: detection form line;Determine the longest crossing vertical form line and the longest water-glass ruling;Calculate the bounding rectangles just covering all form lines;For each form line, bounding rectangles extends, obtain the Virtual table ruling of correspondence;Calculate the position of the central point of the frame divided by Virtual table ruling and form line;Each central point is divided into internal point and external point, and wherein internal point is the central point surrounded by form line;Internal point being grouped, if the path between adjacent inner point is not stopped by form line, the most described internal point belongs to identical group;The frame combination corresponding to the internal point of identical group will be belonged to, be formed as cell;And identify the handwritten content in unit lattice.Hand-written table recognition methods according to embodiments of the present invention and equipment, it is possible to hand-written table is identified efficient and highly accurately.
Description
Technical field
The present invention relates to hand-written table identification technology, more particularly, to a kind of hand-written table identification side
Method and equipment, it is possible to identify hand-written table efficient and highly accurately and extract content therein.
Background technology
Along with the development of information technology, electronic equipment (such as, personal digital assistant, hand-held electricity
Brain, mobile phone) etc. use more and more universal in the life of people.These electronic equipments defeated
Enter mode and substantially can be divided into input through keyboard and the big class of handwriting input two.Keyboard entry method is adopted
With the keyboard on electronic equipment, each button in user point beating keyboard inputs.Hand-written
Input mode receives the input of user by the touch screen of electronic equipment, processes input
To identify the content of user's input.This needs electronic equipment to have higher identification ability, the most right
Its accuracy rate identified proposes higher requirement.
Chinese patent application 200710178961.1 proposes a kind of hand-written table identification equipment.This sets
For extracting hand-written table from input content.But, this equipment supposes that the size of table is more than certain
Individual threshold value, and suppose to there are not irregular table unit lattice.
Therefore, there is too much parameter and limit in the method for prior art, and this will have a strong impact on institute
The scene that can apply.Such as, most existing methods according to the line width of specific form line and
Line length extracts form line.Additionally, the irregular unit of method None-identified of prior art
Lattice.These methods are only capable of the cell processing regular shape.
Accordingly, it would be desirable to a kind of mechanism that can hand-written table be identified efficient and highly accurately.
Summary of the invention
Present disclosure proposes a kind of hand-written table recognition methods and equipment, it is possible to efficient and highly accurately
Identify hand-written table the content extracting in table.
According to an aspect of the present invention, it is proposed that a kind of hand-written table recognition methods, including: inspection
Survey form line;Calculate the bounding rectangles just covering all form lines;For each form line,
Bounding rectangles extends, obtains the Virtual table ruling of correspondence;Calculate by Virtual table ruling and table
The position of the central point of the frame that ruling divides;Each central point is divided into internal point and external point,
Wherein internal point is the central point surrounded by form line;Internal point is grouped, if adjacent inner
Path between point is not stopped by form line, and the most described internal point belongs to identical group;To belong to
Combine in the frame corresponding to the internal point of identical group, be formed as cell;And identify that each is single
Handwritten content in unit's lattice.
Preferably, detection form line includes: calculate the size of average stroke;Length is much larger than
The Stroke discrimination of average stroke size is form line.
Preferably, detection form line also includes: whether detection form line exists flex point on path;
And the form line that there is flex point is divided into single form line at flex point.
Preferably, the bounding rectangles calculating all form lines of covering just comprises determining that crossing
The longest vertical form line and the longest water-glass ruling;It is the longest that calculating covering just is intersected
Vertical form line and the bounding rectangles of the longest water-glass ruling.
Preferably, if the top of two root table rulings is at a distance of the pixel less than predetermined number, then recognize
It is that two root table rulings intersect.
Preferably, be divided into internal point and external point also to include each central point: will with adjacent outside
The internal point that portion's point is separated by Virtual table ruling changes into external point;And execution above-mentioned steps,
Until not having internal point to be changed to external point.
