JP2000148907A - Handwritten diagram recognition device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a handwritten diagram recognition device which eliminates the drop of a recognition rate due to a pen fluctuation generated when an input is performed and due to the quantization error of a tablet, is not influ enced by a strode order and also does not make a big difference from an input ted diagram. SOLUTION: This on-line handwritten diagram recognition device consists of a detecting means which detects a pen locus written on a display incorporated tablet, a processing means 3 which produces time sequentially recorded stroke data and performs recognition processing by using the stroke data, a displaying means 7 which shows processed results made by the means 3 on the tablet 1 and a selecting means which selects recognition results. In such a case, the handwritten diagram recognition device has a discriminating means which discriminates between a display request for the next candidate and the next diagram input by tapping the display screen with a pen when the recognition results are shown on the screen.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a handwritten figure recognition apparatus, and more particularly, to a recognition apparatus that performs recognition without using a figure dictionary, displays a figure closer to an input, and recognizes a target object based on the size of the input figure. The present invention relates to a handwritten figure recognition device suitable for actual use with limited figures. For example, the present invention is suitable for a word processor, a personal computer, or the like that identifies a graphic input from a tablet using a pen and displays the result.

[0002]

2. Description of the Related Art As a method of recognizing a figure entered on a tablet, an input figure stroke is converted from a feature of the shape into a set of line element string data of an arc or a straight line, and is previously converted into similar line element string data. Recognition was performed by matching with defined registration figures. When exchanging data with line segment element data, there is a method that introduces ambiguity, but registration of a figure is required. There is also a method of examining the distribution of the angle histogram in the writing direction, determining the line segment based on the distribution state, and identifying the number of sides and the like. In either method, uniform processing is performed regardless of the size of the input graphic and the length of the stroke. Further, as a method of selecting a candidate of a figure recognition result, there is a method of indicating the following candidate. In other words, this is a method in which the next candidate appears when a predetermined area is touched with a pen. Alternatively, tapping (lightly touching the tablet with a pen) brings up a list of candidates, and you can select from them. Further, there is a method of canceling an input figure when the input figure is canceled by touching a pen in a range determined to be canceled.

As a known document describing the prior art according to the present invention, for example, a “pattern recognition method” is proposed in Japanese Patent Application Laid-Open No. 63-82584. This publication is suitable for an apparatus for recognizing a handwritten character figure with ambiguity. Even if the stroke shape has ambiguity, the input stroke is formed into a shape in order to enable correct matching with the registered figure. Separation into distinctly arcuate parts, distinctly straight parts, and arcs or straight lines or ambiguous parts, and changing the interpretation of the ambiguous parts as arcs or straight lines, changing the arc or straight line of the input stroke A plurality of structural patterns based on the segment elements are created, and the degree of difference is obtained by comparing the structure of the structural pattern with the structure pattern of each dictionary graphic, and checking the likelihood of arcs or straight lines of the line segment elements. It is to choose one.

[0004]

As described above, there are the following problems in making a low-priced apparatus or mounting it in a personal device in the conventional graphic recognition apparatus. That is, in the method of converting into a line segment element sequence, in order to perform accurate recognition, it is necessary to track the coordinate sequence in detail for all the entered strokes and determine the line and arc elements. Also, it takes a long processing time to determine the element sequence, and a personal device such as a CPU (Central Processing Unit) having a low processing capability cannot perform processing in real time. There is a cost issue with increasing throughput. Further, in the method using the distribution of angles and histograms, it is not possible to correctly recognize an absolutely small figure because of a quantization error caused by the resolution of the tablet. In addition, increasing the resolution of a tablet has problems in manufacturing and costs, and cannot be adopted for personal devices.

Furthermore, if a person writes a figure on a tablet with a pen by handwriting, the pen fluctuates. This fluctuation may adversely affect recognition when the figure is small. For example, an extremely small circle is erroneously recognized as a polygon having a large number of sides, such as a regular hexagon. In general, when drawing a figure, if the size of the figure is smaller than a certain value, the drawing is usually accompanied by deformation, and the drawing cannot be performed correctly. On the other hand, in the case of the display-integrated tablet, the graphics may be deformed and displayed due to the low resolution on the display side, and a third party may not be able to determine. For example, if a regular hexagon with a maximum width of 5 mm for a millimeter three figure is to be entered, a handwriting on the tablet is often viewed as a circle by a third party. In such a case, in the above-described method, a hexagon or the like is output as a recognition result, and a mismatch may occur with the written handwriting.

