JPH10334183A - Black-and-whilte inverted area discriminating device and table recognition device using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately discriminate, specially, a black-and-white inverted area and to perform table recognition according to the discrimination result as to the recognition device which recognizes image data in table format such as a document. SOLUTION: Image data read by a scanner, etc., are stored in an image area storage part 31 through an image input part 30, a segment extraction part 32 uses a segment extraction mask for the image data to segment local areas, and a segment integration part 33 integrates segments whose black pixels in the area exceed a specific quantity value horizontally and vertically; and a black-and-white area determination part 35 discriminates segments above a certain threshold among the integrated segments as a black-and-white inverted area, thereby determining an area where black-and-white inverted areas overlap, for example, horizontally and vertically as a black-and-white inverted area. Further, a ruled-line extraction part 34 recognizes a table such as a document according to extracted ruled lines and the black-and-white inverted area.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a black-and-white inversion area discriminating apparatus for inputting tabular image data such as a form using an input device such as a scanner and performing table recognition processing on the image data. The present invention relates to a table recognition device using a device.

[0002]

2. Description of the Related Art In recent years, an input device such as a scanner has been widely used as a device for supplying a computer with characters and forms on a document as image data. Further, image data including a table supplied from such an input device is recognized as characters and ruled lines by an image recognition device,
It is stored in a computer, for example, as a corresponding character code or vector data.

[0003] Among the above-mentioned image recognition apparatuses, in order to recognize a table having a relatively thin line such as a ruled line and a long line vertically and horizontally, a table recognition apparatus using a run (Japanese Patent Laid-Open No. 6-33790) is particularly preferred. ) Has been proposed. This device is
As shown in FIG. 34, from the image input unit 101 to the image area 10
2, a line image having a certain length or more is extracted from the input image using two runs in the vertical and horizontal directions by the solid line run extraction unit 103, and a ruled line is formed by connecting them. It is to extract. Also, among the extracted ruled lines,
A portion surrounded by a ruled line having a width equal to or larger than the threshold value is output to the black-and-white inverted cell candidate extraction unit 104 as a black-and-white inverted cell candidate. When the density of black pixels inside the extracted black-and-white inverted cell candidate is, for example, 50% or more, the black-and-white inverted cell determination unit 105 identifies the corresponding area (cell) as a black-and-white inverted cell,
The pixel inversion unit 106 performs a black-and-white inversion process on the cell.

After the above-described black-and-white inversion processing, a character cutout processing is performed on the data written in the black-and-white inversion area, and a recognition processing is performed on the cut-out characters.

[0005]

However, the above-mentioned conventional table recognition apparatus has the following problems. (A) First, in the conventional table recognition apparatus, as shown in FIG. 35A, a long horizontal run and a vertical run are used. Therefore, as shown in FIG. There is no problem if the image is composed of images, for example, FIG.
As shown in (1), a problem arises when a black-and-white inversion area and a normal cell area are mixed in one cell. That is, FIG.
When the number of black pixels is counted using the horizontal run and the vertical run having the length shown in FIG. 3A, the number of white pixels becomes extremely large, and the accurate number of black pixels cannot be counted. For this reason, the identification of the black-and-white inversion area is incorrect. (B) In the conventional table recognition device, as shown in FIG. 35A, a long horizontal run and a vertical run (for example, the same length as the length of the black-and-white inversion area shown in FIG. Horizontal run,
And vertical run), it is not possible to accurately count the number of black pixels when recognizing a black-and-white inversion area having different horizontal and vertical lengths of the cell to be recognized. For example, as shown in FIG. 3A, when the length of the horizontal run is L1 and the length of the vertical run is L2, the width of the black-and-white inverted cell to be recognized is longer than L1 and the vertical width is L1. If the length is longer than L2, the number of white pixels also increases, and it is not possible to accurately identify a black-and-white inversion area.

[0006] A similar situation occurs when the character size in the black-and-white inverted cell area is large. That is, in this case, the character size of the white character of the character string “company name” shown in FIG. 7B is large, and the white character portion is large when the number of black pixels is counted according to the horizontal run or the vertical run.
The count value of the number of black pixels is reduced, and it is not possible to accurately recognize a black-and-white inverted area (black-and-white inverted cell). (C) Furthermore, when the reading process of a table image such as a form by a scanner is performed, for example, in a tilted state, the black-and-white inversion area is tilted, and a long horizontal run and a vertical run without tilt are used. Counting the number of black pixels by counting results in counting the wrong (smaller) number of black pixels, resulting in erroneous recognition of the black-and-white inverted cell.

In such a case, for example, it is necessary to perform the counting process by inclining the horizontal run and the vertical run by the same angle.
However, such inclination correction requires complicated processing. In order to solve the above problem, the present invention uses a short line segment extraction mask in the horizontal direction and a short line segment extraction mask in the vertical direction, and counts and integrates the number of black pixels included in the line segment extraction mask. If the line width of the integrated line segment exceeds a predetermined threshold value, it is determined as a black-and-white inversion candidate area, and the black-and-white inversion candidate area is created in the horizontal direction and the vertical direction. By doing
It is an object of the present invention to provide a black-and-white inversion area identification device that solves the above-mentioned problems (a) to (c), and a table recognition device that uses the device.

[0008]

According to the first aspect of the present invention, in order to solve the above-described problems, binary image data in a table format is scanned in a horizontal direction using a line segment extraction mask that is long in the horizontal direction, Region extracting means for scanning in a vertical direction using a line segment extraction mask which is long in the vertical direction, and extracting regions in which the black pixel density is higher than a predetermined threshold value; and the region in which the black pixel density extracted by the region extracting means is high. A first integration unit that integrates in a direction orthogonal to the respective mask scanning directions, and if the line width of the region integrated by the first integration unit is equal to or greater than a predetermined threshold, the region is determined to be a black-and-white inversion candidate region. Second integration means for integrating the black-and-white inversion candidate areas in the respective mask scanning directions, and horizontal and vertical line segment segments for the contours of the black-and-white inversion candidate areas integrated by the second integration means Line segment And a black-and-white inversion area setting means for setting an area surrounded by the horizontal line segment segments and the vertical line segment segments created by the line segment segment generation means as a black-and-white inversion area. This can be achieved by providing a black-and-white inversion area identification device provided with:

That is, in this embodiment, the length is long in the horizontal direction and the length × width (m × n) is, for example, 1 × 6, 1 × 15, 1 × 30, etc.)
The line segment extraction mask is used to sequentially scan every fixed dot in the horizontal direction, and when a rectangular region cut out by the line segment extraction mask includes black pixels of a predetermined threshold or more, the region is scanned as described above. If the line width is equal to or greater than a predetermined threshold value based on the result of the integration, the area is regarded as a black-and-white inversion candidate area, and the area (line segment) of the black-and-white inversion candidate area is horizontally set by the second integration means. Extend in the direction.

On the other hand, the same processing is performed in the vertical direction, and a vertical m × n (for example, 6 × 1, 15 ×
Using a line segment extraction mask of (1, 30 × 1 etc.), scanning is performed sequentially at regular intervals in the vertical direction at regular intervals, and a region cut out by the line segment extraction mask contains black pixels of a predetermined threshold or more. At this time, the area (line segment) is integrated in a direction orthogonal to the vertical direction, and if the line width obtained as a result of the integration is a line width equal to or greater than a predetermined threshold, the area is determined as a black-and-white inversion candidate area. Thus, the line segment of the black-and-white inversion candidate area is extended in the vertical direction.

[0011] Then, for the black-and-white reversal candidate area created in the horizontal and vertical directions, line segment segments indicating the respective contours are created in the horizontal and vertical directions. The area enclosed by the rows is a black-and-white inversion area.

As described above, the invention according to claim 1 uses a line segment extraction mask which is extremely thin (for example, 1 dot) and relatively short (for example, 6 dots, 15 dots, 30 dots) in the horizontal direction. Extremely thin in the vertical direction (for example, 1
(Dots) and relatively short (for example, 6 dots, 15 dots, and 30 dots) line segment extraction masks are used to integrate regions with a large number of black pixels to identify a black-and-white inversion region. Even if a black-and-white inversion area exists in a part of the area, the area can be accurately identified (solution of the above-mentioned problem (a)). Even when a large size inverted character exists in the black-and-white inversion area, the line-slice extraction mask used in this example uses an extremely thin and relatively short line-segment extraction mask as described above. (Solution of the above problem (b)). Furthermore, even when the front image is read with an inclination, the black-and-white inversion area is identified by the integration processing while recognizing the number of black pixels in a narrow range, so that it can be recognized as an inclined line segment, and the black-and-white inversion in an inclined state. It can be recognized as an area (solution of the above problem (c)).

According to a second aspect of the present invention, in order to solve the above-described problem, binary image data in a table format is scanned in a horizontal direction using a line segment extraction mask that is long in a horizontal direction, and a line segment that is long in a vertical direction is scanned. An area extraction unit that scans in the vertical direction using an extraction mask to extract regions in which the black pixel density is higher than a predetermined threshold, and extracts the regions in which the black pixel density is high extracted by the region extraction unit in the respective mask scanning directions. A first integrating means for integrating in a direction orthogonal to the direction, and if the line width of the area integrated by the first integrating means is equal to or greater than a predetermined threshold, the area is determined to be a black-and-white inversion candidate area. A black-and-white inversion area is set based on the overlapping state of the horizontal black-and-white inversion candidate area and the vertical black-and-white inversion candidate area integrated by the second integration means. It can be achieved by providing a rolling tone reversal region identification apparatus provided with an area setting means.