According to the second aspect of the invention, it is provided that a kind of hand-written table identification equipment, including table
Ruling detection module, is configured to detect form line;Table area detection module, is configured to
Calculate the bounding rectangles just covering all form lines;List structure identification module, for each table
Ruling, extends on the bounding rectangles that table area detection module is calculated, and obtains the void of correspondence
Intend form line;Calculate the position of the central point of the frame divided by Virtual table ruling and form line;
Each central point is divided into internal point and external point, during wherein internal point is surrounded by form line
Heart point;Internal point is grouped, if the path between adjacent inner point is not stopped by form line,
The most described internal point belongs to identical group;The frame combination corresponding to the internal point of identical group will be belonged to,
Be formed as cell;And table unit lattice content extraction module, it is configured to identify unit
Handwritten content in lattice.
Unlike the prior art, hand-written table recognition methods and equipment according to embodiments of the present invention exists
Many aspects improve the performance of hand-written table identification, at least include:
Even if 1. there is different size of table in sample space, it is also possible to detect these different big
Little table;
Even if 2. there is irregular cell, it is also possible to correctly identify the structure of table;
3. for direction and the threshold value not requirement pre-set of table.Such as, big for character
Little not restriction, nor need the priori about threshold value.
Accompanying drawing explanation
By the preferred embodiments of the present invention are described below in conjunction with the accompanying drawings, by making, the present invention's is above-mentioned
And other objects, features and advantages are clearer, wherein:
Fig. 1 shows the schematic block diagram of hand-written table identification equipment according to embodiments of the present invention.
Fig. 2 shows the table of an example, wherein illustrate only form line.
Fig. 3 shows the table shown in Fig. 2, illustrated therein is Virtual table ruling.
Fig. 4 shows the structure of the table shown in Fig. 2.
Fig. 5 shows the central point division result of the table shown in Fig. 2.
Fig. 6 shows the final analysis result to the table shown in Fig. 2.
Fig. 7 shows the division result of the central point of the table of another example.
Fig. 8 shows the flow chart of hand-written table recognition methods according to embodiments of the present invention.
Detailed description of the invention
Referring to the drawings, the example embodiment of the present invention is described in detail.Retouch following
In stating, some specific embodiments are only used for describing purpose, and should not be construed and have the present invention
Any restriction, and the example of the simply present invention.It is mixed the understanding of the present invention may be caused to cause
When confusing, conventional structure or structure will be omitted.
Fig. 1 shows the schematic block of hand-written table identification equipment 100 according to embodiments of the present invention
Figure.This hand-written table identification equipment 100 includes: handwriting input trapping module 110, form line are examined
Survey module 120, table area detection module 130, list structure identification module 140 and table unit lattice
Content extraction module 150.
Handwriting input trapping module 110 is for capturing the handwriting input content of user.Such as, exist
Line data capture module can be touch screen and the processor of hand-written table identification equipment, and user utilizes
Writing pencil or finger input the most on the touchscreen, and it is defeated that processor records user in real time
The stroke entered.
Table line detecting module 120 is configured to detect form line.According to an embodiment, can
Form line is identified with the size according to stroke.Such as, average stroke size is calculated.If one
The size of a little strokes is much larger than average stroke size, then it is assumed that these strokes are form lines.This is
Because the length that the form line of level is in the horizontal direction is at least above the length of a stroke, hang down
Straight form line length in vertical direction is at least above the length of a stroke.Therefore, logical
Cross and compare the size/length of stroke and can identify form line.