In the prior art, the method of selecting a candidate involves a large amount of movement of a pen, and it takes time to input a figure with the pen.
In addition, a writer may inadvertently input a stroke into a tablet while drawing a graphic. In this case,
In the above method, erroneous recognition may occur. Thus, there is a problem of operability.

[0007] The present invention has been made in view of such circumstances, except for the fluctuation of the pen and the reduction of the recognition rate due to the quantization error of the tablet which occur at the time of input. It is an object of the present invention to provide a handwritten figure recognition device having a small difference between the two.

[0008]

In order to achieve the above object, the present invention provides: (1) detecting means for detecting the trajectory of a pen written on a display-integrated tablet, and stroke data recorded in time series. And an on-line handwritten figure recognizing device comprising: a processing unit for performing a recognition process using the stroke data; a display unit for displaying a processing result by the processing unit on a tablet; and a selection unit for selecting a recognition result. When the recognition result is displayed on a screen, the display device includes a determination unit that determines a request for displaying a next candidate and a next graphic input by tapping the display screen with a pen. Provided is a handwritten figure recognition device. (2) In a handwritten figure recognition device having a function of selecting a candidate, when a tablet on which a recognition result is displayed is tapped with a pen, the position of the tapped on the tablet and the position of the displayed figure are displayed. By the relationship
There is provided a handwritten figure recognition apparatus characterized by having a selection means for selecting a process such as selection of a candidate or cancellation of an input. (3) In an online handwritten figure recognition device having a function of selecting a candidate, when a tablet on which a recognition result is displayed is tapped with a pen, a tapped position on the tablet and a displayed figure are formed. The present invention provides a handwritten figure recognition apparatus characterized by having a selection means for selecting a stroke that has been erroneously recognized based on a relationship with a stroke position to be performed, and selecting a process such as selection of a candidate for only that portion. (4) detecting means for detecting the trajectory of the pen written on the tablet, processing means for creating stroke data recorded in time series, and performing recognition processing using the stroke data; An on-line handwritten graphic recognition device comprising display means for displaying a processing result on a tablet, characterized in that it has a restriction means for restricting a category of a graphic to be output as a recognition result according to the number of strokes input to the tablet. And a handwritten figure recognition device. (5) The determination means of (1) determines that a tap is made within the figure of the recognition result as a request for displaying the next candidate, and if the tap is made outside, the determination means of the next figure is input. Provided is a handwritten figure recognition device characterized by determining a request. (6) The handwriting figure recognition device is characterized in that the determination means of (1) deletes a figure as a recognition result when the display screen is tapped with a pen.

From the viewpoint of operability, even when selecting a candidate,
Eliminating complicated operations and shortening the input time improves the efficiency of graphic input with the pen. Thus, a small, high-performance, and easy-to-use graphic recognition device can be realized.

When the operator wants to correct the figure of the recognition result, he taps the display screen with a pen,
The figure of the recognition result is automatically deleted, and the selection of the next candidate is displayed.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a configuration diagram for explaining an embodiment of a handwritten graphic recognition device according to the present invention, and is a configuration diagram of an online graphic input device for inputting a graphic by turning a pen on a tablet. In the figure, 1 is a tablet, 2 is a stroke end determination unit, 3 is a stroke processing unit, 4 is a graphic end determination unit, 5 is a graphic recognition unit, 6 is a display coordinate generation unit, and 7 is a display unit. In addition, the numerical value of the following example is an example,
Appropriate settings can be made in consideration of the usage status of the graphic recognition device.

Embodiment 1 A handwritten figure recognition apparatus according to the present invention detects a trajectory of a pen written on a tablet, creates stroke data recorded in time series, performs recognition processing using the stroke data, The present invention relates to an online handwritten figure recognition device that displays results on a tablet. The data written by the writer on the display-integrated tablet 1 is converted into coordinate point sequence data represented by electric signals, and the data is transferred to the stroke processing unit 3. Next, when the stroke end determining unit 2 determines that the entry of the stroke is completed, the stroke processing unit 3 performs a stroke process. That is, every time a stroke is input, data processing is performed for each stroke, and the number of sides and arc-like portions included in the stroke are obtained. Next, when the end of the figure is determined by the figure end determining unit 4, the figure recognizing unit 5 is activated, and determines whether the input stroke is an open figure or a closed figure. That is, after the graphic input is completed, the features formed in all strokes are extracted, a closed graphic (a loop that is formed by all strokes, no intersection) is determined, the stroke is integrated with the features of each stroke, and the stroke order is determined. And recognize figures without depending on the number of strokes. The result of performing the recognition process is sent to the display coordinate generation unit 6, where it is calculated to the coordinates corresponding to the input position, and the result is sent to the display unit 7 using liquid crystal and displayed.