That is, in this embodiment, the invention described in claim 1 creates a line segment segment row for the outline of the black and white reversal candidate area, and identifies an area surrounded by the line segment segment row as a black and white reversal area. On the other hand, the invention of this example has a configuration in which the black-and-white inversion area is identified based on the overlapping state of the horizontal and black-and-white inversion candidate areas in the horizontal direction and the vertical and black-and-white inversion candidate areas.

That is, this embodiment also uses a horizontally long mxn line segment extraction mask, sequentially scans in the horizontal direction at regular intervals, and sets a predetermined threshold value within a rectangular area cut by the line segment extraction mask. When the above black pixels are included, the rectangular area is integrated in the direction orthogonal to the horizontal direction, and if the integrated result indicates that the line width is equal to or greater than a predetermined threshold, the area is regarded as a black-and-white inversion candidate area, and in the vertical direction. The same processing is performed for the vertical direction, using a long line segment extraction mask, sequentially scanning in the vertical direction at a constant dot interval, and setting a black area of a predetermined threshold or more within a rectangular area cut by the line segment extraction mask. When pixels are included, the rectangular areas are integrated in a direction orthogonal to the vertical direction, and if the integrated result indicates that the line width is equal to or greater than a predetermined threshold, the area is determined as a black-and-white inversion candidate area.

Then, the overlapping state of the black-and-white inversion candidate areas created in the horizontal direction and the vertical direction is determined, and the black-and-white inversion area is identified. For example, this overlapping state indicates that the area where the black-and-white inversion area in both the horizontal and vertical directions overlaps is one area (either the horizontal direction or the vertical direction).
If the value obtained by dividing the area of the black-and-white inversion area in the direction is equal to or greater than a predetermined threshold, the area surrounded by the black-and-white inversion area in both the horizontal and vertical directions is defined as the black-and-white inversion area. With this configuration, the black-and-white inversion area can be accurately identified, and the above-described problems (a) to (c) can be solved.

The discrimination of the black-and-white inversion area is not limited to the above example. The overlapping sum of the black-and-white reversal candidate areas in the horizontal and vertical directions is obtained, and when the sum is equal to or more than a predetermined value, the black-and-white inversion candidate in both the horizontal and vertical directions is obtained. A region surrounded by the region may be set as a black-and-white inversion region, and other various methods are conceivable.

A third aspect of the present invention embodies the invention of the first or second aspect. For example, it is determined that the line width of the area integrated by the first integration means is equal to or smaller than a predetermined threshold. In this case, a ruled line judging means for judging the area integrated by the first integrating means to be a ruled line.

That is, the ruled line has a relatively narrow line width,
When it is determined that the line width of the area integrated by the first integration means is equal to or greater than a predetermined threshold, the area is considered as a black-and-white inversion candidate area as described above. Judge as a ruled line.

With this configuration, according to the present embodiment, the ruled line can be extracted simultaneously with the identification of the black-and-white inversion area, and an extremely efficient black-and-white inversion area identification device can be provided.

According to a fourth aspect of the present invention, the ruled line is created by integrating the ruled lines determined by the ruled line determining means in respective mask scanning directions to form a ruled line. It is a configuration having means.

With this configuration, when recognizing the tabular image data such as a form, the ruled line can be recognized together with the identification of the black-and-white inverted area (black-and-white inverted cell), and the identification processing of the black-and-white inverted area is completed. At the same time, for example, the creation of ruled lines on the form is also completed, and a highly efficient black-and-white inversion area identification device can be obtained.

A fifth aspect of the present invention further embodies the fourth aspect of the present invention. In the case where the ruled line creating means creates a vertical ruled line, for example, the line segment segment creating means creates the ruled line. Horizontal line segment column
An area surrounded by the vertical ruled lines is a black-and-white inversion area.

The line segment segment row is the outline of the horizontal black-and-white inversion candidate area. For example, the area surrounded by the horizontally long black-and-white inversion candidate area and the vertical ruled line is set as the black-and-white inversion area. As described above, in this example, the area surrounded by the line segment segment sequence (black and white inversion candidate area) and the ruled line is identified as a black and white inversion area.

Further, the sixth aspect of the present invention embodies the invention of the fourth aspect. When the ruled line setting means sets a horizontal ruled line, for example, the line segment segment string generating means The area surrounded by the vertical line segment segments created in step (1) and the horizontal ruled lines is a black-and-white inversion area.

In the description of the present embodiment, the line segment segments are the outlines of the vertical black-and-white inversion candidate regions, and are surrounded by the line segment segments and the horizontal ruled lines. The discriminated area is identified as a black-and-white inversion area.

A seventh aspect of the present invention embodies the first or fifth or sixth aspect of the present invention, wherein a margin for absorbing irregularities in the outline of the black-and-white inversion area is added to the line segment row. Configuration.

That is, in this example, when the line segment extraction processing is performed using the line segment extraction mask, the individually extracted regions are
For example, in this example, a predetermined number of dots are added in the vertical direction or the horizontal direction, for example, in the direction outside the black-and-white inversion area (outside of the normal vector of the line segment), and the contour of the black-and-white inversion area is added. It is a configuration that absorbs irregularities.

An eighth aspect of the present invention embodies the invention of the first or second aspect, wherein the black-and-white inversion area setting means sets a horizontal level when the number of connected components is equal to or less than a predetermined threshold value. In this configuration, a region surrounded by the direction line segment segment row and the vertical direction line segment segment row is determined as a black-and-white inversion area.

Therefore, in the present embodiment, even if the black-and-white inversion area is identified by the first aspect of the present invention, the black-and-white inversion area is not determined unless the number of connected components in the black-and-white inversion area is less than a predetermined threshold value. is there. Here, the number of connected components indicates the number of connected portions of the black printing portion or the white printing portion in the region of the black-and-white inversion candidate region. When a dot is included, the influence of a halftone dot or the like is eliminated by limiting the number of the connected components.

The ninth aspect of the present invention is a more specific example of the eighth aspect, wherein, for example, a calculating means for calculating the number of connected components, and the number of connected components calculated by the calculating means is a predetermined number. Identification means for identifying that the value is equal to or smaller than the threshold value.

Here, the calculation of the number of connected components by the calculation means is, for example, an operation of calculating an independent black printing portion in a black-and-white inversion area. It is a configuration to do. On the other hand, the discriminating means is a means for discriminating whether the number of connected components calculated by the calculating means is equal to or less than a predetermined threshold value. Thus, the cell to which such a pattern is added and the black-and-white inverted cell are reliably identified.

According to a tenth aspect of the present invention, the region extracting means extracts an isolated ruled line candidate, and the area extracting means extracts the black and white around the isolated ruled line candidate. When there is an inversion candidate area, an isolated ruled line integrating means for integrating the isolated ruled line candidate into a black and white inversion candidate area is provided.

That is, as a result of integrating a region having a high black pixel density in a direction orthogonal to the mask scanning direction, if the line width of the region is equal to or greater than a predetermined threshold, the region is extracted as a black-and-white inversion candidate region. However, when the black-and-white inversion candidate area exists around the ruled line candidate, it is originally a black-and-white inversion candidate area,
For example, a ruled line candidate may be erroneously extracted as an isolated ruled line candidate due to “blurring” or the like. Therefore, in order to remedy such a case, it is integrated into the black-and-white inversion candidate area by the isolated ruled line integrating means.

With this configuration, even if there is a faint black print such as "fading" in the black-and-white inversion area, it can be reliably detected, corrected, and identified as a black-and-white inversion area.

The eleventh aspect of the present invention makes the above-described tenth aspect more concrete. The extraction of the isolated ruled line candidate includes means for distinguishing a character region from a ruled line region, and a ruled line extracting process. In this configuration, a black-and-white inversion area is extracted later.

That is, the ruled line extracted by the ruled line extraction may have a character string or the like erroneously recognized as a ruled line in addition to the original ruled line. This is for extracting ruled lines.

The twelfth aspect of the present invention makes the invention of the first aspect more concrete, wherein the line segment segment row has an attribute indicating that the line segment row is a black-and-white inversion area. Is added.

The addition of the attribute to the line segment segment sequence is as follows.
For example, this is realized by setting a dedicated flag in a memory storing the line segment sequence and storing the segment sequence indicating the black-and-white inversion area.

According to a thirteenth aspect of the present invention, the position of the line segment segment with respect to the black-and-white inversion area is vertically or horizontally shifted. Is added as position information of the side of.

For example, in the case of two line segment segments in the horizontal direction, upper or lower position information is added to the line segment segment depending on the position with respect to the black-and-white inversion area. In the case of two line segment segments in the vertical direction, right or left position information is added depending on the position with respect to the black-and-white inversion area.

With this configuration, the position of the created line segment line with respect to the black-and-white inversion area becomes clear, and, for example, when the black-and-white inversion area is distinguished from the normal cell area, it can be easily performed.