Table area detection module 130 is configured to detect table area, i.e. calculates and just covers
The bounding rectangles of lid form line.In one embodiment, divide according to the length of the form line of detection
The other form line to level and vertical form line are ranked up.Select the longest vertical form line
The longest water-glass ruling.Check the longest vertical form line and the longest water-glass ruling it
Between whether intersect.If both intersect, then mean to there is a table.Calculate and just can cover
Cover the longest vertical form line and the bounding rectangles of the longest water-glass ruling intersected, obtained
Bounding rectangles be the table area existing for table.Table is to be limited by the form line intersected vertically
Region, therefore, by detect whether exist intersect form line, it may be determined whether
Table, and by detecting the longest form line intersected, it may be determined that the region of table.Due to hands
The randomness write, two form lines may will not intersect just.Therefore, according to an embodiment,
If the top of two root table rulings is at a distance of the pixel less than predetermined number, then it is assumed that two root table rulings
Intersect.Such as, the pixel of predetermined number can be 10 pixels.Additionally, in the present embodiment,
" level " and " vertically " that be previously mentioned also needs not to be horizontal and vertical completely, and detects
To form line also need not to be straight line.In reality, user's hand-drawn line is unlikely to be the most straight
Line, it is also not possible to be horizontal and vertical completely.It may for instance be considered that at vertical " 10% "
Line in the range of Zuo You is vertical line, and the line in the range of " 15% " left and right of level is
Horizontal line.In another embodiment, table can be detected by detecting the coordinate of form line
Lattice region.Such as, detection is positioned at the most left and the rightest form line, as the horizontal extent of table,
Detection is positioned at the form line of highest and lowest, as the vertical range of table, then calculates and covers table
Go up most, under, the bounding rectangles of the most left and the rightest form line, as the region of table.
List structure identification module 140 is configured to the structure of identification table.First, for all of
Form line L, extends on bounding rectangles, obtains the Virtual table ruling Lv of correspondence.Virtual tables
Form line L corresponding for line Lv has identical direction.Such as, if Virtual table ruling
Be horizontal line, then it is line from left to right in the input space.If Virtual table ruling is vertical
Straight line, then it is line from top to bottom in the input space.Fig. 2 shows the table of an example,
Wherein thick lines represent the form line detected.Fig. 3 shows the table shown in Fig. 2, wherein shows
Having gone out the Virtual table ruling of correspondence, wherein hachure represents Virtual table ruling.Such as, for water
Flat form line, the Virtual table ruling of its correspondence is on bounding rectangles at same vertical position, respectively
Extend the line obtained to the left and to the right.For vertical form line, the Virtual table ruling of its correspondence is
Place located horizontally from the emitter on bounding rectangles, extend the line obtained the most up and down.In other words,
Form line is extended on bounding rectangles, available corresponding Virtual table ruling.Therefore, border
Rectangle is divided into multiple frame by Virtual table ruling or form line.Then, according to the position of each frame,
Calculate the position of the central point of each frame.Fig. 4 shows the structure of the table of an example, wherein
Show the central point of each frame.Then, central point is divided into internal point and external point.If
One central point is surrounded on four direction by form line, and the most lower left and right all exists form
Line, then it is assumed that this central point is internal point.Here, the most whether central point exists
Form line is not limited to the form line/Virtual table ruling of its place frame, and refers to that it is in the direction
Above, in bounding rectangles, whether there is form line.If at least one direction of a central point
On there is not form line, i.e. it only exists Virtual table ruling at least one direction, then recognize
It is external point for this central point.Fig. 5 shows the division result of the central point of the table of an example.
Then, central point is grouped.If the path between two adjacent central points is not by form line
Stopped, i.e. be Virtual table ruling between two central points, then it is assumed that the two central point belongs to
In identical group.It is combined as a cell by belonging to the frame corresponding to the internal point of identical group,
Even if the cell obtained is irregularly shaped.It is final that Fig. 6 shows the table shown in Fig. 2
Analysis result.
Handwritten content in the table unit lattice content extraction module 150 unit lattice to being identified
It is identified.
Fig. 1 also show display 160.Hand-written table identification equipment 100 can be by recognition result
Display is on display 160.For example, it is possible to represent different cells with different background colours,
User is made to easily verify that unit lattice.In some instances, display is also this equipment
Touch screen.
According to an embodiment, table line detecting module 120 also includes that stroke splits module 1210,
It is configured to the form line of even pen is divided into single form line.User connects pen sometimes and writes out phase
Adjacent form line.Needed to separate these form lines being connected before carrying out hand-written table identification.
Whether detection form line exists flex point, if there is flex point on their path, then it is assumed that this
A little form lines are the form lines of even pen, split by these form lines, obtain multiple list at flex point
Root table ruling.
