JPH09185726A - Ruled line removing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain high speed and accuracy in processing at the time of realizing the removing of a ruled line while preventing the chipping of the stroke of a character at a slip picture. SOLUTION: A ruled line position storing memory 5 stores ruled line position information of a ruled line in the slip picture within a picture memory 1. A ruled line erasing means 2 erases the ruled line in the slip picture within the picture memory 1 through the use of this ruled line position information and stores a coordinate value showing the cutting position of a character component generated at the interference place of the character and the ruled line in the process of erasing the ruled line in a cutting position storing memory 3. A character chipping restoring means 4 analyzes the graphic structure of the slip picture in the neighborhood of this cutting position based on the reference of ruled line position information within the ruled line position storing memory 5, the coordinate value showing the cutting position within cutting position storing memory 3 and the slip picture within the picture memory 1, estimates the interference pattern between the character and the ruled line generated at this cutting position by the analyzing, and restores the chipping part of the character based on the estimate.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a form in which characters formed of a small number of strokes (characters such as numbers (arithmetic numerals), alphabets, katakana, and hiragana; numbers are the most common) are filled in. Regarding the ruled line removal method that removes ruled lines from the image (form image) of (forms with ruled lines), it is possible to remove ruled lines without losing strokes of characters (characters that touch or intersect) that overlap the ruled lines. The ruled line removal method for doing.

[0002]

2. Description of the Related Art Conventionally, this type of ruled line removal method has been
CR (Optical Character Read
er), it is used to remove a ruled line that hinders character recognition before the character recognition processing.

The conventional ruled line removing method of this kind stores an image memory in which form images are stored in advance and a ruled line position storage in which information indicating the positions of ruled lines on the form image (ruled line position information) is stored in advance. It is composed of a memory, an image analysis means, and a ruled line erasing means.

The image analysis means, for example, traces the contour of the figure on the form image, compares and analyzes the ruled line position information and the position information of the contour tracking result, and determines whether the ruled line and the character overlap. .

That is, the contour inside the ruled line can be specified by comparing the ruled line position information with the position information of each contour detected by the contour tracing (position information of the contour tracing result). There is no interference between the ruled line and the characters written inside it if the ruled line position indicated by the ruled line position information read out from the storage memory does not largely deviate and if it has a simple rectangular shape. You can judge. On the other hand, if this condition (the condition that the contour has a simple rectangular shape without being greatly displaced from the position of the ruled line) is not satisfied and it is determined that there is interference, further analysis is performed and the characters overlap. The section of the ruled line that has not been specified is specified.

Further, the image analysis means realizes the removal of the ruled line while avoiding the loss of the stroke of the character by erasing the section specified as the section of the ruled line which does not overlap with the character.

[0007]

In the above-described conventional ruled line removal method, it is necessary to perform contour tracing and analysis based on the contour tracing result, so that it is possible to realize the ruled line elimination while avoiding the loss of the stroke of the character. There is a problem that the processing amount (calculation amount of the computer system) becomes large, and the calculation cost (the amount of money for constructing a large-scale computer system, the processing time and the number of processing required for the ruled line removal processing, etc.) becomes large. .

In addition, in the conventional ruled line removal method, if there is a break in the ruled line component on the image, it is necessary to incorporate a pre-process for interpolating this break. There is also a problem that the calculation cost required for the (pre-processing of step 1) increases.

Further, in the conventional ruled line removing method, when the characters and the ruled lines interfere with each other, there is a risk that the characters may be separated with some of the ruled line components remaining. There is a problem in that the accuracy of removal is poor.

In view of the above points, an object of the present invention is to realize removal of ruled lines while avoiding loss of strokes of characters in a form image having characters (characters formed by a small number of strokes) and ruled lines. Another object of the present invention is to provide a ruled line removal method that requires only a small amount of calculation (thus enabling high-speed processing) and realizes highly accurate processing.