The description of claim 14 makes the description of claim 13 more concrete. When analyzing the cell structure of the table, the position information of the cell is added to the line segment sequence. If the position information is the same as the position information of the upper and lower or left and right sides, it is determined that the region is a black-and-white inversion region.

This embodiment is a specific application example of the thirteenth aspect, and the above configuration makes it possible to easily distinguish between a black-and-white inversion area and a normal cell area. In addition to the above, the vertical and horizontal position information added to the line segment segment row can be used, for example, for calculating the number of black-and-white inversion areas and the size of black-and-white inversion areas included in the table.

[0045] Claim 15 is based on claim 1 or 2 above.
In a case where the black-and-white inversion area includes a plurality of cell areas, a ruled line connected to the black-and-white inversion area is extended to extract the cell area in the black-and-white inversion area. Configuration.

For example, when the black-and-white inversion area is formed in the horizontal direction, the vertical ruled line is extended into the black-and-white inversion area, and the black-and-white inversion area is divided into, for example, strips and separated by the vertical ruled line. The region that has been set is a cell region.

With this configuration, the black-and-white inversion area can be easily divided into cells. According to a sixteenth aspect of the present invention, when the cells in the black-and-white inversion area are divided, an area obtained by extending a ruled line connected to the black-and-white inversion area is subdivided. And a means for determining a black pixel density.

That is, when dividing the black-and-white inversion area into cells as described in claim 15, for example, when the vertical ruled line is extended into the black-and-white inversion area and the black-and-white inversion area is divided into strips, the ruled line is divided into strips. If there is a black-and-white inverted character at the position of the extended black-and-white inverted area, this position cannot be separated by ruled lines, so this processing is performed to confirm the presence of the black-and-white inverted character.

Therefore, for example, a line segment extraction mask that is long in the vertical direction is used, and the number of black pixels is extracted in the target black-and-white inversion area using the line segment extraction mask. If the number is large, it can be seen that there is a black-and-white inverted character (white character) at the position, and the position cannot be divided into cells.

With such a configuration, the inside of the black-and-white inversion area can be accurately divided into cells. Claim 17
In order to solve the above problems, the described invention scans binary image data in table format in the horizontal direction using a line segment extraction mask that is long in the horizontal direction, and vertically scans the binary image data using the line segment extraction mask that is long in the vertical direction. Region extracting means for scanning in the direction, and extracting regions in which the black pixel density is higher than a predetermined threshold, and integrating the high black pixel density regions extracted by the region extracting means in directions orthogonal to the respective mask scanning directions. If the line width of the area integrated by the first integrating means is equal to or greater than a predetermined threshold, the area is determined to be a black-and-white inversion candidate area;
Second integrating means for integrating the black-and-white inversion candidate areas in the respective mask scanning directions; and horizontal and vertical line segment segments for the outline of the black-and-white inversion candidate areas integrated by the second integration means. A line segment segment sequence creating means to be created, and a second region for setting an area surrounded by the horizontal line segment segment sequence and the vertical line segment segment sequence created by the line segment segment sequence creating means as a black-and-white inversion area. A black-and-white inversion area identifying unit having setting means; a black-and-white inversion processing unit for performing black-and-white inversion processing on the black-and-white inversion area identified by the black-and-white inversion area recognition unit; This can be achieved by providing a table recognizing device having a recognizing means for performing a recognizing process on a character string located in an area.

The present invention is a table recognizing apparatus using the black-and-white inversion area discriminating apparatus according to claim 1. In other words, using a line segment extraction mask that is long in the horizontal direction and a line segment extraction mask that is long in the vertical direction, regions where the black pixel density is higher than a predetermined threshold are respectively extracted, and for the region where the black pixel density is high,
Integrate in the direction orthogonal to the mask scanning direction, and further integrate in the mask scanning direction.For example, a pair of line segment segments is formed on the contour of the black and white reversal candidate area, and the area surrounded by the line segment row is black and white reversed. A recognizing means for performing black-and-white reversal processing by the black-and-white reversal processing means on the area set in the area, and further performing recognition processing on a character string positioned in the area reversed by the black-and-white reversal processing means; Is a table recognition device having

With this configuration, it is possible to provide a table recognizing device that performs black-and-white inversion processing on a correctly identified black-and-white inversion area and performs character recognition processing on a character string in the black-and-white inversion area. In this case, the character string that has been subjected to the black-and-white inversion processing is accurately printed in black, so that a reliable table recognition processing can be performed.

According to an eighteenth aspect of the present invention, in order to solve the above problem, binary image data in a table format is scanned in the horizontal direction using a line segment extraction mask that is long in the horizontal direction, and the line segment that is long in the vertical direction is scanned. Scan vertically using the extraction mask,
A region extracting means for extracting regions having black pixel densities higher than a predetermined threshold value, and a first integration for integrating the regions having high black pixel densities extracted by the region extracting means in directions orthogonal to the respective mask scanning directions. Means and a second integration for integrating the black-and-white inversion candidate area in the respective mask scanning directions if the line width of the area integrated by the first integration means is equal to or greater than a predetermined threshold value Means and the second
A black-and-white inversion area identification unit including black-and-white inversion area setting means for setting a black-and-white inversion area based on the overlapping state of the horizontal and black-and-white inversion candidate areas integrated by the integration means; Black-and-white inversion processing means for performing black-and-white inversion processing on the black-and-white inversion area identified by the area recognition unit; and recognition means for performing recognition processing on a character string located in the area inverted by the black-and-white inversion processing means. This can be achieved by providing a table recognition device having

According to the present invention, there is provided a table recognizing apparatus using the black-and-white inversion area discriminating apparatus according to claim 2.
7. A black-and-white inversion area identification device that sets a black-and-white inversion area based on the overlapping state of the horizontal and black-and-white inversion candidate areas integrated by the second integration means with respect to the table recognition apparatus described in 7 above. It is a table recognition device.

With this configuration, it is possible to provide a table recognizing apparatus that performs black-and-white inversion processing on a correctly identified black-and-white inversion area and performs character recognition processing on a character string in the black-and-white inversion area. In this case, the character string that has been subjected to the black-and-white inversion processing is accurately printed in black, so that it is possible to perform a reliable recognition processing.

The nineteenth aspect of the present invention embodies the invention according to the seventeenth aspect, wherein the black-and-white inversion processing means includes a painting-out means for painting out an outline segment of the black-and-white inversion area, and the outline segment row. It is configured to include processing means for performing black-and-white inversion on a black-and-white inversion area other than the above.

That is, in this embodiment, the outline portion of the black-and-white inversion area is painted out with, for example, black by the painting means and framed, and the black-and-white inversion area therein is subjected to inversion processing by the processing means.

A twentieth aspect of the present invention embodies the specifics of the nineteenth aspect, wherein the black-and-white inversion processing means has a cell area painting means for painting an area divided by cells.

Also in this case, the dividing line for dividing the cells in the black-and-white inversion area is painted in black, for example, so that the cell dividing line can be recognized when the black-and-white inversion area is inverted.

According to a twenty-first aspect of the present invention, the line width of a character string positioned in a black-and-white inversion area, which has been inverted by the black-and-white inversion processing means, is specified. The character string is expanded or reduced according to the calculated line width, and the character string is shaped.

With such a configuration, an abnormality in the line width of a character string that can occur due to the black-and-white inversion processing is corrected, and the character string is shaped into an accurate character string. The invention of claim 22 embodies the invention of claim 17 or 18, wherein the monochrome inversion area setting means sets the monochrome inversion area when the number of connected components is equal to or less than a predetermined threshold. This is a configuration for performing processing.

This embodiment has a configuration corresponding to the above-described claim 8, and the configuration is applied to a table recognition device.
That is, when setting the black-and-white inversion area, the number of connected components is calculated, and when the number of connected components is within a predetermined threshold value, the black-and-white inversion area cell can be reliably identified by determining the black-and-white inversion area. On the other hand, it is possible to provide a table recognizing device that can perform a recognition process such as character recognition in a state where the reversal process has been accurately performed.

The invention according to claim 23 embodies the invention according to claim 17 or 18, wherein the area extracting means extracts an isolated ruled line candidate and places the isolated ruled line candidate around the isolated ruled line candidate. When a black-and-white inversion candidate area is present, an isolated ruled line integrating means is also provided for integrating the isolated ruled line candidate into the black-and-white inversion candidate area.

This embodiment has a configuration corresponding to the above-mentioned claim 10 and is applicable to a table recognition device. That is, as a result of integrating a region having a high black pixel density in a direction orthogonal to the mask scanning direction, if the line width of the region is equal to or less than a predetermined threshold, the region is extracted as a ruled line candidate, and the black-and-white inversion candidate is surrounded around the ruled line candidate. When an area exists, it is set as a black-and-white inversion candidate area. With this configuration, even if it is determined that the ruled line is an isolated ruled line candidate due to, for example, "blurring," the isolated ruled line integrating means integrates it into the black-and-white inversion candidate region, and performs the inversion process on the accurate black-and-white inverted region. It is possible to provide a table recognizing device that can perform a recognition process such as character recognition in a state where the recognition is performed.