According to an embodiment, in list structure identification module 140 is additionally configured to for identifying
Portion's point and external point perform adjustment, the internal point change that will separate by Virtual table ruling with external point
For external point.Such as, Fig. 7 shows the division result of central point of table of an example.Should
Table in example and the table shown in Fig. 5 differ only in rightmost form line unclosed,
Therefore, whether the internal point obtained and external point division knot are surrounded by form line according to central point
Fruit is as shown in Fig. 7 (a).In this embodiment, list structure identification module 140 is for dividing knot
Fruit is adjusted, and the internal point separated by Virtual table ruling with adjacent external point is changed into outside
Point.That is, for the central point of frame of the second row first row in Fig. 7 (a), changed into
External point, obtains the division result shown in Fig. 7 (b).This is because in this example, table
The L-shaped region in the lower left corner does not form a cell closed, and therefore, is not the one of table
Part.List structure identification module 140 performs repeatedly to adjust for division result, until in not having
Till portion's point is changed to external point, i.e. all of central point has been adjusted by.
Fig. 8 shows the flow chart of hand-written table recognition methods 800 according to embodiments of the present invention.
Hand-written table recognition methods according to embodiments of the present invention is applied to handwriting input device, and user is at hands
Write and on input equipment, carry out handwriting input.When needs identify that user inputs, real according to the present invention
The hand-written table recognition methods executing example starts.First, at step S810, detect form line.
Then, at step S820, calculate the bounding rectangles just covering all form lines.Permissible
It is determined by the longest vertical form line that intersects and the longest water-glass ruling, calculates and just cover
Cover the longest vertical form line and the bounding rectangles of the longest water-glass ruling intersected, covered
Cover the bounding rectangles of all form lines.At step S830, for each form line, on limit
Extend on boundary's rectangle, obtain the Virtual table ruling of correspondence, calculate by Virtual table ruling and form line
The position of the central point of the frame divided.Each central point is divided into internal point and external point, wherein
Internal point is the central point surrounded by form line.By internal point be grouped, if adjacent inner point it
Between path do not stopped by form line, the most described internal point belongs to identical group.Phase will be belonged to
With the combination of the frame corresponding to internal point of group, be formed as cell.Finally, in step S840
Place, identifies the handwritten content in unit lattice.
In step S810, form line can be identified according to the size of stroke.Such as, meter
Calculate average stroke size.If the size of some strokes is much larger than average stroke size, then it is assumed that
These strokes are form lines.This is because the length that the form line of level is in the horizontal direction is at least
More than the length of a stroke, vertical form line length in vertical direction is at least above one
The length of individual stroke.Therefore, the size/length by comparing stroke can identify form line.Root
According to this embodiment, can identify that there is different size of table.Because interior in different size of table
Container has different stroke sizes.According to average stroke size relatively detect form line not
Need to pre-set any threshold value, it is not required that know set threshold value.
In step S820, can be according to the length of the form line of detection table to level respectively
Ruling and vertical form line are ranked up, then identify intersect the longest vertical form line and
The longest water-glass ruling.Due to hand-written randomness, two form lines may phase just
Hand over.Therefore, according to an embodiment, if the top of two root table rulings is at a distance of less than predetermined number
Purpose pixel, then it is assumed that two root table rulings intersect.Such as, the pixel of predetermined number can be 10
Individual pixel.
The form line detected in step S810 is probably many root tables ruling of connection.Therefore,
In step S810, also include that the form line by even pen is divided into the step of single form line.With
Family connects pen sometimes and writes out adjacent form line.Needed these before carrying out hand-written table identification
The form line being connected separates.Whether detection form line exists flex point on their path, as
There is flex point in fruit, then it is assumed that these form lines are the form lines of even pen, are being turned by these form lines
Split at Dian, obtain multiple single form line.
The internal point obtained in step S830 and external point may need to adjust further.At this
In the case of Zhong, the internal point separated by Virtual table ruling with external point is changed into external point.Pin
Perform repeatedly to adjust to division result, until not having internal point to be changed to external point.