[0011]

The ruled line removing method of the present invention includes an image memory that stores in advance a form image having characters and ruled lines formed by a small number of strokes.
A ruled line position storage memory that stores ruled line position information of ruled lines in the form image in the image memory, and a cut position save memory that saves coordinate values of the cut positions of characters generated at the points of interference between characters and ruled lines. The ruled line position information read from the ruled line position storage memory is used to erase the ruled lines in the form image in the image memory, and the character components generated at the points of interference between the characters and the ruled lines in the process of deleting the ruled lines are deleted. Ruled line erasing means for storing coordinate values indicating a cutting position in the cutting position saving memory, ruled line position information in the ruled line position saving memory, coordinate values indicating a cutting position in the cutting position saving memory, and in the image memory Based on the reference of the form image, the graphic structure of the form image near the cutting position is analyzed, the interference pattern between the character and the ruled line occurring at the cutting position is estimated by the analysis, and based on the estimation. There and a character defect restoring means for restoring the lost portion of the character.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the ruled line removing method of the present invention.

In the ruled line removing method of this embodiment, the image memory 1 in which the image of the form (form image) is stored in advance, and the ruled line erasing means 2 for erasing the ruled line on the form image stored in the image memory 1. And a cutting position storage memory 3 for storing the coordinate value of the position (cutting position) of the position (cutting position) interrupted by the ruled line erasure for a figure (a figure forming a stroke of a character, etc.) that touches or intersects the ruled line. A character loss restoring means 4 for restoring a character loss caused by erasing ruled lines, and a ruled line position storage memory 5 that stores ruled line position information.

It should be noted that some of the constituent elements in FIG. 1 will be described in the following items.

Image memory 1 The image memory 1 stores in advance a form which is a binary two-dimensional digital image obtained by capturing a form in which numbers such as amounts are entered by an image scanner or the like. Here, the pixel value of the part without the form is "1", and the background part of the form (the part where the graphic pattern such as characters and ruled lines does not exist)
Has a pixel value of "0", and the pixel value of a figure pattern portion such as characters and ruled lines in the form is "1". As described above, in the present embodiment, the case where the character that interferes with the ruled line is a number will be described. However, even for a character other than a number, a character having a small number of strokes (alphabet, katakana, hiragana, etc.) can be used. For example, the present invention is effective.

Ruled Line Position Saving Memory 5 The ruled line position saving memory 5 indicates the result of detecting the position of the ruled line in the form image stored in the image memory 1 by means or device other than the constituent elements of the present invention. Information is stored in advance.

As shown in FIG. 13, the ruled line position information in the ruled line position storage memory 5 includes variables A, B, C, D, and E.
Expressed as a set of variable values for.

Here, the variable A is a code for identifying whether the ruled line is a horizontal ruled line or a vertical ruled line.

Variables B and C are expressed when the ruled line is expressed by the following linear expression: y = ax + b (horizontal ruled line) ... (Formula 1) or x = cy + d (vertical ruled line) ... (Formula 2) Value of the coefficient a or c of the first-order term (variable B) and the value of the constant term b or d (variable C).

Variables D and E are the ranges (x ₀ ≤x≤x) that can be taken by x in the case of the horizontal ruled line in the expressions 1 and 2.
_M ), which is the range of y that can be taken by vertical ruled lines (y
₀ ≦ y ≦ y _N ). That is, the variable D is x ₀
Or y ₀ , and the variable E indicates x _M or y _N.

Character Loss Restoring Unit 4 As shown in FIG. 4, the character loss restoring unit 4 includes an intersection point coordinate calculating unit 401, an intersection point coordinate storing unit 402, a cutting position information discarding unit 403, and a section setting unit 404. Feature extraction means 405, first determination means 406, second determination means 407, restoration means 408, cutting position information selection means 4
09 and a local cutting position information storage means 410.

FIG. 2 is a flow chart showing the processing of the ruled line removal method of this embodiment (processing relating to the overall operation). This processing includes ruled line reading step 201 and ruled line erasing step 2
02, a character loss restoration step 203, and an end determination step 204.

FIG. 3 is a flow chart showing the processing of the ruled line erasing means 2. This process is performed by the provisional ruled line boundary detection step 301.
An average line width calculation step 302, a ruled line boundary correction step 303, and a ruled line boundary undetected point interpolation step 30
4 and a ruled line component erasing step 305.