According to a twenty-fourth aspect of the present invention, the line segment segment sequence includes, for example, an attribute indicating that the line segment segment sequence is a black-and-white inversion area. Is added.

This embodiment has a configuration corresponding to the above-mentioned claim 12 and is applicable to a table recognition device. That is, the line segment line corresponds to the black-and-white inversion area, and the position of the line segment line is added by adding an attribute such as up / down or left / right as a position attribute to the black-and-white inversion region to the line segment line. Is clear, the distinction between the black-and-white inverted cell and the normal cell is simplified, and a table recognition device that facilitates the table recognition processing can be provided.

A twenty-fifth aspect of the present invention embodies the invention according to the seventeenth or eighteenth aspect. When the black-and-white inversion area includes, for example, a plurality of cell areas, a connection to the black-and-white inversion area is made. The cell area in the black-and-white inversion area is extracted by extending the ruled line to be used.

This embodiment has a configuration corresponding to the above-described claim 15 and is applicable to a table recognition device. That is, when there are a plurality of cell areas in the black-and-white inversion area, the black-and-white inversion area is divided into a plurality of cells by using a ruled line whose extension is connected in a strip shape to the black-and-white inversion area.

With this configuration, even when a plurality of cell areas exist in one black-and-white inversion area, the cell area can be surely divided, and the correctly divided cell area can be recognized. A table recognition device capable of performing processing is provided.

The invention according to claim 26 embodies the invention according to claim 25, wherein, when dividing the cells in the black-and-white inversion area, for example, an area in which a ruled line connected to the black-and-white inversion area is extended Is subdivided to obtain a black pixel density.

This example describes one method for dividing a cell in a black-and-white inversion area. According to this method, a table recognizing device that prevents a character string in a black-and-white inversion area from being erroneously divided into cells. Can be provided. The method of dividing the inside of the black-and-white inversion area into cells is not limited to the above example.

According to a twenty-seventh aspect of the present invention, in order to solve the above-mentioned problem, the binary image data in the table format is scanned in the horizontal direction using a line segment extraction mask that is long in the horizontal direction, and the line segment that is long in the vertical direction is scanned. Scan vertically using the extraction mask,
A function of extracting regions in which the black pixel density is higher than a predetermined threshold value, a function of integrating the regions in which the black pixel density is high extracted by the function in a direction orthogonal to the respective mask scanning directions, and a function of integrating the functions. If the line width of the area is equal to or more than a predetermined threshold, it is determined that the area is a black-and-white inversion candidate area, and the function of integrating the black-and-white inversion candidate area in each mask scanning direction and the outline of the black-and-white inversion candidate area integrated by this function The ability to create horizontal and vertical line segment segments,
The computer storing a program including an instruction for causing the computer to execute a function of setting a region surrounded by the horizontal line segment segments and the vertical line segment segments created by the function as a black-and-white inversion region is provided. This can be achieved by providing a readable storage medium.

According to the present invention, a program for performing the above processing is stored in a storage medium such as a hard disk, an IC memory card, a magnetic tape, a floppy disk, or an optical disk. When the program is executed, the program is read from the storage medium. Run.

According to a twenty-eighth aspect of the present invention, in order to solve the above-mentioned problem, the binary image data in the table format is scanned in the horizontal direction using a line segment extraction mask which is long in the horizontal direction, and the line segments which are long in the vertical direction are scanned. Scan vertically using the extraction mask,
A function of extracting regions in which the black pixel density is higher than a predetermined threshold value, a function of integrating the regions in which the black pixel density is high extracted by the function in a direction orthogonal to the respective mask scanning directions, and a function of integrating the functions. If the line width of the area is equal to or greater than a predetermined threshold, it is determined that the area is a black-and-white inversion candidate area, a function of integrating the black-and-white inversion candidate area in each mask scanning direction, and a horizontal black-and-white inversion candidate area integrated by the function. This can be achieved by providing a computer-readable storage medium storing a program including an instruction for causing a computer to execute a function of setting a black-and-white inversion area according to the overlapping state of the black-and-white inversion candidate areas in the vertical direction.

The present invention also has a configuration in which a program for performing the above processing is stored in a storage medium such as a hard disk, an IC memory card, a magnetic tape, a floppy disk, or an optical disk. When the program is executed, the program is read from the storage medium. Run.

Further, claim 29 specifically shows the meaning of the black-and-white inversion area according to the invention of claim 1 or 2, and the black and white of the black-and-white inversion area is formed by three elements of color. For example, various combinations such as white and red, white and blue, yellow and red, and yellow and green are conceivable.

Further, the description of claim 30 similarly specifically indicates the meaning of the black-and-white inversion area of the invention according to claim 17 or 18, and the black and white of the black-and-white inversion area is represented by three elements of color. It includes any two colors to be created, and various combinations such as white and red, white and blue, yellow and red, yellow and green, and the like can be considered.

[0078]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a system configuration diagram of a table recognition device including the black-and-white inversion area recognition device of the embodiment. This example uses m × n to extract a black-and-white inversion area in a table used for, for example, spreadsheet software or a form.
In this method, a region having a high black pixel density is locally obtained by using the line segment extraction mask described above, and a black-and-white inversion region existing locally can be extracted by integrating those regions.

In the figure, 1 is a central processing unit (CPU) for performing various processes, 2 is a program memory for storing programs executed by the CPU 1, and 3 is an image memory for storing image data in a bitmap format. , 4
Is a work memory used for image processing, 5 is a scanner for optically reading images, 6 is a memory for temporarily storing information read by the scanner 5, 7 is a dictionary file storing characteristics of each character image, 8 Is a display for displaying the recognition result, 9 is a printer for printing the recognition result, 10 is an interface circuit for the display 8 and the printer 9, 11 is a CPU 1, a program memory 2, an image memory 3, a work memory 4, a memory 6, and a dictionary file 7. , A bus connecting the interface circuit 10 and the driver 12, 13 a hard disk, 14 an IC memory card, 15 a magnetic tape, 16 a floppy disk, 1
Reference numeral 7 denotes an optical disk such as a CD-ROM.

The system of the present apparatus temporarily stores the image data read by the scanner 5 in the memory 6 and develops the image data in the image memory 3 in a bitmap format. Then, a pattern extraction process is performed on the binary image data copied from the image memory 3 to the work memory 4. Based on the result, a character image is cut out from the image data read by the scanner 5, the feature of the cut out character image is compared with the feature data stored in the dictionary file 7, and the recognition result is displayed on the display 8. Or, it outputs to the printer 9.

In this system, the extraction of the black-and-white inversion area is realized as a function of the CPU 1 which performs processing according to a program stored in the program memory 2. Here, the program for performing the black-and-white inversion area extraction processing is stored in the ROM of the program memory 2 in advance. Also, after loading the program for performing the black-and-white inversion area extraction processing from a storage medium such as the hard disk 13, the IC memory card 14, the magnetic tape 15, the floppy disk 16, or the optical disk 17 into the RAM of the program memory 2,
This program may be executed by the CPU 1.

Further, a program for performing the black-and-white inversion area extraction processing can be extracted from the communication network via the communication interface. For example, as a communication network connected to the communication interface 19, for example, a LAN (Local Area Network)
k), WAN (Wide Area Network)
k), the Internet, an analog telephone network, a digital telephone network (ISDN: Integral Service D)
wireless network such as digital network, PHS (personal handy system), and satellite communication.

Next, the processing shown in FIG. 2 is based on the program stored in the program memory 2 as described above.
This is a recognition process performed by the CPU 1. In FIG.
1 performs the processing of the image input 20 first, and performs the input processing of the image data (image data) stored in the image memory 3 as described above. Next, a process of line segment extraction 21 for sequentially extracting image data using a line segment extraction mask is performed. For example, since this line segment extraction mask is a dot extraction mask with vertical and horizontal dimensions of m × n and is a mask for extracting line segments, when extracting a horizontal line segment, it is long in the horizontal direction (for example, 1 × n). When a vertical line segment is extracted using a mask of 6, 1 × 30, etc., it is long in the vertical direction (for example, 6 × 1, 3).
0 × 1). Then, the image data is sequentially cut out using the line segment extraction mask that is long in the horizontal direction or the line segment extraction mask that is long in the vertical direction, and the number of black pixels included in the cut out data is a fixed threshold value (threshold value). ) If it is greater than or equal to, it is all assumed to be a black pixel component line segment.

Next, the process of black-and-white inversion area extraction 22 is performed. This process integrates the horizontally long line segment or the vertically long line segment in the above-described line segment extraction 21 process, and determines whether the integrated region is a ruled line or a black-and-white inverted region. Extract the black and white reversal area. A predetermined threshold value is used for this determination, and an integrated line segment having a line width equal to or larger than the predetermined threshold value is set as a target line segment in the black-and-white inversion area.

Next, in the cell area extraction 23, a predetermined line segment among the line segments extracted in the line segment extraction process 21 and leaked to the black-and-white inversion area extraction 22, that is, the integrated line segment is extracted. An integrated line segment having a line width equal to or smaller than the threshold value is set as a ruled line candidate, and a cell region extraction process is performed.