FIG. 2 is a flowchart for explaining the operation of the handwritten figure recognition apparatus according to the present invention. Less than,
Each step will be described in order. First, a stroke is input (Step 1), and the stroke processing unit removes the same point and the intermediate point in the coordinate point sequence from the data as preprocessing (Step 2). Next, it is determined whether the stroke is shorter than the threshold (Step 3). If the stroke is shorter than the threshold, the stroke is skipped (Step 4).
Return to 1. If the stroke is not shorter than the threshold, a histogram in the absolute direction is calculated, and the feature of the stroke is extracted (Step 5). Next, the figure end determination unit determines whether the figure input is completed (Step 6). If the figure input is not completed, the process returns to Step 1. If the figure input is completed, the absolute direction histogram is calculated, and the feature of the stroke is extracted (Step 7). Next, the figure recognizing unit determines whether or not the input stroke is a closed figure (Step 8). If the stroke is not a closed figure, an open figure recognition process is performed (Step 10), and the feature amount of the figure is calculated. (Step 1
1). If the figure is a closed figure in Step 8, the closed figure recognition processing is performed (Step 9), and the feature amount of the figure is calculated (Step 11) and displayed on the tablet. (Step 1
2).

FIGS. 3A to 3D are diagrams showing examples of stroke processing, showing examples of features for each stroke. FIG. 7A shows an example of an input stroke, and black circles indicate points after the row processing. FIG. 2B is a diagram showing an example of a direction code, FIG. 2C is a diagram showing an angle histogram (windowed column), and FIG. 2D is a diagram showing an angle histogram (after windowing).

In the stroke processing unit 3, the same point and the intermediate point in the coordinate point sequence are removed from the data as preprocessing. An intermediate point here is a point on the line when the coordinates of the preceding and following points are connected by a straight line, or when there is no point on the line,
It means the point where it crosses for the first time when it goes in parallel with the X axis or Y axis. The portion indicated by the black circle in FIG. After that, for all points except the end point of the continuous coordinate sequence in the corresponding stroke in the writing direction, the direction and length from that point to the next point are obtained. As the direction, as shown in FIG. 3B, a value obtained by quantizing 360 degrees in all directions in 32 directions is used. This is shown in FIG.
As shown in (1), an angle histogram obtained by summing the lengths is obtained. Next, as shown in FIG. 3 (d), after performing windowing of width 3 on this, an angle having the maximum value of the histogram,
A peak, that is, an angle having a maximum value is obtained. When the maximum value of the peak is extremely smaller than the maximum value of the histogram, this peak is ignored. At the same time, the total of the circumscribed frame and the length of the stroke is totaled. The above processing is repeated for each stroke input.

When the graphic end judging section 4 judges the end of the graphic, the graphic recognizing section 5 is started, and the input stroke is first determined as to whether it is a closed graphic or an open graphic as follows. That is, when the input graphic is one image, the distance between the start point and the end point is 3 mm or less, and when the coordinates indicating the stroke are connected by a straight line and there is no intersection in the middle, the figure is a closed graphic. If the input figure or two or more strokes, the end points of each stroke are connected, and if all the strokes are included and the distance between all the end points is 3 mm or less, it is determined that the figure is a closed figure.

When it is determined that the figure is a closed figure, if it is a closed figure, it is obvious that if it is formed only by a straight line, it is an N-gon (N> 3). Estimate the number of sides by the number of
Detailed identification is performed for each number of the sides. First, vertices of the estimated number of sides are extracted. The vertex formed by the end points of a plurality of strokes adopts the middle point. If a vertex exists in the stroke, the stroke is first divided so that the length is equal to the number of sides estimated to be included in the target stroke. Between the start point and the end point of the stroke, the above dividing points are arranged in the stroke writing direction, and the following processing is performed. First, the first point of the division point is changed to a point farthest from a straight line connecting the preceding and following points. Similar processing is repeated for the second and subsequent points. By repeating this process until all the division points are not changed, the position of the estimated vertex is detected. Subsequently, after the vertices are extracted, the linearity defined by the following equation, linearity = (linear distance between vertices) / (road ride on stroke) Is determined based on the variance from the straight line direction. The threshold for making the determination is arbitrarily set according to the performance of the tablet and the like. As described above, it is possible to discriminate between a polygon and a sector.