FIG. 4 is a block diagram showing the configuration of the character loss restoring means 4 as described above.

FIG. 5 is a flow chart showing the processing of the character loss restoring means 4. This process is performed by the intersection detection step 501.
, Intersection point cutting position coordinate value rejection step 502, section setting step 503, and local feature extraction step 504
And a first determination step 505 and a second determination step 50
6 and a restoration step 507.

FIG. 6 is a diagram for explaining the operation of the ruled line erasing means 2.

7 to 12 are diagrams for explaining the specific operation of the ruled line removing method of this embodiment.

FIG. 13 is a diagram showing the structure (data structure) of the ruled line position information in the ruled line position storage memory 5, as described above.

Next, the operation of the ruled line removing method of the present embodiment having such a configuration will be described.

First, the overall operation of the ruled line removing method of this embodiment will be described with reference to FIGS. 1 and 2.

The ruled line erasing means 2 first reads ruled line position information for one ruled line indicated by the values of variables A, B, C, D and E in FIG. 13 from the ruled line position storage memory 5 ( Step 201).

Next, the ruled line erasing means 2 executes step 201.
The ruled line in the form image in the image memory 1 corresponding to the ruled line position information read in step 3 is erased (step 202). That is, the pixel value of the ruled line component in the binary two-dimensional digital image in the image memory 1 is converted from "1" to "0". In this process, the part of the character component that overlaps the ruled line is also erased, but at this time the location of the pixel area of pixel value "1" (one area where the pixels of pixel value "1" are continuous without interruption) The coordinate value (coordinate value indicating the cutting position) of (cutting point) is output and saved in the cutting position saving memory.

When the ruled line erasing means 2 has finished erasing the ruled lines, the character loss restoring means 4 is activated.

The character loss restoring means 4 reads the cutting position of the pixel area (cutting point) having a pixel value of "1" from the character cutting position storage memory 3 and utilizes the ruled line position information in the ruled line position saving memory 5. The shape of the character component before the loss is estimated by analyzing the graphic structure of the component near the corresponding coordinate value in the image memory 1, and the lost character is restored based on the estimation (step 203). The restoration result is written in the image memory 1.

When the above processing is completed, the ruled line erasing means 2
Determines whether processing has been completed for all ruled line position information in the ruled line position storage memory 5 (step 20).
4).

If the ruled line erasing means 2 determines in step 204 that "the processing has been completed for all ruled line position information in the ruled line position storage memory 5," the ruled line removal method of this embodiment completes the entire operation. Let

Further, the ruled line erasing means 2 executes step 204.
When it is determined that "processing has not been completed for all ruled line position information in the ruled line position storage memory 5", unprocessed ruled line position information is read from the ruled line position storage memory 5 and the above-mentioned ruled line position information is described above. Steps 201-
The same process as 203 is executed.

Next, detailed operations of the ruled line erasing means 2 and the character loss restoring means 4 will be described.

First, the operation of the ruled line erasing means 2 will be described with reference to FIGS. 3 and 6.

The ruled line erasing means 2 has the following (1) to (5)
Is performed.

(1) First, the boundary of a temporary ruled line is detected as follows (step 301).

First, the ruled line position information for one ruled line is read from the ruled line position storage memory 5. Let the read ruled line be L. After reading the ruled line position information, ruled lines La and Lb that are parallel to the ruled line L and are separated from the ruled line L by Δs are generated. Here, the area sandwiched between these ruled lines La and Lb is referred to as area Z (see FIG. 6).

Now, a case where the ruled line L is a horizontal ruled line will be described (note that the process when the ruled line L is a vertical ruled line is the process in which the x axis and the y axis are exchanged in the process described below. ).

On the image in the area Z in the image memory 1, x =
From x ₀ (value of variable D in ruled line position saving memory 5) to x = x _M (value of variable E in ruled line position saving memory 5)
Up to 1 pixel unit, the y coordinate value is changed in the y direction within the range Z at each x coordinate value. The pixel value at each coordinate value (x, y) designated in this process is read.

