The present invention relates generally to digital ink processing and in particular, to a method and tool for recognizing a hand-drawn table.
Interactive input systems that allow users to input ink into an application program using an active pointer (eg. a pointer that emits light, sound or other signal), a passive pointer (eg. a finger, cylinder or other suitable object) or other suitable input device such as for example, a mouse or trackball, are known. These interactive input systems include but are not limited to: touch systems comprising touch panels employing analog resistive or machine vision technology to register pointer input such as those disclosed in U.S. Pat. Nos. 5,448,263; 6,141,000; 6,337,681; 6,747,636; 6,803,906; 7,232,986; 7,236,162; and 7,274,356 assigned to SMART Technologies ULC of Calgary, Alberta, Canada, assignee of the subject application, the contents of which are incorporated by reference; touch systems comprising touch panels employing electromagnetic, capacitive, acoustic or other technologies to register pointer input; tablet personal computers (PCs); laptop PCs; personal digital assistants (PDAs); and other similar devices. Although these interactive input systems are able to recognize handwriting well, their ability to recognize and represent the meaning of hand-drawn objects and in particular, charts or tables, has been limited.
Microsoft Office software includes a feature that enables tables to be created by moving a cursor over an open document. During table creation, cursor movements over the open document are translated immediately into one of rectangles, squares and straight lines. Although tables can be created, the functionality of this feature limits its ease of use in interactive input systems that accept freeform or handwritten ink as input. As will be appreciated, improvements in hand-drawn table recognition are desired.
It is therefore an object of the present invention to provide a novel method and tool for recognizing a hand-drawn table.
Accordingly, in one aspect there is provided a computerized method of recognizing an input hand-drawn table formed by freeform line objects, said method comprising:
transforming each freeform line object of said table into one of a vertical line segment and a horizontal line segment;
generating a grid system based on the vertical and horizontal line segments; and
converting the generated grid system into a table object.
According to another aspect there is provided a computerized digital ink recognition method comprising:
selecting digital ink presented on a graphical user interface, said digital ink at least comprising intersecting freeform lines representing a hand-drawn table; and
performing a recognition procedure on said intersecting freeform lines to recognize the hand-drawn table;
generating a table object corresponding to the hand-drawn table in response to recognition thereof.
According to another aspect there is provided a computer readable medium embodying computer program code for digital ink recognition, said computer program code comprising:
program code for selecting digital ink presented on a graphical user interface, said digital ink at least comprising intersecting freeform lines representing a hand-drawn table;
program code for performing a recognition procedure on said intersecting freeform lines to recognize the hand-drawn table;
program code for generating a table object corresponding to the hand-drawn table in response to recognition thereof.
According to yet another aspect there is provided a computer readable medium embodying computer program code for recognizing an input hand-drawn table formed by freeform line objects, said computer program code comprising:
program code for transforming each freeform line object of said table into one of a vertical line segment and a horizontal line segment;
program code for generating a grid system based on the vertical and horizontal line segments; and
program code for converting the generated grid system into a table object.
According to still yet another aspect there is provided an interactive input system comprising:
a coordinate input device having a display surface on which an image is presented, said coordinate input device being responsive to user input; and
processing structure communicating with said coordinate input device and updating the image presented on said display surface to reflect user input, said processing structure executing a table recognition procedure when invoked to recognize a hand-drawn table entered on said coordinate input device and convert the recognized hand-drawn table into a table object.
Embodiments will now be described more fully with reference to the accompanying drawings in which:
a is a block diagram of an interactive input system;
b is a block diagram showing the components of a hand-drawn table recognizing tool;
a to 2e show a graphical user interface presented on the display surface of a coordinate input device during hand-drawn table creation and recognition;
a to 3c, 4a and 4b show the graphical user interface during irregular hand-drawn table creation and recognition;
a to 5d show the graphical user interface during hand-drawn table creation in the presence of picture and text objects and recognition of the created hand-drawn table.
In the following, a software tool and method for recognizing freeform or hand-drawn tables input into an interactive input system such as those referred above, is described. The software tool when executed by a processing device transforms the freeform line objects or “lines” forming the hand-drawn table into horizontal and vertical line segments, generates a grid system based on the vertical and horizontal line segments and converts the generated grid system into a table object that is compatible with conventional word processing software such as for example Microsoft Word™.