Hand-written table recognition methods and equipment according to embodiments of the present invention can apply to online handwriting
Input equipment, including flat board PC, has the desktop PC of touch screen, mobile phone, PDA etc.
There is the electronic equipment of online handwriting input function.User can be carried out on this electronic equipment
Handwriting input and editor.Hand-written table can be identified by electronic equipment efficient and highly accurately.
Such as, hand-written table recognition methods according to embodiments of the present invention, can after identifying a table
To identify further for hand-written space, to identify other table that may be present.Due to
Utilize stroke size to identify form line, so the size for table does not limit.With existing skill
Art uses threshold value, will be greater than the stroke of threshold value and be considered as the technology of form line and compare, according to this
The method of bright embodiment can identify that different size of table is without arranging any threshold value.Additionally,
By the structure of table is analyzed, even if table exists irregular cell, it is also possible to just
True ground recognition unit lattice.
Other of the embodiment of the present invention disclosed herein arrange the side including performing formerly to summarize
The software program of the steps and operations of method embodiment.More specifically, computer program be as
Under a kind of embodiment: there is computer-readable medium, on computer-readable medium coding have meter
Calculating machine program logic, when performing on the computing device, computer program logic provides relevant
Operation, thus technique scheme is provided.When holding at least one processor in the system of calculating
During row, computer program logic makes processor perform the operation (side described in the embodiment of the present invention
Method).This set of the present invention is typically provided as arranging or encoding at such as light medium (such as
CD-ROM), the software on the computer-readable medium of floppy disk or hard disk etc., code and/or its
On his data structure or the most one or more ROM or RAM or PROM chip
Other media of firmware or microcode or special IC (ASIC) or one or more
Downloadable software image in module, shared data bank etc..Software or firmware or this configuration
May be installed on calculating equipment, so that the one or more processors in calculating equipment perform basis
Technology described in inventive embodiments.In conjunction with in such as one group of data communications equipment or other entities
The software process that calculating equipment carries out operating can also provide the equipment according to the present invention.According to this
The multiple software process that can also be distributed in multiple data communications equipment of equipment of invention or
Transport on all software process run on one group of small, dedicated computer or single computer
Between all software process of row.
It should be understood that strictly say, embodiments of the invention can be implemented as on computer equipment
Software program, software and hardware or individually software and/or individually circuit.
It should be noted that, in the above description, the most in an illustrative manner, it is shown that this
Bright technical scheme, but it is not intended that the invention be limited to above-mentioned steps and cellular construction.?
When possible, as required step and cellular construction can be adjusted and accept or reject.Cause
This, some step and unit not implement element necessary to the overall invention thought of the present invention.
Therefore, technical characteristic essential to the invention is limited solely by the overall invention being capable of the present invention
The minimum requirements of thought, and do not limited by above instantiation.
So far already in connection with preferred embodiment, invention has been described.It should be understood that ability
Field technique personnel without departing from the spirit and scope of the present invention, can carry out various its
Its change, replace and add.Therefore, the scope of the present invention is not limited to above-mentioned particular implementation
Example, and should be defined by the appended claims.
Claims (14)
1. a hand-written table recognition methods, including:
Detection form line;
Calculate the bounding rectangles just covering all form lines;
For each form line, bounding rectangles extends, obtain the Virtual table ruling of correspondence;
Calculate the position of the central point of the frame divided by Virtual table ruling and form line;
Each central point is divided into internal point and external point, and wherein internal point is to be surrounded by form line
Central point;
Internal point is grouped, if the path between adjacent inner point is not stopped by form line,
The most described internal point belongs to identical group;
The frame combination corresponding to the internal point of identical group will be belonged to, be formed as cell;And
Identify the handwritten content in unit lattice.
Hand-written table recognition methods the most according to claim 1, wherein, detects form line bag
Include:
Calculate the size of average stroke;And
It is form line by the Stroke discrimination that length is much larger than average stroke size.
Hand-written table recognition methods the most according to claim 1 and 2, wherein, detects form
Line also includes:
Whether detection form line exists flex point on path;And
The form line that there is flex point is divided into single form line at flex point.