Based on the read result, the coordinate value (x, y) closest to the ruled line L satisfying the following conditions is detected. Here, f (x, y) represents the pixel value of the pixel at the coordinate value (x, y). Condition: f (x, y) = 1 and f (x, y−1) = 0 (Equation 3) or f (x, y) = 1 and f (x, y + 1) = 0 ... (Equation 4)

When the coordinate value closest to the ruled line L satisfying the expression 3 is detected in the area Z, the value of y is increased by 1 pixel from the pixel having the coordinate value as the starting point, and the coordinate satisfying the expression 4 is obtained. Detect the value. On the contrary, when the coordinate value closest to the ruled line L satisfying the expression 4 is detected, the value of y is decreased by one pixel starting from the pixel having the coordinate value, and the coordinate value satisfying the expression 3 is detected.

The position of the pixel having the coordinate value (x, y) satisfying the expression 3 detected as described above indicates the upper end position of the temporary ruled line. Further, the position of the pixel having the coordinate value (x, y) satisfying the expression 4 detected as described above indicates the lower end position of the temporary ruled line. Each coordinate value is stored in the array shown in and provided below in the ruled line erasing means 2 as follows.

The horizontal ruled upper coordinate storage _{array: KUPP = {(x 0,} y u0), (x 1, y u1),
_{(X 2, y u2),} ..., (x M, y uM)}

Horizontal ruled line lower end coordinate value storage array: KLOW = {(x ₀ , y _w0 ), (x ₁ , y _w1 ),
(X ₂ , y _w2 ), ..., (x _M , y _wM )}

Where x ₀ , x ₁ , x ₂ , ..., X _M are
The x-coordinate value of the range in which the horizontal ruled line of interest exists is saved for each pixel.

[0052] In _{addition, y u0, y u1, y} u2, ..., y uM , the corresponding x-coordinate value _{x 0, x 1, x 2} , ..., the upper end of the y-coordinate value of the lateral borders in the x _M store It was done. If some of the upper end y-coordinate values of the horizontal ruled line cannot be detected by the above processing, “−1” is set as the upper end y-coordinate value.
Is stored (“−1” is information indicating “not detected” and is not necessarily limited to the value “−1”).

Further, y _w0 , y _w1 , y _w2 , ..., Y _wM are
The lower end y coordinate value of the horizontal ruled line at the corresponding x coordinate value x ₀ , x ₁ , x ₂ , ..., X _M is stored. In addition, when some of the lower end y coordinate values of the horizontal ruled line cannot be detected by the above processing, the lower end y coordinate value is set to "-".
Save 1 ”.

As described above, the boundaries of the temporary ruled lines (in this case, the upper and lower ends of the horizontal ruled lines) are detected.

(2) Based on the boundaries of the provisional ruled lines detected in step 301, the average line width and the allowable line width are calculated as follows (step 302).

First, an array KUP for storing the horizontal coordinate upper end coordinate value
The difference between the y coordinates at the same x coordinate value x _m is calculated between P and the horizontal ruled line lower end coordinate value storage array KLOW, and the distance between the upper and lower ends at the x coordinate value x _m , that is, the line width d.
_{Find uwm} . However, m takes the value of 0, 1, 2, ..., M.

From this calculation result, the average line width d is calculated by the following equation. d = ( _{m = 0} ^M Σd _uwm ) / (M + 1-M ′) (Equation 5)

Note that d in the equation (5) represents the total sum of m = 0 to m = M for d _uwm . Where y
d _uwm where _um = -1 or / and y _wm = -1
Is not added. _M'represents the number in that case (when y _um = -1 or / and y _wm = -1).

Further, d _limit, which is a value obtained by adding a predetermined minute amount α to the average line width d, is obtained as the allowable line width.

(3) Horizontal ruled line upper end coordinate value storage array KU
In the PP and horizontal ruled line bottom coordinate value storage array KLOW, the m-th element that _satisfies "d _uwm > d _limit " (K
Array data of UPP and KLOW start from the 0th element) and the y coordinate values (y _um and y _wm ) are both “−1”.
Change to That is, the boundary of the ruled line of interest is corrected based on the calculation result of step 302 (step 3
03).