Next, the processing of the in-cell item extraction 24 is performed by inverting the black-and-white inversion of the items in the cell area extracted by the processing of the cell area extraction 23 and the black-and-white inversion area extracted by the processing of the black-and-white inversion area extraction 22. Then, an in-cell item extraction process is performed on, for example, the character string after the inversion process.

Finally, the character recognition 25 is performed on the item data in the cell. In this processing, for example, the character (bitmap data) in the extracted item is compared with the characteristic data of the character registered in the dictionary 7 in advance, so that the character data whose characteristics match or the characteristic is most approximated. The character data to be recognized is a recognition character.

FIG. 3 is a system configuration diagram specifically explaining the above processing. Note that, in FIG.
Performs the processing of the image input 20 described with reference to FIG. 2 described above.
Is performed. Also, the line segment integration unit 33, ruled line extraction unit 34, black-and-white inversion area determination unit 35, black-and-white inversion area cell determination unit 36, and pixel inversion unit 37 shown in FIG. The corresponding parts.

The image area storage section 31 shown in FIG. 3 is a part of the area of the work memory 4 shown in FIG. <First Embodiment> The first embodiment will be described below using the above system configuration.

First, the flowchart shown in FIG. 4 explains the processing of this embodiment. Hereinafter, description will be given with reference to FIG. It is assumed that the above-mentioned image memory 3 has already read, for example, a document in the form of a table from the scanner 5 and stores the corresponding image data via the bus 11.

First, the CPU 1 reads out the image data stored in the image memory 3 and performs line segment extraction 22 and black-and-white inversion area extraction 23 processing. These processes are performed separately in the horizontal direction or the vertical direction. In the description of this example, the horizontal processing, which is the processing on the left side shown in FIG. 4, will be described.

First, horizontal line segment extraction processing (step (hereinafter referred to as S) 1) is performed using a line segment extraction mask that is long in the horizontal direction. This line segment extraction mask is mxn
However, for example, 1 × 6 (the line extraction mask is 1 × 6
Not only the mask of (1), but also a mask of 1 × 15, 1 × 30, etc. can be used).

FIG. 5 is a flowchart specifically explaining the above-described line segment extraction processing (S1). In this process, as shown in the figure, first, the corresponding 1 ×
6 is determined, and it is determined whether a value obtained by dividing the number of black pixels in the area by the mask size (the number of pattern black pixels in the mask / mask size) is larger than a preset threshold (S).
1-1). As a result, when the divided value is larger than the predetermined threshold value, the number of black pixels is large in the line segment cut by the line segment extraction mask (that is, the thin and relatively short line segment). (Black pixel) (S1-
1 is YES, S1-2). On the other hand, when the value obtained by dividing the number of black pixels in the mask by the mass size is smaller than the threshold value, it is determined that the number of black pixels is small in the region cut by the line segment extraction mask, and the cut region is not a line segment ( S1-1 is N
O).

FIG. 6 is a diagram specifically explaining the above-described processing. For example, a horizontal line extraction mask that specifies an area of pixels (dots) whose m × n (vertical × horizontal) is 1 × 6 is used. This is to explain a case where a directional component is extracted. FIG.
Indicates mask areas "1" to "9" overlapping each other, and a horizontal line segment is extracted by performing mask processing on the image data of FIG. Indicates that Through this processing, a large number of extremely thin and short regions (m × n is 1 × 6 line segments) are extracted in the horizontal and vertical directions.

Next, a line segment integration process (S2) shown in FIG. 4 is executed. Specific examples of the line segment integration processing are the determination (S2-1) and the integration processing (S2-2) shown in FIG.
In addition, at the time of the integration processing, the black-and-white inversion area candidate extraction processing (S3) shown in FIG. 4 is also executed at the same time. Done). The line integration process is a process performed by the line integration unit 33 shown in FIG. That is, according to the above-described processing (S1) of FIG. 4, it is determined whether the upper and lower line segments or the left and right line segments are in contact with the extracted line segment (S2-1). (S2
-1 is YES), these are integrated (S2-2), and a rectangular area is created (S2-3).

Here, the line integrating unit 33 has a function as a first integrating unit and a function as a second integrating unit. First, as a function as the first integrating unit,
The vertical direction (vertical direction) of the line segments is integrated. That is, when a region (line segment) is cut out in the horizontal direction using the line segment extraction mask as described above, another region (line segment) perpendicular to the region (vertical direction) is integrated. In this case, the distance between the integrated regions (line segments) must be shorter than a predetermined threshold value (thd) (S3-1), and if the distance between the line segments is longer than the threshold value (thd), it is determined that they touch. do not do. In this case, because the distance between the line segments is too large, they cannot be regarded as the same line segment. This integration is performed when the lengths of the line segments are the same (S3-2), and when the lengths are different, partial integration is performed (S3-3).

By the above processing, the 1 × 6 region (line segment) in the horizontal direction extends in the vertical direction if the number of black pixels is equal to or more than a predetermined threshold value, and the region (line) integrated by the above processing is obtained. If the width of (minute) exceeds a predetermined threshold, it is determined as a black-and-white inversion candidate area. If it is determined that the area is the black-and-white inversion area,
Next, the integrated line segments are integrated in the mask scanning direction.

This process is the second function of the line segment integration unit 33. This integration process sequentially integrates the black-and-white inversion candidate regions adjacent in the mask scanning direction. Then, the black-and-white inversion candidate area extends in the mask scanning direction in this way, and the above-described rectangular area is formed (S2-3).

FIG. 8 and FIG. 9 are diagrams showing the results of the above-described line segment integration processing. FIG. 8 shows a case where there is no “blurring” or the like in the integrated area, and FIG. 9 shows a case where there is “blurring” or the like in the integrated area.

First, the example of FIG. 8 will be described. FIG. 8 shows a line segment obtained by integrating regions having a line width equal to or more than a threshold value (not less than a threshold value). That is, the line width obtained as a result of integrating in the vertical and horizontal directions using an mxn line segment extraction mask that is long in the horizontal direction is equal to or larger than the threshold value (threshold value (thd1) shown in the figure). Min, which is a black and white reversal candidate area. If the size of the mxn line segment extraction mask that is long in the horizontal direction is M, for example, the size relationship shown in FIG.

On the other hand, FIG. 9 is an example in which the integrated area includes a "blurred" portion as described above.
"" Is a line segment obtained by integrating regions having a line width equal to or larger than the threshold value (threshold value (thd1) shown in FIG. 9). These line segments are also line segments of the black-and-white inversion candidate region. The line segment of "-" is a state in which the portion corresponding to the "blurred" portion is cut away as compared with the line segment of FIG. The reason for this is that in the above-described integration processing, the integration line segment distance in the vertical direction was insufficient, and the integration processing in the vertical direction was not performed thickly (S3-1 (> thd)), resulting in a thin integrated line segment. Things.

Next, ruled line candidate extraction processing (S
Perform 4). This processing is performed by the ruled line extracting unit 3 shown in FIG.
4 is the process performed. FIG. 10 is a flowchart specifically illustrating this processing. First, an integrated line segment not included in the black-and-white inversion candidate area is extracted from the integrated line segments integrated by the above-described processing (S4-1). Next, it is determined whether the line width of the integrated line segment extracted by the extraction processing is smaller than a predetermined threshold value (thd1) (S4-2). Here, when the line width of the integrated line segment is narrower than a predetermined threshold value (thd1), it is determined as a ruled line candidate (S4-2).

Here, with reference to FIGS. 8 and 9 described above, when the threshold value (thd1) is the width shown in both figures, the integrated line segments “,” are ruled line candidates.
Also, in the case of FIG. 9, the integrated line segment, ', ",'
Is a ruled line candidate. 8 is different from FIG. 9 in that the ruled line candidate is present in FIG. 9 with “blurred.” The ruled line extraction process (S5) in FIG. The extracted ruled line candidates are integrated in the mask scanning direction to create a line segment long in the horizontal direction.

Next, the isolated ruled line extraction processing (S6) of FIG. 4 is performed. FIG. 11 is a flowchart specifically illustrating this processing. In this process, it is determined whether or not there is a black-and-white inversion candidate region within the threshold value (thd2) of the distance left and right or up and down with respect to the ruled line extracted by the above-described ruled line candidate extraction process (S6-1). . In the case of the example of FIG. 8 described above, the integrated line segment, ',' does not correspond to the above-mentioned isolated ruled line. That is, although the black-and-white inversion candidate area or the right side of the integrated line segment exists on the right side of the line segment, the black-and-white inversion candidate area does not exist on the left side of the line segment. In addition, the black and white inversion candidate area exists on the left side of the integrated line segment ',', but the black and white inversion candidate area does not exist on the right side of the line segment.

On the other hand, in the case of the example of FIG.
8 are the same as in the example of FIG. 8 described above, but are different in the case of the integrated line segment ". That is, the black and white reversal candidate regions", "exist on the left and right of the integrated line segment" (S6-1). Therefore, the integrated line segment "" is extracted as an isolated ruled line.