Hand-written table recognition methods the most according to claim 1 and 2, wherein, calculates just
The bounding rectangles covering all form lines includes:
Determine the longest crossing vertical form line and the longest water-glass ruling;
Calculate and just cover the longest vertical form line intersected and the limit of the longest water-glass ruling
Boundary's rectangle.
Hand-written table recognition methods the most according to claim 1 and 2, wherein, if two
The top of form line is at a distance of the pixel less than predetermined number, then it is assumed that two root table rulings intersect.
Hand-written table recognition methods the most according to claim 1 and 2, wherein, by each
Heart point is divided into internal point and external point also to include:
The internal point separated by Virtual table ruling with adjacent external point is changed into external point;And
Perform above-mentioned steps, until not having internal point to be changed to external point.
Hand-written table recognition methods the most according to claim 3, wherein, described method is applied
In online handwriting input equipment.
8. a hand-written table identification equipment, including:
Table line detecting module, is configured to detect form line;
Table area detection module, is configured to calculate the shape based moment just covering all form lines
Shape;
List structure identification module, is configured to, for each form line, detect mould at table area
Extend on the bounding rectangles that block is calculated, obtain the Virtual table ruling of correspondence;Calculate by Virtual table
The position of the central point of the frame that ruling and form line are divided;Each central point is divided into internal point
And external point, wherein internal point is the central point surrounded by form line;Internal point is grouped, as
Really the path between adjacent inner point is not stopped by form line, and the most described internal point belongs to identical
Group;The frame combination corresponding to the internal point of identical group will be belonged to, be formed as cell;And
Table unit lattice content extraction module, is configured to the handwritten content in identification unit lattice.
Hand-written table recognition methods the most according to claim 8, wherein, described form line is examined
Survey module is configured to:
Calculate the size of average stroke;
It is form line by the Stroke discrimination that length is much larger than average stroke size.
Hand-written table identification equipment the most according to claim 8 or claim 9, wherein said form
Line detection module includes:
Stroke segmentation module, is configured to detect whether form line exists flex point on path;With
And the form line that there is flex point is divided into single form line at flex point.
11. hand-written table identification equipment according to claim 8 or claim 9, wherein, described table
Lattice region detection module is configured in the following manner and calculates the limit just covering all form lines
Boundary's rectangle:
Determine the longest crossing vertical form line and the longest water-glass ruling, and
Calculate and just cover the longest vertical form line intersected and the limit of the longest water-glass ruling
Boundary's rectangle.
12. hand-written table identification equipment according to claim 8 or claim 9, wherein, if two
The top of root table ruling is at a distance of the pixel less than predetermined number, then it is assumed that two root table rulings intersect.
13. hand-written table identification equipment according to claim 8 or claim 9, wherein, list structure
Identification module is also configured to
The internal point separated by Virtual table ruling with adjacent external point is changed into external point;And
Perform above-mentioned steps, until not having internal point to be changed to external point.