(4) In the following manner, a ruled line boundary undetected portion is obtained, and the ruled line boundary undetected portion is interpolated (step 304).

First, an array KUP for storing the upper coordinate value of the horizontal ruled line
In P and the horizontal ruled line lower end coordinate value storage array KLOW, an element having a y coordinate value of "-1" is searched. Such an element corresponds to a position where a boundary between the ruled line and the background is not found (a ruled line boundary undetected position) on the form image in the image memory 1.

Next, an array KUP for storing the horizontal coordinate upper end coordinate value
Interpolation is performed for the undetected part of the ruled line boundary on each array data of P and the horizontal ruled line lower end coordinate value storage array KLOW. That is, in a series of continuous x-coordinate values, the ruled line boundary undetected portion is interpolated by the detection successful point around the ruled line boundary undetected portion. As a result, the position of the ruled line of interest is determined (the horizontal ruled line upper end coordinate value storage array KUPP and the horizontal ruled line lower end coordinate value storage array KLOW.
Final generation of is completed).

In this case, the x-coordinate value and the y-coordinate value at both ends of the ruled line boundary undetected portion are output to the cutting position storage memory 3 as the coordinate value of the position (cutting position) of the cutting position.

(5) Finally, a horizontal ruled line upper end coordinate value storage array KU generated by the processing up to step 304
PP and horizontal ruled line lower end coordinate value storage array KLOW x _m
In each of the coordinate values (m = 0 to M), to erase the pixels (pixel border component) in the section sandwiched between the position indicated by the coordinate values of the position and y _wm represented by the coordinate values of y _um ( Step 305). That is, the pixel values of those pixels are converted from “1” to “0” (the pixel value is originally “0”).
The pixel that is is left as "0").

Secondly, the operation of the character loss restoring means 4 will be described with reference to FIGS.

The character loss restoring means 4 is activated after the processing of the ruled line erasing means 2 for the ruled line of interest (the ruled line of interest as the processing target) is completed, and performs the following processes (1) to (8).

(1) The intersection point coordinate calculating means 401 in the character loss restoring means 4 reads the ruled line position information of the ruled line of interest and the ruled line position information of the ruled line orthogonal to this ruled line from the ruled line position storage memory 5. Further, the intersection coordinate calculation means 401
Calculates (detects) the coordinate values of the intersections of these ruled lines based on these ruled line position information, and the intersection point coordinate storage means 402
The coordinate value (coordinate value of ruled line intersection) is output to (step 501). The intersection coordinate storage means 402 stores the ruled line intersection coordinate values.

(2) The cutting position information rejection means 403 is
Using the ruled line intersection point coordinate value stored in the intersection point coordinate storage means 402, the coordinate value of the cutting point generated at the intersection point is detected from the coordinate values of the cutting point stored in the cutting point storage memory 3 Is rejected (step 502).

(3) The section setting means 404 reads the ruled line intersection point coordinate values in the intersection point coordinate storage means 402, extracts one of the sections sandwiched by consecutive intersection points, and several pixels at both ends of the section. Set the widened section (Step 50)
3). As described above, in the present embodiment, in the recognition of one section (area for one character), when the horizontal ruled line is the target ruled line, the recognition is performed based on one section divided by the vertical ruled line (vertical direction). If the ruled line is the target ruled line, the recognition is performed based on one section delimited by the horizontal ruled line). However, other recognition methods, such as designating a section in advance, are possible.

(4) The feature extracting means 405 extracts the following local feature groups (step 504). However, for convenience of description, the description is limited to the case where the horizontal ruled line is the target ruled line.

Now, the cutting position read from the cutting position storage memory 3 will be referred to as follows. ○ Cutting point {ai}, i = 0, generated at the top of the horizontal ruled line
1, ..., N _upp -1 ○ Cutting point {bj}, j = 0, generated at the bottom of the horizontal ruled line
1, ..., N _low -1

However, N _upp is the number of cut points formed above the ruled line, and N _low is the number of cut points generated below the ruled line. It is assumed that the coordinate values are arranged in order from the smallest X coordinate value to the largest X coordinate value. In addition, the coordinate value of the left end of each cutting point is (x _aLi , _yaLi ) or (x _bLj ,
y _bLj ), the rightmost coordinate value is (x _aRi , y _aRi ), or (x _bRj , y _bRj ), and the central coordinate value is (x _bRj ).
_aMi , _yaMi ) or ( _xbMj , _ybMj ).