Turning now to
The processing structure 14 in this embodiment is a general purpose computing device in the form of a computer. The computer comprises for example, a processing unit, system memory (volatile and/or non-volatile memory), other non-removable or removable memory (a hard disk drive, RAM, ROM, EEPROM, CD-ROM, DVD, flash memory etc.) and a system bus coupling the various computer components to the processing unit.
The processing structure 14 runs a host software application such as SMART Notebook™ offered by SMART Technologies ULC. As is known, during execution, the SMART Notebook™ application provides a graphical user interface comprising a canvas page or palette, that is presented on the display surface of the coordinate input device 12 on which freeform or handwritten ink objects and other objects can be input and manipulated via pointer interaction with the coordinate input device. The SMART Notebook™ application in this embodiment, is provided with a tool for recognizing hand-drawn tables and converting such tables into corresponding table objects.
b is a block diagram showing the components of the hand-drawn table recognizing tool 16. As can be seen, the hand-drawn table recognizing tool comprises an object classifier 16a to classify selected objects into table outline elements and table cell content elements, a freeform line object transformer 16a to transform table outline elements into vertical and horizontal line segments, a grid system generator 16c to create a grid system identifying the intersection points of the vertical and horizontal line segments and a table object generator 16d to create a table object based on the intersection points.
a and 2b show the graphical user interface 20 presented on the display surface of the coordinate input device 12. In
c and 2d show the hand-drawn table 22 when selected as well as a drop-down menu 30 that is presented to the user allowing the user to select and invoke the hand-drawn table recognizing tool 16. Once invoked, the hand-drawn table recognizing tool 16 converts the hand-drawn table into a table object 32 with straight horizontal and vertical lines 34 and 36 respectively as shown in
a to 3c and
a shows side-by-side picture objects 50 and 52 and associated underlying text objects 54 and 56 that have been input on the canvas page 24 of the graphical user interface 20. In
Turning now to
The steps performed during transformation of the table outline elements into vertical and horizontal line segments (step 102 of
(step 156) and the orientation value is checked to determine if it is greater than or equal to one (1) (step 158).
If the orientation value of the temporary line segment is greater than or equal to one, the temporary line segment is transformed into a horizontal line segment extending from the minimum x coordinate value to the maximum x coordinate value and having a y coordinate value equal to (ymax+ymin)/2 and denoted yavg (step 160). Thus, the horizontal line segment has a start point (xmin, yavg) and an end point (xmax, yavg). The horizontal line segment is then compared to a horizontal line segment list to determine if the horizontal line segment is located at substantially the same y coordinate position as a horizontal line segment that already exists in the horizontal line segment list (step 162). In this embodiment, the horizontal line segment is deemed to be at the same y coordinate position as an existing horizontal line segment if the y coordinate of the horizontal line segment is within ten (10) pixels of the y coordinate of an existing horizontal line segment. If the horizontal line segment is deemed to be at the same y coordinate position as an existing horizontal line segment, the start point and end point of the horizontal line segment are compared with the start point and end point of that existing horizontal line segment to determine if the horizontal line segment and the existing horizontal line segment overlap. The horizontal line segment and the existing horizontal line segment are deemed to overlap if one of the following conditions holds true:
XEmin≦Xmin≦XEmax
XEmin≦Xmax≦XEmax
Xmin≦XEmin≦Xmax
Xmin≦XEmax≦Xmax
where:
XEmin and XEmax are the x coordinates of the start and end points respectively of the existing horizontal line segment; and
Xmin and Xmax are the x coordinates of the start and end points respectively of the horizontal line segment.
If the horizontal line segment is determined not to overlap with the existing horizontal line segment, the horizontal line segment is added to the horizontal line segment list (step 164). A check is then made to determine if any other classified table outline element exists that has not been selected (step 166). If no non-selected classified table outline element exists, the process is deemed completed. Otherwise, the process reverts back to step (150) and the next classified table outline element is selected.