Next, the integration process (S7) of the isolated ruled line and the black-and-white inversion candidate region in FIG. 4 is performed. This process is a process performed by the black and white reversal area determination unit 35 shown in FIG. FIG. 12 is a flowchart specifically illustrating this processing. This processing is to calculate a rectangular area circumscribing the upper and lower, or left and right, black and white reversal candidate areas surrounding the isolated ruled line extracted in the above processing (S6) (S7-1). That is, as described above, the isolated ruled line is, for example, the integrated line segment of "" in FIG. Accordingly, processing is performed to integrate an isolated ruled line as shown by "" in FIG. Specifically, a rectangular area enclosed by a square in FIG. 9 is calculated.

The extraction of the black-and-white inversion area in the horizontal direction and the extraction processing of the ruled line area are completed according to the above description.
In addition, the above processing is performed in a black-and-white inversion area in the horizontal direction,
And the extraction process of the ruled line area, but the same is performed in the vertical direction.

That is, this is the processing on the right side of the flow chart shown in FIG. 4, in which the vertical line segment extraction processing (step (hereinafter referred to as S ') 1) is performed by using the vertical long line segment extraction mask and extracted. A line segment integration process is performed on the line segment thus obtained (S2 '). This specific processing is performed in accordance with the flowcharts of FIGS. 5 and 7 as described above. next,
A black-and-white inversion candidate area extraction process (S3 ') is performed to extract a black-and-white inversion candidate area, a ruled line candidate extraction process (S4'), and a ruled line extraction (S5 '). Further, extraction processing is performed on the isolated ruled line (S6 ′), and finally, integration processing of the black-and-white reversal candidate area in the vertical direction is performed (S7 ′). In each of the above-described processes, as in the case of the horizontal direction, FIGS.
Is executed in accordance with the specific flowchart of FIG.

With the above-described processing, the processing of extracting the black-and-white inversion area and the ruled line area in the vertical direction is also completed, and then the processing shifts to the integration processing of horizontal / vertical black-and-white inversion candidate areas (FIG. 4).
S8).

FIG. 13 is a flowchart specifically explaining this processing. FIG. 14 is a view for explaining the processing in accordance with the flowchart of FIG. First, FIG.
According to the flowchart of 3, a black-and-white inversion candidate area in the horizontal direction is extracted (S8-1). The black-and-white inversion candidate areas are ', ", and in Fig. 14. Next, a vertical black-and-white inversion area overlapping with the horizontal black-and-white inversion area is extracted (S8-2). This is shown in Fig. 14. Note that the black-and-white inversion candidate area in the vertical direction is an area extracted in the black-and-white inversion area extraction processing (S3 ') in Fig. 4.

Next, the area of the overlapping portion between the horizontal black-and-white inversion candidate area and the vertical black-and-white inversion candidate area is calculated (S8-3). That is, the area A in which the horizontal black-and-white inversion candidate area ′ and the vertical black-and-white inversion candidate area in FIG. 14 overlap, the area B in which the black-and-white inversion candidate area overlaps with the vertical black-and-white inversion candidate area, the horizontal black-and-white The respective areas of the area C where the inversion candidate area and the vertical black-and-white inversion candidate area overlap each other and the area D where the black-and-white inversion candidate area and the vertical black-and-white inversion candidate area overlap each other are calculated.
The areas of B, C, and D are added.

Next, the following calculation is performed using the area of the overlapped region calculated by the above processing. That is,
(Area of overlapping portion) / Area of black-and-white inversion area in horizontal direction is calculated, and this value and threshold value (thd
3) is compared (S8-4). When the area of the (overlapping portion) / the area of the black-and-white inversion region in the horizontal direction is larger than the threshold value (thd3) (≧ thd3),
A rectangular area where the horizontal black / white inversion candidate area and the vertical black / white inversion candidate area are circumscribed is calculated (S8-6). That is, in this case, the horizontal black-and-white inversion candidate area and the vertical black-and-white inversion area have a large (overlapping area) and may be a black-and-white inversion area. On the other hand, in the above determination (S8-4),
When the area of the (overlapping portion) / the area of the horizontal black-and-white inversion area is smaller than the threshold value (thd3) (<thd3), the horizontal black-and-white inversion candidate area and the vertical black-and-white inversion area overlap. This is a case where the area cannot be identified as a black-and-white inversion area.

However, in this case, the next judgment (S8-5)
Execute That is, contrary to the above, the area of the black-and-white inversion area in the vertical direction (the area of the overlapping part) / the vertical direction is calculated,
This value is compared with a threshold value (thd3). When the area of (overlapping portion) / the area of the black-and-white inversion region in the vertical direction is larger than the threshold value (thd3) (≧ thd3), the ratio of (overlapping portion area) to the vertical direction is large. Assuming that there is, a rectangular area circumscribing the black-and-white inversion candidate area and the vertical black-and-white inversion candidate area is calculated (S8-6). That is, in this case, the area of the horizontal black-and-white inversion candidate area and the vertical black-and-white inversion area (the area of the overlapping portion) is sufficiently large with respect to the area of the vertical black-and-white inversion candidate area. This is because it can be determined that the area may be identified as a reverse area.

When the rectangular region where the horizontal black-and-white inversion candidate region and the vertical black-and-white inversion candidate region are circumscribed by the above processing is calculated, a solid line region E surrounded by a square in FIG. 14 is obtained. That is, this area is a black-and-white inversion area identified from the horizontal and vertical black-and-white inversion candidate areas.

However, in the present embodiment, the black-and-white inversion area is not immediately determined from the result obtained in the integration processing of the horizontal / vertical black-and-white inversion candidate areas, but the process of calculating the connected component in the next black-and-white inversion area (see FIG. Step S9 in FIG. 4 is executed, and a process of determining a black-and-white inversion area is performed based on the result (S10 in FIG. 4).
Hereinafter, a specific description will be given.

First, the calculation of the connected components is executed according to the flowchart shown in FIG. 15. For example, the number of connected components is calculated by labeling the inside of the black-and-white inversion area (S9-1). That is, the calculation of the number of connected components is, for example, a process of attaching a label to an independent black print portion in a black-and-white inversion area and calculating the number of labels (number of connected components). Therefore, for example, when there is a halftone dot or the like in the black-and-white inversion area, the number of black dots (the number of labels and the number of connected components) has an extremely large value.

Next, a process of determining a black-and-white inversion area is performed (S
10). This determination process is performed by the process shown in FIG. 16, and the above-described connected component is converted to a predetermined threshold value (thd
4) If it is equal to or more than the threshold value, the black-and-white inversion candidate area is rejected, and if the connected component is equal to or smaller than a predetermined threshold value, the black-and-white inversion candidate area is determined as the black-and-white inversion area (S10-1).

That is, this is a process for identifying whether or not the number of connected components calculated as described above is equal to or less than a preset threshold value. For example, when a pattern such as a halftone dot is added to the black-and-white inversion area, this process is performed. The cells identified and provided with such a pattern are the original black-and-white inverted areas (black-and-white inverted cells)
And is excluded from the black-and-white inversion area.

Here, for example, in the case of the example shown in FIG. 17, the black printing portion is a total of seven black printing portions (for example, six) included in the character string "company name" and one black portion in the background. is there. Therefore, in this case, the threshold value (t
hd4) is “10”, the judgment (S10-1) is (≦
thd4), in this case, a black-and-white inversion area.

Next, a contour segment expression of the black-and-white inversion area determined by the above processing is performed (S11). This process is specifically performed according to the flowchart of FIG. In other words, paying attention to the extracted line segment located on the outer periphery of the black-and-white inversion area, the line width of the line segment is incremented in the direction outside the black-and-white inversion area (outside the radiation vector of the line segment) (S11).
-1). Then, the vertex coordinates of the rectangle of the line segment are stored (S
11-2).

More specifically, as shown in FIG. 19, the region (line segment) extracted by the horizontal and vertical line segment extraction masks may be inclined up to about 4 degrees due to the inclination of the original image. , The region (line segment) to be extracted is also shifted vertically and horizontally. Therefore, as described above, correction is performed to increase the line width of the line segment in the outward direction of the black-and-white inversion area.

Next, an attribute assigning process for the black-and-white inversion area is performed (S12). In this process, specifically, as shown in the flowchart of FIG. 20, the position information of the contour segment created in the above process in the black-and-white inversion area is stored as attribute data (S12-1). For example, in the above-described example of the line segment segments shown in FIG. 19, the line segment segments a1 to a4 located above the black-and-white inversion area are added with an attribute indicating the position “up” of the black-and-white inversion area. An attribute indicating the position of “right” in the black-and-white inversion area is added to the line segment segment rows b1 and b2, and the line segment segment rows c1 to c4 located below the black-and-white inversion area are “below” the black-and-white inversion area. Is added, and an attribute indicating the “left” position of the black-and-white inversion area is added to the line segment segment rows d1 and d2.

With such a configuration, the above-described attributes are added to the line segment segments surrounding the black-and-white inversion area. For example, these attributes are stored in the work memory 4 together with the position information of the line segment segments. Thus, when the table recognition processing is performed later, the line segment (column) is a segment (column) surrounding the black-and-white inversion area, and it can be easily determined from the attribute data which position of the black-and-white inversion area corresponds. .

Next, cell division within the black-and-white inversion area is performed (S13 in FIG. 4). This process is performed by the black / white inversion area cell determining unit 36 shown in FIG. That is, the inside of the black-and-white inversion area may be composed of a plurality of cells, and the inside of the black-and-white inversion area set as described above is divided into cells.

