JPH1145339A - Broken line identification method and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To remove a thinned solid line and to precisely identify a broken line by identifying the broken line or a pseudo broken line based on the sizes of broken line elements and the variance value of a distance between the elements. SOLUTION: A picture inputted from a binary picture input part 100 is stored in a binary image memory 110. A rectangle extraction part 120 uses data of the binary image memory 110, extracts a rectangle and stores it in a rectangle memory 130. A broken line/ruled line extraction part 140 uses rectangle information in the memory 130 and extracts the broken line. The broken line extracted in the broken line/ruled line extraction part 140 is stored in a broken line/ruled line memory 150, and a processing terminates when it is not extracted. A broken line judgement part 160 removes the pseudo broken line and sorts the pseudo broken line from the real broken line. The removal processing of the pseudo broken line is executed on all the extracted broken lines. The remaining real broken lines from which the pseudo broken lines are removed by the broken line judgment part 160 are stored in a broken line/ruled line memory 170.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for identifying a broken line and a pseudo broken line with high accuracy and a recording medium.

[0002]

2. Description of the Related Art As a method of recognizing dotted ruled lines constituting a table, there is, for example, a ruled line recognizing method described in JP-A-7-230525. According to this method, a rectangle that is equal to or smaller than a predetermined threshold is extracted as a dotted line element, and a rectangle obtained by integrating elements whose intervals between the elements are within a predetermined threshold is extracted as a dotted ruled line. In addition, the validity of the ruled line is determined based on the respective variances of the size of the rectangle and the interval between the rectangles.

[0003]

However, if the rectangle extraction process is performed at a predetermined threshold value as described above, for example, there is a possibility that a rectangle constituting a character is erroneously extracted as a broken line component. is there. In addition, when the image is skewed or has noise, there is a problem that a broken line cannot be correctly extracted.

An object of the present invention is to calculate the variation (statistics such as variance) of the size of the elements constituting the broken line and the distance between the elements (broken line interval) after or during the broken line identification processing, An object of the present invention is to provide a dashed line identification method and a recording medium with improved dashed line identification accuracy by removing those having a variation amount equal to or greater than a predetermined threshold value as pseudo dashed lines.

[0005]

In order to achieve the above object, according to the first aspect of the present invention, a rectangle in which black pixel connected components are integrated is extracted from a document image, and a dashed line is formed from the extracted rectangle. A method for extracting a rectangle (hereinafter referred to as a dashed line element) and identifying whether it is a dashed line or a pseudo dashed line composed of a part of a character rectangle based on the dashed line element, wherein the size of the dashed line element is The method is characterized in that it is distinguished between a broken line and a pseudo broken line based on the variance of the distance between elements.

According to the second aspect of the present invention, a broken line or pseudo line is based on the ratio of the width and the height of the broken line element, the ratio of the width of the broken line element to the distance between the elements, and the ratio of the height and the distance between the elements. It is characterized by identifying whether it is a broken line.

According to a third aspect of the present invention, it is characterized in that a dashed line or a pseudo dashed line is identified based on the area of a black pixel or the area of a white pixel occupying the broken line element.

According to a fourth aspect of the present invention, it is characterized in that it is distinguished as a broken line or a pseudo broken line based on the difference between the position information of the broken line and the position information of the character rectangle.

According to a fifth aspect of the present invention, it is characterized in that it is distinguished as a broken line or a pseudo broken line based on the difference between the position information of the broken line element and the position information of the character rectangle.

In the invention according to a sixth aspect, the identified broken line is a broken line including a dotted line, an alternate long and short dash line.

[0011] According to the invention described in claim 7, according to claims 1, 2,
All of the methods described in 3, 4, and 5, or a part thereof are combined to identify a broken line or a pseudo broken line.

[0012] According to the invention described in claim 8, according to claims 1 and 2,
The method is characterized in that the broken line identified by the method described in 3, 4, 5, or 7 is labeled as a rectangle constituting the broken line.