At step 162, if the horizontal line segment is determined to overlap with the existing horizontal line segment, the horizontal line segment and existing horizontal line segment are merged to update the start point and end point of the existing horizontal line segment in the horizontal line segment list (step 170). During merging, the start point of the horizontal line segment is checked to determine if it is smaller than the start point of the existing horizontal line segment and the end point of the horizontal line segment is checked to determine if it is greater than the end point of the existing horizontal line segment. If the start point of the horizontal line segment is smaller, the start point of the horizontal line segment is used to overwrite the start point of the existing horizontal line segment in the horizontal line segment list. If the end point of the horizontal line segment is greater, the end point of the horizontal line segment is used to overwrite the end point of the existing horizontal line segment in the horizontal line segment list.
Once the horizontal line segment and existing horizontal line segment have been merged at step 170, the process proceeds to step 166 where a check is made to determine if any other classified table outline element exists that has not been selected.
As step 158, if the orientation value of the temporary line segment is less than one, the temporary line segment is transformed into a vertical line segment extending from the minimum y coordinate value to the maximum y coordinate value and having an x coordinate value equal to (xmax+xmin)/2 and denoted xavg (step 180). Thus, the vertical line segment has a start point (xavg, ymin) and an end point (xavg, ymax). The vertical line segment is then compared to a vertical line segment list to determine if the vertical line segment is located at substantially the same x coordinate position as a vertical line segment that already exists in the vertical line segment list (step 182). In this embodiment, the vertical line segment is deemed to be at the same x coordinate position as an existing vertical line segment if the x coordinate of the vertical line segment is within ten (10) pixels of the x coordinate of an existing vertical line segment. If the vertical line segment is deemed to be at the same x coordinate position as an existing vertical line segment, the start point and end point of the vertical line segment are compared with the start point and end point of that existing vertical line segment to determine if the vertical line segment and existing vertical line segment overlap. The vertical line segment and the existing vertical line segment are deemed to overlap if one of the following conditions holds true:
YEmin≦Ymin≦YEmax
YEmin≦Ymax≦YEmax
Ymin≦YEmin≦Ymin
Ymin≦YEmax≦Ymax
where:
YEmin and YEmax are the y coordinates of the start and end points respectively of the existing vertical line segment; and
Ymin and Ymax are the y coordinates of the start and end points respectively of the vertical line segment.
If the vertical line segment is determined not to overlap with existing vertical line segment, the vertical line segment is added to the vertical line segment list (step 184) and the process proceeds to step 166 where check is made to determine if any other classified table outline element exists that has not been selected.
At step 182, if the vertical line segment is determined to overlap with the existing vertical line segment, the vertical line segment and existing vertical line segment are merged to update the start point and end point of the existing vertical line segment in the vertical line segment list (step 190). During merging, the start point of the vertical line segment is checked to determine if it is smaller than the start point of the existing vertical line segment and the end point of the vertical line segment is checked to determine if it is greater than the end point of the existing vertical line segment. If the start point of the vertical line segment is smaller, the start point of the vertical line segment is used to overwrite the start point of the existing vertical line segment. If the end point of the vertical line segment is greater, the end point of the vertical line segment is used to overwrite the end point of the existing vertical line segment.
Once the vertical line segment and existing vertical line segment have been merged at step 190, the process proceeds to step 166 where a check is made to determine if any other classified table outline elements exists that has not been selected.
The steps performed during generation of the grid system (step 104 in
If the vertical and horizontal line segments of the selected combination intersect, the intersection point is calculated (step 210) and the calculated intersection point is added to an intersection point list (step 212). A check is then made to determine if the intersection point list comprises four (4) intersection points (step 214). If not, the process reverts back to step 206 and the next combination of vertical and horizontal lines segments is selected.
At step 208, if the vertical line segment and horizontal line segment of the selected combination do not intersect, a check is made to determine if the gap separating the horizontal and vertical line segments of the selected combinations is within a threshold, in this example one (1) pixel (step 216). If the gap separating the horizontal and vertical line segments is within the threshold, the horizontal and vertical line segments are deemed to intersect and the effective intersection point is calculated (step 218). The process then proceeds to step 212 where the calculated effective intersection point is added to the intersection point list.