If it is determined that the figure is an open figure, a line segment approximation is performed for each stroke, and a recognition result including a straight line or a plurality of line segments and an arc is output as a recognition result. The result of performing the recognition processing as described above is sent to the display coordinate generation unit 6, and the coordinates of the display device corresponding to the input position are calculated. The result is sent to a display unit 7 using liquid crystal and displayed. Thereafter, the same processing is repeated until the online graphic recognition ends.

As described above, the open figure and the closed figure are determined, and if the figure is a closed figure, an N polygon (N
> 3), the stroke straight line portion is detected, and the input figure is recognized based on its position and length. Also, when extracting the size and rotation angle of the input graphic, in the case of an N-gon, the rotation angle obtained by stroke recognition is converted into a fixed rotation direction (for example, clockwise) at the time of stroke entry to convert each figure. It is the angle of the side. In this case, the recognition and shaping of the figure is performed using the features such as the absolute angle of the other side based on the position of the start point and the end point, the length of the stroke and the length of the side with respect to the size of the figure. If it is an open figure, approximation is made to a line segment and a curve for each section.

In recognizing a closed figure, a stroke entry method is classified using an angle histogram, and then feature points such as estimated vertices are detected, and detailed discrimination is performed using the feature points. Further, when the size of the graphic is smaller than a certain size, the recognition target is narrowed, thereby preventing erroneously recognizing a small written graphic. Also, in terms of operability, a part is provided to determine whether the next pen input after displaying the result on the tablet is the next input or a candidate selection, and the next step such as selecting a candidate based on the result. Select the process. When this process is selected, the relationship with the display position of the immediately preceding result can be used.

Embodiment 2 In the online graphic recognition device, the minimum value and the maximum value in the X-axis direction and the Y-axis direction are obtained for the coordinate point sequence of all the strokes input from the tablet, respectively. Each width is calculated, and the larger width is set as the maximum width of the figure. If this maximum width is smaller than 4 mm, only "circles" or "straight lines" will be recognized.
If it is larger than 7mm and smaller than 7mm,
Add “regular triangle” and “square”. If it is larger than 7 mm and smaller than 10 mm, “Right triangle” and “Rectangle” are added.

Embodiment 3 After performing stroke preprocessing on a stroke coordinate point sequence input from a tablet, a distance between a start point and an end point is obtained. The stroke determination unit compares the lengths of the strokes before performing the recognition process, and determines the stroke having the previous maximum length. If there is a stroke having a length of 1/10 or less based on the length of this stroke, it is determined that the writer has made an erroneous input, and this stroke is deleted from the stroke buffer. After performing the above processing, the recognition processing is executed.

Embodiment 4 FIG. 4 is a flowchart for explaining the online graphic recognition processing. The description will be made in order according to each step. In the input device, online graphic recognition is started by an operation using a keyboard or the like. When entering the online graphic recognition mode, first, it enters the state of waiting for pen input (step
1). In inputting a figure, the user inputs a figure using a pen (step 2). After input is completed, perform figure recognition processing (step
3) Display the first candidate of the recognition result (step
4). After the display, it waits for pen input (step 5). When any input is made with the pen, it is determined whether the input is a tap (step 6). If it is a tap, it is interpreted that the next candidate has been requested, the figure of the current result is deleted, and the next candidate is displayed (step 7), and a state of waiting for input (step 7)
5). If it is not a tap, it is determined that the input of the next figure has been started, and a figure input process is performed (step 2). Thereafter, the same processing is repeated until the online graphic recognition ends.

Embodiment 5 FIG. 5 is another flowchart for explaining the online graphic recognition processing. Hereinafter, the steps will be sequentially described. When you enter online shape recognition mode,
The process enters a state of waiting for pen input (step 1). In inputting a figure, the user inputs a figure using a pen (step 2). After the input is completed, a figure recognition process is performed (step 3), and the first candidate of the recognition result is displayed (step 4). After the display, it waits for pen input (step 5). When something is input with the pen,
It is determined whether it is a tap (step 6). If it is determined that the tap is a tap, the figure position of the displayed recognition result is compared with the tapped position (step 7), and the tapped position is inside or close to the figure (for example,
If it is within 3 mm of the straight line distance), the figure of the current result of the recognition result is deleted, and the next candidate is displayed (step 8). If the figure is outside the figure, it is determined that cancel has been input, and the figure as the recognition result is erased (step 9), and the state is awaiting input (step 1). Thereafter, the same processing is repeated until the online graphic recognition ends.