According to the ninth aspect of the present invention, a function of extracting a rectangle integrating black pixel connected components from a document image, and a rectangle forming a broken line from the extracted rectangle (hereinafter, a broken line element)
And a function for identifying whether a dashed line or a pseudo dashed line composed of a part of a character rectangle based on the size of the dashed line element and the distance between the elements. It is a computer-readable recording medium in which a program for causing a computer to implement a function of labeling a rectangle forming a broken line is recorded.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be specifically described below with reference to the drawings. Before describing each embodiment of the present invention, first, terms used in the present invention will be defined. Rectangle / rectangle extraction: A continuous image or a continuous image portion (for example, a continuous black pixel portion or a continuous white image portion in the case of a binary image) that is equal to or more than a predetermined threshold value is bundled into an image A rectangle enclosed by a circumscribed rectangle so that they are included is defined as a rectangle, and extracting the position coordinates of the rectangle is defined as rectangle extraction.

Broken line / broken line element / pseudo broken line: A line other than a solid line is defined as a broken line. That is, the broken line is a general term including a dotted line, a one-dot chain line, a two-dot chain line, and the like. The rectangle forming the broken line is defined as a broken line element. For example, a binary image indicates a rectangle of black pixels. A dashed line that is incorrectly identified as a dashed line despite being not a dashed line is called a pseudo dashed line. For example, a part of a rectangle forming a character is a target of the broken line identification processing, and a broken line extracted as a result of erroneous recognition corresponds to the pseudo broken line.

[0016] Inter-element distance: the distance between the elements constituting the broken line. This distance represents, for example, the difference between the position coordinates of the elements and, if the element is a binary image composed of black pixels, the length of the continuous white image portion between the black pixel elements.

FIG. 1 shows a configuration common to the first to fifth embodiments of the present invention. FIG. 2 shows Embodiments 1 to 5 of the present invention.
7 is an overall processing flowchart common to FIG. The binary image input unit 100 is an image input unit such as a scanner, for example. Here, the image is captured as a binary image, but may be a multivalued image. The image input from the binary image input unit 100 is stored in the binary image memory 110 (Step 1).

Next, the rectangle extracting unit 120 extracts a rectangle using the data of the binary image memory 110 and stores it in the rectangle memory 130 (step 2). The dashed line extraction unit 140 extracts a dashed line by using the rectangle information in the memory 130 (step 3). The dashed line extraction process
Any method may be used as long as it identifies a broken line. The dashed line extracted by the dashed line extraction unit 140 is stored in the dashed line memory 150 (step 4), and if not extracted, the process ends. At this stage, a pseudo broken line is also included as a broken line. As will be described in detail below, the broken line determining unit 160 removes the pseudo broken line and selects a pseudo broken line and a true broken line (step 5). This pseudo dashed line removal processing is performed for all extracted dashed lines (step 6).

The broken line determining unit 160 includes a variance calculating unit 161,
A ratio calculating unit 162, a constant information amount occupancy calculating unit 163,
A rectangular position and a broken line position comparing / calculating unit 164 and a rectangular position / a broken line element position comparing / calculating unit 165 are provided. The remaining true dashed line from which the dashed line is removed by the dashed line determination unit 160 is stored in the dashed line memory 170. Here, although the memory 170 is provided exclusively as a broken line ruled line memory,
May be shared.

In FIG. 1, as the processing order, the dashed line is extracted by the dashed line extraction unit 140 and then the dashed line determination unit 160 determines the dashed line. .

<Embodiment 1> In the first embodiment, the broken line determination unit 1
Only the 60 variance calculation units 161 function. Distributed calculation unit 1
At 61, the extracted dashed line in the dashed line rule memory 150,
Alternatively, the variation of the width and height of the broken-line element being extracted and the distance between the elements is calculated. Here, a description will be given of, for example, the dispersion of statistical processing as an example showing the variation.