FIG. 21 is a flowchart specifically explaining this processing. First, the distance (threshold value (thd
5)) Rule lines separated by more than the above are extracted (S13-1). For example, when the black-and-white inversion area is formed in the horizontal direction, a vertical ruled line is extracted. Next, a strip area in which the ruled line is extended into the black-and-white inversion area is extracted (S13-2). That is, the vertical ruled line extracted as described above is extended into the black-and-white inversion area, and the black-and-white inversion area is divided into, for example, strips. Then, it is determined whether the white pixel density of the strip region is lower than a predetermined threshold (thd5) (S13-
3). That is, when the black-and-white inversion area is divided into strips, if a black-and-white inversion character exists at the position of the black-and-white inversion area where the ruled line is extended, this position cannot be separated by the ruled line. This is a process performed to confirm the above. Therefore, for example, a line segment extraction mask that is long in the vertical direction is used, and the number of black pixels is extracted in the target black-and-white inversion area using the line segment extraction mask. In this case, it can be seen that a black-and-white inverted character exists at the position, and the position cannot be divided into cells.

Here, if the white pixel density of the above-mentioned strip area is smaller than a predetermined threshold value (thd5), contour segmentation of the black-and-white inversion area of the strip area is performed (S1).
3-4). Then, the contour segment expression and attribute assignment of each of the above-described cell-divided black-and-white inversion areas are performed (S13-
5).

Finally, the pixel inversion processing of the black and white inversion area is performed (S14). This process is a process performed by the pixel inverting unit 37 in FIG. FIG. 22 is a flowchart for explaining this specific processing. First, a filling process is performed in the outline segment obtained by dividing the black-and-white inversion area (S14-).
1). Next, a black-and-white inversion process is performed on an area other than the outline segment of the black-and-white inversion area (S14-2). That is, the outline portion of the black-and-white inversion area is filled with, for example, black by the above-mentioned filling means to form a frame, and the inversion processing is performed only on the black-and-white inversion area. For example, referring to the example of FIG. 19, line segment segments a1 to a4,
b1, b2, c1 to c4, d1, and d2 are painted in black, and black and white inversion processing is performed on the black and white inversion area therein.

As described above, when the black-and-white inversion process is performed on the black-and-white inversion area, the white characters in the black-and-white inversion area are black-and-white inverted and black printed. FIG. 23 is a diagram showing a state after the inversion processing is performed by the above-described black-and-white inversion processing.

As described above, the black-and-white inverted region identification apparatus of this embodiment uses the horizontal long line segment extraction mask and the vertical long line segment extraction mask, sets the black-and-white inversion candidate region by integrated processing, and This is a device for identifying an inversion area, and capable of accurately identifying a black-and-white inversion area.

In the black-and-white inversion area discriminating apparatus of this embodiment, the number of connected components is calculated for determining the black-and-white inversion area, and when the number of connected components is equal to or smaller than a predetermined threshold value (thd4), the black-and-white inversion area is determined. However, the present invention does not necessarily require calculation of the number of connected components and determination processing based on the number of connected components.

In the description of the present embodiment, the processing (S9) in FIG. 4 is executed to set the black-and-white inversion area, and (area of the overlapping portion) / area of the black-and-white inversion area in the vertical direction is calculated. And the threshold value (thd3) are compared to set the black-and-white inversion area. However, the black-and-white inversion candidate areas in the horizontal direction and the vertical direction are surrounded by outlines, and the area is set as the black-and-white inversion area. Good.

Further, a horizontal line segment sequence is extracted, and vertical ruled line information is used for the line segment sequence to define an area partitioned by the horizontal line segment sequence and the vertical ruled line. It may be set in a black-and-white inversion area. Also, you do n’t have to use line segment columns,
The black-and-white inversion candidate area in the horizontal direction may be used as it is, and the black-and-white inversion candidate area in the horizontal direction may be separated by a vertical ruled line and set as a black-and-white inversion area.

FIG. 24 shows an example in which a black-and-white inversion area is set by the above-described processing. That is, FIG. 11A shows a horizontal black-and-white inversion candidate area, and a vertical ruled line ′,
The area separated by 'is set as a black-and-white inversion area. In another example shown in FIG. 3B, when the black-and-white inversion candidate areas in the horizontal direction have the same length, the two black-and-white inversion areas are recognized as black-and-white inversion areas. With such a configuration, the black-and-white inversion area can be accurately extracted.

On the other hand, contrary to the above, a vertical line segment sequence is extracted, and the horizontal ruled line information is used for the line segment segment sequence. May be set as a black-and-white inversion area. Further, it is not always necessary to use the line segment line, and the black-and-white inversion candidate area in the vertical direction may be used as it is, divided by horizontal ruled lines, and the black-and-white inversion area may be set. <Second Embodiment> Next, a second embodiment of the present invention will be described.

In this example, the black-and-white inversion area is extracted by the above-described black-and-white inversion area identification device, the area is subjected to black-and-white inversion processing, and the recognition processing is performed on the black-and-white inversion area and character strings in normal cells. And performs a recognition process on the entire table.

The processing of this embodiment is also executed by the CPU 1 in accordance with the table recognition processing program stored in the program memory 2. First, FIG. 25 is a diagram showing an example of a form document to be recognized by the table recognition device of this example. The form in this example is an employee list of a certain company, for example, in which items such as company name, gender of the employee, telephone number, name, charge, etc. are described. Have been.

The form document is read by the scanner 5, and the corresponding image data is stored in the image memory 3 via the bus 11. First, a line segment extraction process (S1) is performed using a line segment extraction mask that is long in the horizontal direction, and a line segment integration process is performed on the extracted line segments (S2). Note that, also in the vertical direction, after the processing in the horizontal direction or the simultaneous direction is completed, a line segment extraction process (S1 ′) is performed using a line segment extraction mask that is long in the vertical direction, and the extracted line segment is processed. A line segment integration process is performed (S2 '). The line segment extraction and line segment integration processing are performed according to the flowcharts of FIGS. 5 and 7, and a ruled line including a black-and-white inversion area cell or a normal cell is extracted.

FIG. 26 shows a table recognition result in a state after the above-described line segment extraction and line segment integration processes have been performed. As shown in the figure, there is an area where line segment integration processing cannot be performed sufficiently in a part of the black-and-white inversion area. That is, there is a portion where line segments cannot be reliably integrated into a part of the description of the “store name”, a part of the description of the “company name”, and the like in the item description. This location is indicated by T, U, V, and W in FIG. This portion is a portion where "blurring" or the like exists in the state of the original document shown in FIG.

Next, an isolated ruled line extraction process is performed as a process for the above-mentioned "blurred" portion (S6, S6).
6 ′), the “blurred” portions as described above are extracted as isolated ruled lines. FIG. 27 shows an output result in the case where, for example, the state of an isolated ruled line extracted by the above-described processing is output as a table recognition. As shown in the drawing, isolated ruled lines are formed at locations corresponding to T, U, V, and W described above. Note that these ruled lines are indicated by T ', U', V ', and W' in FIG.

The table recognizing device of this example corrects all the isolated ruled lines to the black-and-white inversion candidate regions by integrating the isolated ruled lines and the black-and-white inversion candidate regions (S
7). That is, the isolated ruled lines T ′, U ′, V ′,
In either case, there is a black-and-white inversion candidate area on the left and right of W ′, and the above-mentioned isolated ruled lines T ′, U ′, V ′, and W ′ are integrated into the black-and-white inversion candidate area on both sides by the above-described integration processing (S7). You. FIG. 28 shows a result of outputting a state in which the isolated ruled line is integrated into the black-and-white inversion candidate area by the above-described processing.

Next, integration processing of horizontal / vertical black-and-white inversion candidate areas (S8), calculation processing of connected components in black-and-white inversion areas (S9), and determination processing of black-and-white inversion areas based on the results (S10) are performed. After that, an outline segment for the determined black-and-white inversion area is created (S11). FIG. 29 is a diagram showing a state in which a contour segment is formed in the determined black-and-white inversion area.

FIG. 30 shows the result of table recognition and output after performing the cell division process (S13) in the black-and-white inversion area to extract the cell area included in the black-and-white inversion area. As shown in the figure, the cell division processing of this example is performed by, for example, applying a ruled line 4
The cell division is performed by extending 0 to 45 into the black-and-white inversion area, and the cell area in the black-and-white inversion area can be easily divided. Although there is no corresponding part in FIG. 30,
When dividing the cell area in the black and white inversion area, if there is a character string on the extension of the vertical ruled line, the number of black pixels is detected using the vertical line segment extraction mask, so it should not be a simple division Can be recognized.

After the black-and-white inversion area is extracted as described above, the black-and-white inversion area is subjected to inversion processing by executing the black-and-white inversion processing (S14). FIG. 31 shows an output result obtained by, for example, inverting the item area of “company name” in the black-and-white inversion area.

On the other hand, a character string that has been inverted by the above-described black-and-white inversion processing is likely to have an abnormal line width. That is,
White characters are not accurately printed in black due to black and white reversal processing.
For example, there are places where the line width is extremely narrow and places where the line width is wide.
Therefore, the table recognizing device of this example performs a character string shaping process to address this problem.