The feature extraction means 405 extracts the following local features from these cut points.

Local Feature 1: Target Section (Step 50
(Distance set in 3), the distance difference between ai and bj in the x-axis direction: d _ij = | x _aMi −x _bMj | However, it is extracted only when N _upp > 0 and N _low > 0.

Local feature 2: Direction of components such as characters continuous at each cut point: θ _ai , θ _bj (0 ≦ θ _ai , θ _bj <π)

Local feature 3: Height of components such as characters continuous at each cut point (distance extending from ruled line end): h _ai , h _bj

(5) The first determination means 406 is activated after the above-mentioned local feature extraction processing is completed, and the feature extraction means 4
05, the feature amounts of the local features 1 to 3 are received, the coordinate values of the cut points in the target section are received from the cut position storage memory 3, and the cut points on the upper part of the ruled line corresponding to one-to-one (x coordinate values are the closest) The following judgment (first judgment consisting of judgment and judgment) is performed on the set of ai and the cut portion bj under the ruled line (step 505). Note that d
_threshold1 , d _threshold2 , and θ _threshold are predetermined thresholds, and d _threshold1 <d _threshold2 .

Judgment: If d _ij <d _threshold1 , ai
And bj are determined to be cutting points regarding the interference pattern of “intersection 1” (see FIG. 7).

Judgment: d _ij <d _threshold2 and | θ
_{If i} −θ _j | <θ _threshold , it is determined that ai and bj are cutting points regarding the interference pattern of “intersection 1”.

The determination result of the above first determination is output to the second determining means 407.

(6) The cutting position information selecting means 409
It is started after the completion of the first determination in step 505, information regarding cutting positions in each section set by the section setting unit 404 is selected, and the selection result is stored in the local cutting position information storage unit 410.

(7) The second judging means 407 is activated after the cutting processing by the cutting position information selecting means 409, and the judgment result (including the value used in the judgment process) of the first judging means 406 and the local cutting. From the coordinate value of the position (cutting position) of the cutting point in the section of interest received from the position information storage means 410, the cutting point ai (i = 0, ..., N _upp ) above the ruled line.
The following judgment (the judgment, and the second judgment consisting of) is performed on the set of the cut point bj (j = 0, ..., N _low ) under the ruled line (step 506). here,
The interference patterns output as the determination results of, and are the intersection patterns of “intersection 2”, “intersection 1”, and “overlap”. These correspond to the interference patterns shown in FIGS. 8, 7 and 9, respectively.

Judgment: N _upp = 2 and N _low = 1
_Where x _aM0 −x _bM0 > D _short , and x _bL0 −x _aR0 <D _long , and x _aM1 −x _bM0 > D _short , and x _aL1 −x _bR0 <D _long , and x _aL1 −x. _{If aR0} <D _long2 , h _a0 <H _short , h _a0 > H _long , and h _a1 > H _long, then a0, a1, and b0 are cut points related to the “intersection 2” interference pattern. To determine. The values of the respective threshold values are, for example, D _short = 2 (this numerical value indicates the number of pixels. The same applies hereinafter), D _long = 7, D _long2 = 14,
H _short = 5 and H _long = 5.

Judgment: N _upp = 2 and N _low = 2
X _aR0 −x _aL0 <W _th , and x _aR1 −x _aL1 <W _th , and x _bR0 −x _bL0 <W _th , and x _bR1 −x _bL1 <W _th , and | x _aM0 − x _bM0 | <D _th2 and | x _aM1 −x _bM1 | <D _th2 and x _aL1 −x _aR0 <D _th3 and h _a0 > H _Long and h _b1 > H _Long and h _{If b0} <H _short and h _a1 <H _short, then a0, a1, b0, and b1 relate to two sets of “intersection 1” interference patterns (a0 and b0 and a1 and b1). It is determined to be a cut point. In addition,
The value of each threshold is, for example, W _th = 5, D _th2 = 10,
D _th3 = 14, H _Long = 7, and H _short = 5.