FIG. 3 shows a dashed line determination unit 160 according to the first embodiment.
It is a processing flowchart of. The dashed line determination unit 160 extracts the element rectangles constituting the dashed line from the dashed line memory 150 (step 11), and the variance calculation unit 161 calculates the variance A1 for each of the height, width, and inter-element distance of the element rectangle. , A2, A3 (variation) are calculated (step 12). Each threshold A that can be determined to be a pseudo-dashed line in consideration of variations due to blurring or the like.
1 ′, A2 ′, and A3 ′ are determined, and when the above-mentioned variation is equal to or greater than the threshold value, it is determined that it is a pseudo-dashed line (step 13). The dashed line determined as the pseudo dashed line is removed and the true dashed line is stored in the dashed line rule memory 170 (step 1).
4).

<Embodiment 2> In the above-described embodiment, there is a possibility that a pseudo broken line which cannot be removed exists. In this embodiment,
This is an improvement in the accuracy of identification of a broken line, and a broken line determination unit 1
Only the 60 ratio calculators 162 function.

FIG. 4 shows a broken line determining unit 160 according to the second embodiment.
It is a processing flowchart of. The dashed line determining unit 160 extracts the element rectangles constituting the dashed line from the dashed line rule memory 150 (step 21), and the ratio calculation unit 162 determines the width and height of the extracted dashed line or the dashed line element being extracted and the distance between the elements. From the value of the distance, W is the width, H is the height, and D is the distance between elements.
Then, D and W, W and H, D and H ratio B1, B2,
B3 is calculated (step 22). All combinations may be used as the combination of ratios, and the calculated value may be an average value. Then, the characteristic amount of each ratio of a general pseudo-dashed line is measured, the respective thresholds B1 ', B2', B3 'are determined based on the characteristic amounts, and the above-described ratio value is compared with the threshold. When the above ratio is equal to or greater than the threshold value, it is determined that the dashed line is a pseudo dashed line (step 23), and the dashed line determined as the pseudo dashed line is removed (step 24).

<Embodiment 3> In Embodiments 1 and 2 described above,
There may be pseudo-dashed lines that cannot be removed. In the present embodiment, the accuracy of identifying a broken line is further improved, and only the constant information amount occupancy calculating unit 163 of the broken line determining unit 160 functions.

FIG. 5 shows a broken line determining unit 160 according to the third embodiment.
It is a processing flowchart of. The dashed line determining unit 160 extracts the element rectangles constituting the dashed line from the dashed line rule memory 150 (step 31). The constant information amount occupancy calculating unit 163 determines whether the extracted dashed line or the dashed line element being extracted is black. A pixel ratio C is calculated (step 32).

FIG. 6 shows an example of a character rectangle and an example of a dotted line element rectangle. The area of a rectangle is represented by S, and the area of a black pixel in the rectangle (FIG. 6)
In the example, the area of the character a) is defined as Black, and the rectangular black pixel ratio C is defined as C = Black / S. Using the fact that the black pixel ratio is low if the rectangle is a character and the black pixel ratio is high if the rectangle is a dashed line element, it is determined whether it is a pseudo dashed line or a true dashed line. Assuming that the predetermined threshold value is C ′, the black pixel ratio C of the element rectangles constituting the broken line or the average value Cavg of the black pixel ratios of all the element rectangles is calculated, and if C <C ′ or Cavg <C ′, (Step 33) The element rectangle is determined to be a character rectangle, that is, a pseudo dashed line, and is removed (Step 34). The dashed line determination unit 160 removes the dashed line determined as the pseudo dashed line and stores the dashed line in the dashed line rule memory 170.

<Embodiment 4> By using the processing method of Embodiment 3 described above, Line 2 and Line 4 shown in FIG. 7 can be easily removed as pseudo-dashed lines. But,
In the method of the third embodiment described above, it may be difficult to distinguish the pseudo dashed lines Line1 and Line3 composed of the character rectangles from the dashed line elements. Therefore, the present embodiment is an embodiment in which the pseudo broken lines Line1 and Line3 shown in FIG. 7 are removed.