At step 214, if the intersection point list comprises four intersection points, a grid is created having corners corresponding to the (x,y) coordinates of the four intersection points (step 230) and the grid is added to the grid system list (step 232). Following this, a check is made to determine if another pair of adjacent vertical line segments exists in the vertical line segment list that has not been selected (step 234). If another pair of non-selected adjacent vertical line segments exists in the vertical line segment list, the next pair of adjacent vertical line segments in the vertical line segment list is selected (step 236) and the process reverts back to step 204. In this manner, each pair of adjacent vertical line segments in the vertical line segment list is compared to the selected pair of adjacent horizontal line segments to determine the intersection points therebetween and thereby determine the coordinates of a row of horizontal grids in the grid system.
At step 234, if no pair of non-selected adjacent vertical line segments exists, a check is made to determine if another pair of adjacent horizontal line segments exists in the horizontal line segment list that has not been selected (step 240). If another pair of non-selected adjacent horizontal line segments exists in the horizontal line segment list, the next pair of adjacent horizontal line segments in the horizontal line segment list is selected (step 242) and the process reverts back to step 202. Steps 202 to 236 are then repeated for the next pair of adjacent horizontal line segments to determine the intersection points between the vertical line segments and the horizontal line segments and thereby determine the coordinates of the next row of horizontal grids in the grid system. At step 240, if no pair of non-selected adjacent horizontal line segments exists, the grid system generation process is deemed complete.
At step 216, if the gap separating the horizontal and vertical line segments of the selected combination is beyond the threshold, the process proceeds to step 240 to determine if another pair of adjacent horizontal line segments exists in the horizontal line segment list that has not been selected.
Once the table object has been created and table cell content placed in the cells of the table object, if appropriate, the table object can be further processed depending on the table cell content placed in the cells. For example, if numeric objects are placed in the cells of the table object, the table object can be treated as a spreadsheet and operations performed similar to other spreadsheet programs such as for example Excel™. In this case, placing a hand-drawn object resembling a “+” in the cell at the end of one of the cell rows or columns invokes a summation operation so that the sum of the numeric objects in that one row or column is calculated and displayed in the cell having the “+” object therein. Of course, other operations can be invoked by placing objects representing those operations in cells of the table object.
Although particular values for the thresholds described above are provided, those of skill in the art will appreciate that the threshold values may be changed to relax or tighten the process to suit the particular environment in which the hand-drawn table recognizing tool 16 is employed. Also, rather than calculating the run versus rise of each table outline element, other techniques can be employed to determine the orientation of the freeform lines. In addition, although the above embodiment describes freeform line objects as being recognized as the table outline elements, if desired, selected combinations of freeform line objects and shape objects may also be recognized as table outline elements.
If desired, when each table outline element is selected at step 150, the table outline element can be processed to make a determination as to whether the table outline element represents a table line or other non-table line such as handwriting. This can be achieved for example by analyzing maximum deviations of the line coordinates in given directions or by calculating derivatives. Alternatively, the user may be provided with the ability to designate non-table lines using an input gesture.
Although the hand-drawn table recognizing tool 16 has been described with specific reference to its implementation in the SMART Notebook™ application, those of skill in the art will appreciate that the hand-drawn table recognition tool may be used as a stand alone tool or implemented as a feature in other digital ink software environments to facilitate the creation of table objects from hand-drawn freeform lines.
The hand-drawn table recognition tool 16 may comprise program modules including but not limited to routines, programs, object components, data structures etc. and may be embodied as computer readable program code stored on a computer readable medium. The computer readable medium is any data storage device that can store data, which can thereafter be read by a computer system. Examples of computer readable medium include for example read-only memory, random-access memory, flash memory, CD-ROMs, magnetic tape, optical data storage devices and other storage media. The computer readable program code can also be distributed over a network including coupled computer systems so that the computer readable program code is stored and executed in a distributed fashion.
Although embodiments have been described with reference to the drawings, those of skill in the art will appreciate that variations and modifications may be made without departing from the spirit and scope thereof as defined by the appended claims.