Judgment: N _upp = 2 and N _low = 0
Then, if x _aM1 −x _aM0 <D _th4 , and h _a0 > H _Long , and h _a1 > H _Long, then it is determined that a0 and a1 are cutting points related to the “overlapping” interference pattern. The threshold values are, for example, D _th4 = 20 and H _Long = 7.

The second judgment means 407 outputs the judgment result of the above second judgment and the judgment result of the first judgment by the first judgment means 406 to the restoration means 408.

In the second determination, the number of N _upp and N _low in the section is limited as described above. Therefore, it is considered that the numbers are suitable as the characters to which this judgment is applied.

(8) The restoring means 408 executes step 50.
6 is started after the second determination in 6 is completed, the determination result by the first determination means 406 and the determination result by the second determination means 407 are input, and the section of interest received from the cutting position storage memory 3 based on these determination results. Restoration processing is performed on the character loss position in the image memory 1 indicated by the coordinate values of the cut position in the inside (step 507).

This restoration processing is performed by different procedures for the three types of interference patterns of "crossing 1", "crossing 2", and "overlap". below~
Now, the procedure of the restoration process for each interference pattern will be described (see FIGS. 10 to 12 which are specific examples).

The following description will be limited to the case of restoration processing of a character loss caused by a horizontal ruled line, but the x axis and y
By exchanging the relationship of the axes, it is possible to similarly restore the character loss caused by the vertical ruled line.

Procedure of Restoration Process for Interference Pattern of “Cross 1” As shown in FIG. 10, a straight line connecting the right end of the cut point above the ruled line and the right end of the cut point below the ruled line and the left end of the upper and lower cut points as well And a straight line connecting them are generated. Convert the pixel value of the pixel (pixel indicated by "○" in FIG. 10) in the area surrounded by the upper and lower cutting points and the two straight lines generated by the above procedure from "0" to "1" To do.

Procedure of Restoration Processing for Interference Pattern of “Cross 2” For the cut points a0 and a1 above the ruled line and the cut point b0 below the ruled line, a1 and b, and a1 and b
The same procedure as that applied to the above "crossing 1" is applied to each of 0 and (see FIG. 11. The pixel value of the pixel indicated by "○" in FIG. 11 is changed from "0" to "1".
Convert to).

Procedure of Restoration Processing for "Overlapping" Interference Pattern As shown in FIG. 12, for the cut points a0 and a1 above the ruled line, first, two cut points a0 and a1.
The circumscribed rectangular area (area of pixels indicated by “x” in FIG. 12) is set. Next, the pixel value of the pixel included in the set circumscribed rectangular area is converted from "0" to "1" (it is originally "1" in the example of FIG. 12). Further, the pixel values of the pixels (pixels indicated by “◯” in FIG. 12) included in the area of two lines (the line between two pixels) adjacent to the lower part of the circumscribed rectangular area are changed from “0” to “0”. 1 ". However, the length of each line is shortened as it goes away from the rectangular area.

In the present embodiment, the description has been directed to "interference between horizontal ruled lines and characters (particularly, interference between horizontal ruled lines and lower part of characters)", but as mentioned above in part,
For interference between horizontal ruled lines and the upper part of characters, vertical ruled lines and the right part of characters, and vertical ruled lines and the left part of characters,
The same can be dealt with by rotating the coordinates in the above description by 180 °, 90 ° counterclockwise and 90 ° clockwise.

[0096]

As described above, according to the ruled line removing method of the present invention, attention is paid to the cutting position of a defective character generated after the ruled line is erased, and the graphic structure in the vicinity thereof is analyzed to cause interference between the character and the ruled line. By estimating the pattern and performing restoration processing based on the interference pattern, it is more accurate and faster than other methods (conventional ruled line removal method) that use contour tracking, etc. This has the effect of enabling restoration processing of. With this effect, it is possible to improve the processing accuracy and processing speed of the ruled line removal processing in the OCR incorporating the ruled line removal method of the present invention.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of a ruled line removing method of the present invention.