The dashed line determining section 160 of this embodiment includes a rectangular position and dashed line position comparing / calculating section 164. FIG.
13 is a processing flowchart of the dashed line determination unit 160 according to the fourth embodiment. In the comparison calculation unit 164, the extracted broken line,
Alternatively, the position information of the dashed line itself being extracted is compared with the position information of the character rectangle on the image, and the difference D between the position of the dashed line and the position of the character rectangle is calculated (step 42). If it is' (step 43), the dashed line is determined to be a pseudo dashed line composed of a part of the character rectangle, and is removed (step 44).

FIG. 9 shows an example (Line 1) of the pseudo dotted line removed in the fourth embodiment. The dotted line in the figure indicates the position range of the extracted rectangle. If D is a horizontally written character,
This is the horizontal distance between the rectangular position and the Line1 position. As a method of calculating D, a distance to an arbitrary rectangle existing in the vicinity of Line 1 is calculated, a distance to all rectangles existing in the target image is calculated, or a character rectangle existing in the vicinity of Line 1 is calculated. Any calculation method such as calculating the distance from the determined rectangle may be used. In the present embodiment, if the calculated distance D is equal to or less than a predetermined threshold, Line1 is determined to be a pseudo dotted line and removed.

Fifth Embodiment In the processing of the first to fourth embodiments, there is a possibility that a pseudo broken line that cannot be removed exists. In the fourth embodiment, since only the position information of the broken line is compared with the position information of the character rectangle, in the case of a small table or a dense table where ruled lines and characters are close to each other, a true broken line is used. Despite this, there is a possibility that it will be erroneously removed as a pseudo broken line.

Therefore, in the present embodiment, after or during the dashed line identification processing, the position information on the image of each element rectangle forming the dashed line and the position information on the rectangular image forming the character To determine whether the extracted dashed line is a pseudo dashed line composed of character rectangular elements, thereby further improving the accuracy of dashed line identification. .

FIG. 10 shows a broken line determining unit 16 according to the fifth embodiment.
0 is a processing flowchart of FIG. The comparison calculation unit 165 compares the position information of the extracted dashed line or the dashed line element being extracted with the position information of the character rectangle, and determines the difference between the position of the dashed line element and the position of the character rectangle. E is calculated (step 53), and if the difference E is equal to or smaller than a predetermined threshold value E '(step 54), the dashed line element is determined to be a pseudo dashed line element composed of a part of a character rectangle, and counted ( Step 55).

FIG. 11 is an example of a pseudo broken line removed in the fifth embodiment. Line 1 in the figure indicates a pseudo-dashed line in which the character rectangle 1 (ell) is erroneously identified as a broken line element. In the fifth embodiment, the distance E between each element constituting Line 1 and the neighboring rectangle is calculated.

That is, the method of calculating the distance E checks whether or not there is a character rectangle on both sides as described below. Notice the character rectangle of l (el) of the element in FIG.
The Ye coordinate value of (L) is set to baseline.

The Ye coordinate value of a rectangle separately extracted as a character rectangle exists in the range of dye, and the Xe coordinate value of the rectangle separately extracted as a character rectangle falls in the range of dxL, or the Xs coordinate value of the rectangle is extracted. It is in the range of dxR. Here, dye, dxL, and dxR are predetermined thresholds.

If the above condition is satisfied for all the numbers of broken line elements, that is, the calculated E is equal to the predetermined threshold value E '.
If it is less than or equal to a character rectangle (for example, l (el) or I
(Eye)), which can be determined as a pseudo-dashed line element. In this manner, it is determined whether all elements constituting Line 1 are pseudo dashed line elements or dashed line elements.

Next, a method of determining whether or not Line 1 is a pseudo broken line will be described. Assuming that the number of components of the broken line is N and the number determined to be a pseudo-dashed line element in the above processing is NGnum, if N = NGnum or N = NGnum × 2, it is determined to be a pseudo-dashed line (step 57), It is removed as a broken line (step 58). Since the number of components of Line1 in FIG. 11 is three, if the number of elements determined to be a pseudo-dashed line element is three, Line1 is determined to be a pseudo-dashed line. Further, since the number of components of Line 3 in FIG. 11 is six, if the number of elements determined to be a pseudo-dashed line element is three,
Line 3 is determined to be a pseudo broken line.