Number | Name | Date | Kind |
---|---|---|---|
5392130 | Mahoney | Feb 1995 | A |
5448263 | Martin | Sep 1995 | A |
5805168 | Minakata | Sep 1998 | A |
6012073 | Arend et al. | Jan 2000 | A |
6141000 | Martin | Oct 2000 | A |
6337681 | Martin | Jan 2002 | B1 |
6415305 | Agrawal et al. | Jul 2002 | B1 |
6747636 | Martin | Jun 2004 | B2 |
6763148 | Sternberg et al. | Jul 2004 | B1 |
6803906 | Morrison et al. | Oct 2004 | B1 |
6952803 | Bloomberg et al. | Oct 2005 | B1 |
7106904 | Shima | Sep 2006 | B2 |
7184606 | Shindo et al. | Feb 2007 | B2 |
7222986 | Mah | May 2007 | B2 |
7236162 | Morrison et al. | Jun 2007 | B2 |
7250968 | Ito | Jul 2007 | B2 |
7274756 | Kuijk | Sep 2007 | B2 |
7583841 | Lin et al. | Sep 2009 | B2 |
7751623 | Simmons et al. | Jul 2010 | B1 |
7865018 | Abdulkader et al. | Jan 2011 | B2 |
8054971 | Weiner | Nov 2011 | B2 |
8150162 | Du et al. | Apr 2012 | B2 |
8201084 | Konno | Jun 2012 | B2 |
8385651 | Zhang et al. | Feb 2013 | B2 |
20010024520 | Yamaai | Sep 2001 | A1 |
20020107885 | Brooks et al. | Aug 2002 | A1 |
20020159639 | Shima | Oct 2002 | A1 |
20030095708 | Pittel | May 2003 | A1 |
20030113016 | Naoi et al. | Jun 2003 | A1 |
20030123732 | Miyazaki et al. | Jul 2003 | A1 |
20030174220 | Ito | Sep 2003 | A1 |
20030214491 | Keely et al. | Nov 2003 | A1 |
20040008883 | Shi et al. | Jan 2004 | A1 |
20050001824 | Yoshimura | Jan 2005 | A1 |
20050093845 | Brooks et al. | May 2005 | A1 |
20050094854 | Kim | May 2005 | A1 |
20050226535 | Ghosh | Oct 2005 | A1 |
20060126938 | Lee et al. | Jun 2006 | A1 |
20060132455 | Rimas-Ribikauskas et al. | Jun 2006 | A1 |
20060132456 | Anson | Jun 2006 | A1 |
20060132457 | Rimas-Ribikauskas et al. | Jun 2006 | A1 |
20060239560 | Sternby | Oct 2006 | A1 |
20070092140 | Handley | Apr 2007 | A1 |
20070097102 | Keely et al. | May 2007 | A1 |
20070140565 | Lin et al. | Jun 2007 | A1 |
20070140566 | Lin et al. | Jun 2007 | A1 |
20070186152 | Gurcan et al. | Aug 2007 | A1 |
20070195084 | Cheung et al. | Aug 2007 | A1 |
20080075368 | Kuzmin | Mar 2008 | A1 |
20080285069 | Yasuda | Nov 2008 | A1 |
20090245645 | Oct 2009 | A1 | |
20090245654 | Xing et al. | Oct 2009 | A1 |
20100123719 | May 2010 | A1 | |
20100178645 | Ieperen et al. | Jul 2010 | A1 |
20100201812 | McGibney et al. | Aug 2010 | A1 |
20100313155 | Willekes et al. | Dec 2010 | A1 |
20110199297 | Antonyuk et al. | Aug 2011 | A1 |
20110216015 | Edwards | Sep 2011 | A1 |
20120223920 | Brooks et al. | Sep 2012 | A1 |
20130044954 | Ashbrook | Feb 2013 | A1 |
Number | Date | Country |
---|---|---|
101452523 | Jun 2009 | CN |
101452523 | Jun 2009 | CN |
1 519 280 | Mar 2005 | EP |
Entry |
---|
Glenn et al. “Word 2000 in a Nutshell” Chapter 10 1st Edition, Aug. 2000, pp. 1-22. |
Supplementary European Search Report and Annex to the European Search Report for European Patent Application No. EP 09725195 with a date of completion of Feb. 13, 2012. |
Jul. 3, 2012 letter from China Sinda to Sim & McBurney with a summary in English of the Jun. 12, 2012 First Office Action for Chinese Patent Application No. 200980111413.4 and a copy of the Office Action in Chinese. |
Number | Date | Country | |
---|---|---|---|
20090245645 A1 | Oct 2009 | US |