FIG. 2 is a flowchart showing a process (process related to overall operation) of the ruled line removal method shown in FIG.

FIG. 3 is a flowchart showing a process of ruled line erasing means in FIG.

FIG. 4 is a block diagram showing a configuration of a character loss restoring means in FIG.

FIG. 5 is a flowchart showing the processing of the character loss restoring means shown in FIG.

FIG. 6 is a diagram for explaining an operation of ruled line erasing means in FIG. 1.

FIG. 7 is a diagram showing an example of an interference pattern “crossing 1”.

FIG. 8 is a diagram showing an example of an interference pattern “intersection 2”.

FIG. 9 is a diagram showing an example of an interference pattern “overlap”.

FIG. 10 is a diagram for explaining a restoration process for “intersection 1”.

FIG. 11 is a diagram for explaining a restoration process for “intersection 2”.

FIG. 12 is a diagram for explaining a restoration process for “overlap”.

13 is a diagram showing a data structure of ruled line position information accumulated in a ruled line position storage memory in FIG.

[Explanation of symbols]

 1 Image Memory 2 Ruled Line Erasing Means 3 Cutting Position Saving Memory 4 Character Loss Restoring Means 5 Ruled Line Position Saving Memory 401 Intersection Coordinate Calculating Means 402 Intersection Coordinate Storage Means 403 Cutting Position Information Rejecting Means 404 Section Setting Means 405 Feature Extracting Means 406 First Judgment Means 407 Second judging means 408 Restoring means 409 Cutting position information selecting means 410 Local cutting position information storing means

Claims (9)

[Claims] 1. An image memory in which a form image having characters and ruled lines formed by a small number of strokes is accumulated in advance, and a ruled line position in which ruled line position information of ruled lines in the form image in the image memory is accumulated. A storage memory, a cutting position storage memory that stores the coordinate value of the cutting position of the character generated at the point where the character and the ruled line interfere, and the image memory using the ruled line position information read from the ruled line position storage memory. Ruled line erasing means for erasing the ruled lines in the form image in the inside, and for saving the coordinate value indicating the cutting position of the character component generated at the interference point between the character and the ruled line in the process of erasing the ruled lines in the cutting position saving memory, Based on the ruled line position information in the ruled line position saving memory, the coordinate value indicating the cutting position in the cutting position saving memory, and the reference of the form image in the image memory, a book near the cutting position is obtained. A character loss restoration means for analyzing a graphic structure of an image, estimating an interference pattern between a character and a ruled line generated at the cutting position by the analysis, and restoring a defective portion of the character based on the estimation. A characteristic ruled line removal method. 2. A temporary ruled line is once detected, then an average line width and an allowable line width are calculated from the temporary ruled line detection result, and the ruled line detection result is corrected based on the calculation result to determine the ruled line to be erased. The ruled line removing method according to claim 1, further comprising: 3. An image memory for accumulating a form image of a form having vertical ruled lines and horizontal ruled lines, and in recognizing an area for one character, if the horizontal ruled line is a target ruled line, it is separated by vertical ruled lines. The character loss restoration means for performing the recognition based on one section and performing the recognition based on one section divided by a horizontal ruled line when the vertical ruled line is a target ruled line. Alternatively, the ruled line removing method according to claim 2. 4. The character loss restoration means, wherein the “interference pattern between characters and ruled lines” to be estimated is “intersection 1”, “intersection 2”, and “overlap”.
Alternatively, the ruled line removing method according to claim 2. 5. The character loss restoration means having the "interference pattern between characters and ruled lines" to be estimated as "intersection 1", "intersection 2", and "overlap".
Ruled line removal method described. 6. The ruled line removing method according to claim 1, wherein the “character formed by a small number of strokes” is a numeral. 7. The ruled line removing method according to claim 3, wherein the “character formed by a small number of strokes” is a numeral. 8. The ruled line removing method according to claim 4, wherein the “character formed by a small number of strokes” is a numeral. 9. The ruled line removing method according to claim 5, wherein the “character formed by a small number of strokes” is a numeral.