As described above, in this embodiment, it is determined whether or not each dashed element is a pseudo dashed line. Therefore, the pseudo dashed line can be identified and excluded more reliably than in the fourth embodiment. The extraction rate of the broken line is improved. Further, even if the character is a true dashed line, if a character exists near the dashed line, the processing of the fourth embodiment may erroneously exclude the true dashed line as a pseudo dashed line. On the other hand, in the processing of the present embodiment, since the number of dashed line elements and the number of pseudo dashed line elements are compared and determined, a pseudo dashed line can be reliably identified and excluded.

Embodiment 6 In Embodiments 1 and 2 described above,
Although the variation (variance) and the ratio of the distance between the elements constituting the extracted broken line are calculated, in this embodiment, the type of the broken line can be classified by using the calculation result. Further, in this embodiment, in addition to the ruled line information, the result of classifying the type of the broken line can be used to obtain detailed data for ruled line discrimination. Tables close to can be reproduced.

FIG. 12 shows the configuration of the sixth embodiment of the present invention. The configuration shown in FIG.
0, a broken line type classification unit 190 and a ruled line reproduction unit 200 are provided. Other components are the same as those in FIG. FIG.
13 is a processing flowchart of the sixth embodiment.

The calculation results of the variance calculator 161 and the ratio calculator 162 are held in the calculation result memory 180. Based on the calculation result, the dashed line type classification unit 190 classifies the dashed line type into a dotted line, a one-dot chain line, a two-dot chain line, and others (step 66).

FIG. 14 is a flowchart of the process performed by the broken line type classification unit. FIG. 15 shows the height h, width w, and inter-element distance d of the broken line elements in the identification direction. The classification method calculates, for the identified dashed line, values representing variations in the height, width, and distance between elements constituting the dashed line, for example, variance values h, w, and d (step 71). ). Assuming that a predetermined threshold is V, h, w, and d are h <V and w <V and d <
If the condition of V is satisfied (step 72), the dashed line is determined to be a dotted line (a dashed line having the same size of constituent elements), and the type of the dashed line is labeled with the dotted line (step 73). Note that the predetermined threshold V may be different for each of h, w, and d.

Next, if it is not determined to be a dotted line, the process proceeds to a condition for determining whether or not a condition for a one-point and two-dot chain line is satisfied. V2, V3
(V2 <V3), h <V2 and d <V2 and w <
If V3 is satisfied (step 74), it is determined that the broken line is a one-point or two-dot chain line, and labeling is performed as a one-point or two-dot chain line (step 75).

If it is determined that the position is a one-point or two-dot chain line, the process proceeds to a more detailed classification. The variance values h1 and w1 of the height and width are calculated for every other element in the broken line component (step 76). Assuming that the predetermined threshold value is V4, if h1 <V4 and w1 <V4 are satisfied (step 77), it is determined as a dashed line and labeling is performed (step 78).

Next, when no labeling is performed,
Similarly, variance values h2 and w2 of the height and width are calculated for every third element (step 79). Assuming that the predetermined threshold value is V5, if h2 <V5 and w2 <V5 are satisfied (step 80), it is determined as a two-dot chain line and labeling is performed (step 81). Also, the ruled line reproducing unit 200 such as DTP reproduces the ruled line. At this time, a correct ruled line can be reproduced using the classification result of the broken line type classification unit 190.

<Embodiment 7> This embodiment is a combination of the above-described embodiments, and can remove pseudo dashed lines with high accuracy and extract dashed lines with high accuracy.

<Embodiment 8> In this embodiment, a pseudo dashed line is removed in each of the above embodiments, and a correctly identified dashed line is labeled as a rectangle constituting the dashed line, and is used once for identification. By clarifying the rectangle
This is an embodiment in which misidentification during other identification processing is reduced, and the accuracy of table processing structure recognition and table reproduction processing is improved.