Embodiment 6 This will be described with reference to the flowchart shown in FIG. Note that st
The processing from ep1 to step6 is the same as in the fifth embodiment. step
If it is determined in step 6 that the tap is a tap, the stroke position forming the displayed graphic of the recognition result is compared with the tapped position (step 7), and the tapped position is within or close to the figure. (For example, within 3 mm in a straight line distance)
If so, the figure of the current recognition result is erased, and the next candidate having changed the recognition result of the tapped stroke is displayed (step 7). If it is outside the shape, it is determined that cancel has been entered, and the recognized shape is erased (step
9), and enters an input waiting state (step 1). Thereafter, the same processing is repeated until the online graphic recognition ends.

Embodiment 7 Recognition is performed in the same manner as in Embodiment 1. In the figure recognition unit 5,
When it is determined that the figure is a closed figure, polygon recognition is performed on the assumption that one stroke forms at least one side.

[0027]

As is clear from the above description, it is possible to easily select a candidate and to select re-entry.

When the operator wants to correct the figure of the recognition result, he or she taps the display screen with a pen,
The figure as a recognition result is automatically erased, and the selection of the next candidate is displayed, so that the operation is simplified.

[Brief description of the drawings]

FIG. 1 is a configuration diagram for explaining an embodiment of a handwritten figure recognition device according to the present invention.

FIG. 2 is a flowchart illustrating an operation of handwriting recognition according to the present invention.

FIG. 3 is a diagram showing an example of a feature for each stroke in the present invention.

FIG. 4 is a flowchart illustrating an online graphic recognition process according to the present invention.

FIG. 5 is another flowchart for explaining the online graphic recognition processing in the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Tablet, 2 ... Stroke end determination part, 3 ... Stroke processing part, 4 ... Graphic end determination part, 5 ... Graphic recognition part,
6: display coordinate generation unit, 7: display unit.

Claims (6)

[Claims] 1. A detecting means for detecting a trajectory of a pen written on a display-integrated tablet, a processing means for creating stroke data recorded in time series and performing a recognition process using the stroke data; In an online handwritten figure recognizing device comprising a display unit for displaying a processing result by a processing unit on a tablet and a selecting unit for selecting a recognition result, when the recognition result is displayed on a screen, a display screen is displayed with a pen. A handwritten figure recognizing apparatus, comprising: a tapping means for discriminating between a next candidate display request and a next figure input. 2. A handwritten figure recognizing device having a function of selecting a candidate, wherein when a tablet on which a recognition result is displayed is tapped with a pen, a tapped position on the tablet, a position of the displayed figure, A handwriting figure recognition device characterized by having a selection means for selecting a process such as selection of a candidate or cancellation of an input in accordance with the relationship. 3. An on-line handwritten figure recognition device having a function of selecting a candidate, wherein when a tablet on which a recognition result is displayed is tapped with a pen, a tapped position on the tablet and a displayed figure are formed. A handwritten figure recognizing device, comprising: a selection unit that selects a stroke that is erroneously recognized based on a relationship with a stroke position to be performed, and selects a process such as selection of a candidate for only that portion. 4. A detecting means for detecting a trajectory of a pen written on a tablet, a processing means for creating stroke data recorded in time series and performing a recognition process using the stroke data, and An on-line handwritten figure recognition apparatus comprising: a display unit for displaying a processing result on a tablet, characterized in that there is provided a limiting unit for limiting a category of a figure to be output as a recognition result according to the number of strokes input to the tablet. Handwritten figure recognition device. 5. The method according to claim 1, wherein the determining unit determines that the next candidate is requested to be displayed when the tap is performed inside the figure of the recognition result.
2. The handwritten figure recognition apparatus according to claim 1, wherein when the tap is made outside, it is determined that the next figure input request is made. 6. The handwritten figure recognition apparatus according to claim 1, wherein said determination means erases a figure as a recognition result when the display screen is tapped with a pen.