14. hand-written table identification equipment according to claim 10, also include:
Handwriting input trapping module, for capturing the handwriting input content of user.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510305645.0A CN106293185A (en) | 2015-06-05 | 2015-06-05 | Hand-written table recognition methods and equipment |
PCT/CN2016/084680 WO2016192664A1 (en) | 2015-06-05 | 2016-06-03 | Handwriting table recognition method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510305645.0A CN106293185A (en) | 2015-06-05 | 2015-06-05 | Hand-written table recognition methods and equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106293185A true CN106293185A (en) | 2017-01-04 |
Family
ID=57440288
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510305645.0A Pending CN106293185A (en) | 2015-06-05 | 2015-06-05 | Hand-written table recognition methods and equipment |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106293185A (en) |
WO (1) | WO2016192664A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109522805A (en) * | 2018-10-18 | 2019-03-26 | 成都中科信息技术有限公司 | A kind of form processing method for Form ballot paper in community election |
CN109656459A (en) * | 2017-10-10 | 2019-04-19 | 夏普株式会社 | Electronic equipment, control device, the control method of electronic equipment and storage medium |
CN109934160A (en) * | 2019-03-12 | 2019-06-25 | 天津瑟威兰斯科技有限公司 | Method and system for extracting table character information based on table recognition |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0863610A (en) * | 1994-08-18 | 1996-03-08 | Mitsubishi Electric Corp | Document processor |
US7412094B2 (en) * | 2004-09-21 | 2008-08-12 | Microsoft Corporation | System and method for editing a hand-drawn table in ink input |
US7583841B2 (en) * | 2005-12-21 | 2009-09-01 | Microsoft Corporation | Table detection in ink notes |
CN101452523B (en) * | 2007-12-07 | 2011-01-26 | 汉王科技股份有限公司 | Handwriting table identification method and apparatus |
US8600164B2 (en) * | 2008-03-28 | 2013-12-03 | Smart Technologies Ulc | Method and tool for recognizing a hand-drawn table |
JP2014127188A (en) * | 2012-12-27 | 2014-07-07 | Toshiba Corp | Shaping device and method |
-
2015
- 2015-06-05 CN CN201510305645.0A patent/CN106293185A/en active Pending
-
2016
- 2016-06-03 WO PCT/CN2016/084680 patent/WO2016192664A1/en active Application Filing
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109656459A (en) * | 2017-10-10 | 2019-04-19 | 夏普株式会社 | Electronic equipment, control device, the control method of electronic equipment and storage medium |
CN109522805A (en) * | 2018-10-18 | 2019-03-26 | 成都中科信息技术有限公司 | A kind of form processing method for Form ballot paper in community election |
CN109522805B (en) * | 2018-10-18 | 2020-12-08 | 成都中科信息技术有限公司 | Form processing method for form vote in community election |
CN109934160A (en) * | 2019-03-12 | 2019-06-25 | 天津瑟威兰斯科技有限公司 | Method and system for extracting table character information based on table recognition |
Also Published As
Publication number | Publication date |
---|---|
WO2016192664A1 (en) | 2016-12-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103065134B (en) | A kind of fingerprint identification device and method with information | |
CN110232311B (en) | Method and device for segmenting hand image and computer equipment | |
US8977059B2 (en) | Integrating feature extraction via local sequential embedding for automatic handwriting recognition | |
US9104242B2 (en) | Palm gesture recognition method and device as well as human-machine interaction method and apparatus | |
US9135503B2 (en) | Fingertip tracking for touchless user interface | |
JP5211334B2 (en) | Handwritten symbol recognition method and apparatus | |
US20130283378A1 (en) | System and method for distinguishing human swipe input sequence behavior and using a confidence value on a score to detect fraudsters | |
US10311295B2 (en) | Heuristic finger detection method based on depth image | |
US9122353B2 (en) | Kind of multi-touch input device | |
CN105320248A (en) | Mid-air gesture input method and device | |
KR20140045573A (en) | Text detection using image regions | |
Jinda-Apiraksa et al. | A simple shape-based approach to hand gesture recognition | |
CN104484643A (en) | Intelligent identification method and system for hand-written table | |
CN102073870A (en) | Method for recognizing Chinese character handwriting on touch screen | |
CN112464797B (en) | Smoking behavior detection method and device, storage medium and electronic equipment | |
CN104299004A (en) | Hand gesture recognition method based on multi-feature fusion and fingertip detecting | |
CN108960221B (en) | Bank card identification method and device based on image | |
CN104915944A (en) | Method and device for determining black margin position information of video | |
CN106293185A (en) | Hand-written table recognition methods and equipment | |
CN110427909B (en) | Mobile terminal driving license detection method and system, electronic equipment and storage medium | |
KR101967858B1 (en) | Apparatus and method for separating objects based on 3D depth image | |
CN115589786A (en) | Method, device and system for recognizing hand-drawn figure and computer readable storage medium | |
CN108255298B (en) | Infrared gesture recognition method and device in projection interaction system | |
Posner et al. | A single camera based floating virtual keyboard with improved touch detection | |
CN107145883A (en) | Method for text detection and equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170104 |
|
WD01 | Invention patent application deemed withdrawn after publication |