FIG. 16 shows the configuration of the eighth embodiment of the present invention. The configuration of FIG. 1 further includes a rectangular labeling unit 300, a rectangular memory 310, and another identification processing unit 320.
FIG. 17 is a processing flowchart of the eighth embodiment. The dashed line determination unit 160 correctly identifies the dashed line memory 170
(Step 95), the rectangular labeling unit 3
Labeling with 00 (step 96).

Here, labeling refers to, for example, attaching a broken line identification number or the like to a rectangle constituting a correctly identified broken line. Also, for example, not only a broken line but also a ruled line such as a solid line which has been correctly identified,
The rectangles used for other identification processing are labeled in the same manner.

The labeled rectangle is a rectangle memory 310.
Is held. The memory 130 may be used as the rectangular memory 310. After the dashed line identification process ends,
When performing other identification processing 320, it is determined whether or not the rectangle is a rectangle such as a ruled line by using the rectangle labeling information in the rectangle memory 310, and a rectangle not related to the current identification processing is removed. The identification process can proceed.

That is, conventionally, when performing another identification processing, the binary image input 100 → storage into the binary image memory 110 → rectangular extraction 120 → storage into the rectangular memory 130 proceeds again. Alternatively, although the rectangle information of the rectangle memory 130 is used, in the present embodiment, since the labeled rectangle information is added to the rectangle information of the rectangle memory 130, the identification process can be easily performed. Thereby, it is possible to reduce false recognition and realize high-speed processing.

<Embodiment 9> FIG. 18 shows Embodiment 9 of the present invention.
Is an embodiment realized by software. When the present invention is implemented by software, as shown in FIG. 18, a general-purpose processing device including a CPU, a ROM, a RAM, a display device, a hard disk, a keyboard, a CD-ROM drive, and a scanner is prepared. On a computer-readable recording medium such as a ROM, a program for realizing the broken line identification function of the present invention is recorded. Images such as documents input from a scanner or the like are temporarily stored on a hard disk or the like. When the program is started, the temporarily stored image data is read, a broken line identification process is executed, and the identification result is output to a display or the like.

[0054]

As described above, according to the first and ninth aspects of the present invention, a variation in information such as a size of a broken line element and a distance between broken line elements is calculated, and a pseudo value is calculated based on the calculated variation. Since it is identified whether it is a broken line or a broken line, a faint solid line or the like can be removed, and a broken line can be correctly identified.

According to the second and ninth aspects of the present invention, it is determined whether a dashed line or a dashed line is based on the respective ratios of the width, height, and distance between dashed line elements. It is possible to accurately remove pseudo-dashed lines or faint solid lines.

According to the third aspect of the present invention, whether a pseudo-dashed line or a broken line is identified based on the area ratio of a black pixel or a white pixel occupying a broken-line element, the pseudo-dashed line or the blurred portion is accurately detected. Solid lines and the like can be removed.

According to the fourth aspect of the present invention, whether a broken line or a pseudo broken line is identified based on the difference between the position information of the broken line and the position information of the character rectangle, the pseudo broken line is removed. Accuracy is improved, and a broken line can be correctly identified.

According to the fifth aspect of the present invention, whether a line is a broken line or a pseudo broken line is identified based on the difference between the position information of the broken line element and the position information of the character rectangle. Accuracy can be further improved, and a broken line can be correctly identified.

According to the sixth aspect of the invention, the type of the broken line can be identified as a dotted line, a one-dot chain line, a two-dot chain line, etc., so that the ruled line can be accurately reproduced.

According to the seventh aspect of the present invention, since the above-mentioned respective identification methods are appropriately combined, the accuracy of removing the false dashed line is further improved, and the dashed line can be extracted with high accuracy.

According to the eighth aspect of the present invention, labeling is performed on a rectangle that constitutes a dashed line that has been correctly identified. Therefore, erroneous identification is performed by using the labeling information during another identification process. Is prevented, and the identification rate of the broken line is improved.

[Brief description of the drawings]

FIG. 1 shows a configuration common to Embodiments 1 to 5 of the present invention.

FIG. 2 is an overall processing flowchart common to Embodiments 1 to 5 of the present invention.

FIG. 3 is a processing flowchart of a broken line determination unit 160 according to the first embodiment.

FIG. 4 is a processing flowchart of a dashed line determination unit 160 according to the second embodiment.

FIG. 5 is a processing flowchart of a dashed line determination unit 160 according to the third embodiment.

FIG. 6 shows an example of a character rectangle and an example of a dotted line element rectangle.

FIG. 7 shows an example of a pseudo broken line.

FIG. 8 is a processing flowchart of a dashed line determination unit 160 according to the fourth embodiment.

FIG. 9 shows an example of a pseudo-dashed line removed in the fourth embodiment.

FIG. 10 is a processing flowchart of a dashed line determination unit 160 according to the fifth embodiment.

FIG. 11 is an example of a pseudo-dashed line removed in a fifth embodiment.

FIG. 12 shows a configuration of a sixth embodiment of the present invention.

FIG. 13 is a processing flowchart of a sixth embodiment.

FIG. 14 is a processing flowchart of a broken line type classification unit.

FIG. 15 shows the height h, width w,
The distance d between elements is shown.

FIG. 16 shows a configuration of Example 8 of the present invention.

FIG. 17 is a processing flowchart of the eighth embodiment.

FIG. 18 shows a configuration of a ninth embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 100 binary image input unit 110 binary image memory 120 rectangle extraction unit 130 rectangle memory 140 dashed ruled line extraction unit 150 dashed ruled line memory 160 dashed line determination unit 161 variance calculation unit 162 ratio calculation unit 163 constant information amount occupancy calculation unit 164 rectangle Comparison calculation section 165 between position and broken line position Comparison calculation section between rectangle position and broken line element position 170 Ruled line memory

Claims (9)

[Claims] 1. A rectangle integrating black pixel connected components is extracted from a document image, and a rectangle (hereinafter, a dashed line element) constituting a dashed line is extracted from the extracted rectangle, and a dashed line is formed based on the dashed line element. Or a pseudo-dashed line composed of a part of a character rectangle, the method being a broken line or a pseudo-dashed line based on the size of the broken-line element and the variance of the distance between the elements. A method for identifying a broken line. 2. A method for identifying a dashed line or a pseudo dashed line based on the ratio of the width and height of the broken line element, the ratio of the width of the broken line element to the distance between the elements, and the ratio of the height and the distance between the elements. 2. The method for identifying broken lines according to claim 1, wherein: 3. The method according to claim 1, wherein the method determines whether the pixel is a dashed line or a pseudo dashed line based on an area of a black pixel or an area of a white pixel occupying the dashed line element. 4. The method for identifying a broken line according to claim 1, further comprising the step of identifying a broken line or a pseudo broken line based on a difference between the position information of the broken line and the position information of the character rectangle. 5. The dashed line identification method according to claim 1, wherein a dashed line or a pseudo dashed line is identified based on a difference between the position information of the dashed line element and the position information of the character rectangle. 6. The broken line identification method according to claim 1, wherein the identified broken line is a broken line including a dotted line, a dashed line, and a two-dot chain line. 7. A method for identifying a dashed line as a dashed line or a pseudo dashed line by combining all of the methods according to claim 1, 2, 3, 4, and 5, or a combination thereof. . 8. A method for identifying a dashed line, wherein the dashed line identified by the method according to claim 1, 2, 3, 4, 5, or 7 is labeled as a rectangle constituting the dashed line. 9. A function for extracting a rectangle integrating black pixel connected components from a document image, a function for extracting a rectangle forming a broken line (hereinafter, a broken line element) from the extracted rectangle, and a size of the broken line element Now, a function of identifying whether a dashed line or a pseudo dashed line composed of a part of a character rectangle based on the distance between elements, and a function of labeling the identified dashed line as a rectangle. Computer-readable recording medium on which a program for causing a computer to realize the above is recorded.