FIG. 32 is a flowchart for explaining the character string shaping processing after the pixel inversion processing. Hereinafter, a specific description will be given. First, the average line width in the horizontal direction of the pattern in the area is calculated (step (hereinafter, referred to as ST) 1). In this calculation, the counting of the number of dots is started from a position where a white dot changes to a black dot in the horizontal direction, and then the counting is ended at a position where the black dot changes to a white dot. Then, the number of dots counted during that time is defined as the horizontal line width. Next, the same is applied to the vertical direction, and the average line width in the vertical direction of the pattern in the area is calculated (ST2). This calculation also starts counting the number of dots from a position where a white dot changes to a black dot in the vertical direction, and ends counting at a position where a black dot changes to a white dot. Then, the number of dots counted during that time is defined as a vertical line width. Then, the average line width of the pattern is calculated for the result of performing the above processing a plurality of times ((horizontal + vertical))
/ 2) (ST3).

Thereafter, the average line width of the character (string) after the black-and-white inversion processing obtained as described above is compared with a predetermined threshold value (thw) (ST4). When it is determined that the average line width is equal to or smaller than a predetermined threshold value (thw) (ST4 is <thw), expansion processing is performed (ST4).
5). On the other hand, when it is determined that the average line width of the character (string) is equal to or greater than a predetermined threshold value (thw) (ST4 is>
thw), a contraction process is performed (ST6). That is, the average line width of the inverted character (column) is equal to the threshold value (th
w) If it is narrower, expansion processing is performed, if line width is thicker, contraction processing is performed, and characters (rows) after inversion processing are shaped.

As described above, after performing the shaping process on the character (string) after the inversion process, the cell region extraction 23
At this time, in the table recognition apparatus of this example, when the black-and-white inverted region is already extracted, the line segment segments corresponding to the ruled line extraction and the black-and-white inverted region are added with an attribute. It can be distinguished from a ruled line surrounding a cell or a ruled line surrounding a normal cell.

For example, FIG. 33 is a diagram specifically explaining the above contents. As described above, the vertical and horizontal attributes are added to the line segment line corresponding to the black-and-white inversion area, so that it is easy to determine which position in the black-and-white inversion area each line segment (column) corresponds to. For example, as shown in the figure, the area surrounded by the upper and lower attributes of the line segment row is a black-and-white inversion area, and the area surrounded by the line segment row with the upper and lower attributes reversed is a normal cell area. .

After the above processing, recognition processing is performed on the character string in each cell, and a table including the recognition result of the character string is output together with the form. FIG. 30 also shows the final output result of the table recognition device of this example.

[0150]

As described above, according to the present invention, the following effects can be obtained. That is, according to the black-and-white inversion area identification device of the present invention, even when a black-and-white inversion area exists in a part of one cell, the area can be accurately identified. Further, even when a large size inverted character exists in the black-and-white inverted region, the line segment extraction mask used in the present invention uses an extremely thin and relatively short mask, so that the black-and-white inverted region can be accurately identified. Furthermore, even when the front image is read with inclination, the black-and-white inversion area is identified by the integration processing while recognizing the number of black pixels in a narrow range.
It can be recognized as a black-and-white inversion area with an inclined state.

Further, according to one aspect of the present invention, even when, for example, “blurring” or the like exists in the black-and-white inversion area, the black-and-white inversion area can be accurately identified because it is repaired by the isolated rule processing.

Further, according to one aspect of the present invention, the position of the created line segment sequence with respect to the black-and-white inversion area is clarified, and, for example, when the black-and-white inversion area is distinguished from the normal cell area, it can be easily performed. Can be.

Further, according to one aspect of the present invention, the black-and-white inversion area is divided into, for example, strips, and the area partitioned by horizontal or vertical ruled lines is used as the cell area. Can be divided into cells.

According to the table recognizing device of the present invention, even when a black-and-white inverted area exists in a part of one cell, the inverted area can be accurately identified, and a large-size inverted character is included in the black-and-white inverted area. Is present, the area can be accurately identified, and even when the front image is read with inclination, the black-and-white inversion area can be accurately identified.

Further, according to the present invention, the processing program is stored in a storage medium such as a hard disk, an IC memory card, a magnetic tape, a floppy disk, or an optical disk, and read out as necessary to identify a black-and-white inversion area and to perform table recognition processing. It can be performed.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of a table recognition device including a black-and-white inversion area recognition device of the embodiment.

FIG. 2 is a diagram illustrating a recognition process performed by a CPU.

FIG. 3 is a system configuration diagram specifically illustrating processing performed by a CPU;

FIG. 4 is a flowchart illustrating a process according to the first embodiment;

FIG. 5 is a flowchart specifically illustrating a line segment extraction process (S1).

FIG. 6 illustrates a case where a horizontal component is extracted by a line segment extraction mask that specifies an area of pixels (dots) whose m × n (length × width) is 1 × 6.

FIG. 7 is a flowchart specifically illustrating a line segment integration process;

FIG. 8 is a diagram showing a result of a line segment integration process.

FIG. 9 is a diagram illustrating a result of a line segment integration process.

FIG. 10 is a flowchart of a ruled line extraction process.

FIG. 11 is a flowchart illustrating an isolated ruled line extraction process.

FIG. 12 is a flowchart specifically illustrating an integration process of an isolated ruled line and a black-and-white inversion candidate region.

FIG. 13 is a flowchart specifically illustrating a process of integrating horizontal / vertical black-and-white inversion candidate areas.

FIG. 14 is a diagram illustrating the integration processing of horizontal / vertical black-and-white inversion candidate areas.

FIG. 15 is a flowchart specifically illustrating a connected component calculation process.

FIG. 16 is a flowchart of determining a black-and-white inversion area.

FIG. 17 shows a total of seven black printing parts, including one black part in the background and independent black printing parts (for example, six) included in the character string “company name”.
It is a figure explaining the number of connected components.

FIG. 18 is a flowchart specifically illustrating a process of expressing a contour segment of a black-and-white inversion area.

FIG. 19 is a diagram illustrating an example in which an area (line segment) to be extracted is extracted with a shift;

FIG. 20 is a flowchart specifically showing an attribute assigning process for a black-and-white inversion area.

FIG. 21 is a flowchart of cell division in a black-and-white inversion area.

FIG. 22 is a flowchart illustrating a pixel inversion process in a black-and-white inversion area.

FIG. 23 is a diagram illustrating a state after the inversion processing is performed by the black-and-white inversion processing.

FIG. 24 is an example of setting a black-and-white inversion area.

FIG. 25 is a diagram showing an example of a form document to be subjected to table recognition for explaining the second embodiment.

FIG. 26 is a flowchart of a ruled line extraction process as an output result from the table recognition device of the present example when it is assumed that table recognition and output have been performed after the line segment extraction and line segment integration processes have been performed.

FIG. 27 shows an output result when the state of the isolated ruled line to be extracted is output as a table recognition.

FIG. 28 shows a result of outputting a state in which an isolated ruled line is integrated into a black-and-white inversion candidate area.

FIG. 29 is a diagram showing a state in which a contour segment is formed in the determined black-and-white inversion area.

FIG. 30 shows a result of performing table recognition output by performing a cell division process in a black-and-white inversion area to extract a cell area included in the black-and-white inversion area.

FIG. 31 is a diagram showing an output result obtained by inverting an item area of, for example, “company name” in a black-and-white inversion area.

FIG. 32 is a flowchart illustrating a character string shaping process after the pixel inversion process.

FIG. 33 is a diagram illustrating that a ruled line surrounding a black-and-white inverted area cell or a ruled line surrounding a normal cell can be easily distinguished.

FIG. 34 is a diagram illustrating a conventional table recognition system.

FIG. 35 (a) is a diagram illustrating the configuration of a run.
(B) is a diagram for explaining a black-and-white inversion region, and (c) is a diagram showing that all regions in the black-and-white inversion region are not black-and-white inversion regions.

[Explanation of symbols]

 Reference Signs List 1 Central processing unit (CPU) 2 Program memory 3 Image memory 4 Work memory 5 Scanner 6 Memory 7 Dictionary file 8 Display 9 Printer 10 Interface circuit 11 Bus 12 Driver 13 Hard disk 14 IC memory card 15 Magnetic tape 16 Floppy disk 17 Optical disk 19 Communication interface 20 Image input 21 Line segment extraction 22 Black-and-white inverted region extraction 23 Cell region extraction 24 Cell item extraction 25 Character recognition 30 Image input unit 31 Image region storage unit 32 Line extraction unit 33 Line integration unit 34 Rule line extraction unit 35 Black / white inversion area determination unit 36 Black / white inversion area cell determination unit 37 Pixel inversion unit 40-45 Ruled line

Claims (30)

[Claims] 1. A table format binary image data is scanned in the horizontal direction using a line segment extraction mask that is long in the horizontal direction, and is vertically scanned using a line segment extraction mask that is long in the vertical direction. Region extracting means for respectively extracting regions having a pixel density higher than a predetermined threshold; and first integrating means for integrating the regions having a high black pixel density extracted by the region extracting means in a direction orthogonal to the respective mask scanning directions. If the line width of the area integrated by the first integration means is equal to or greater than a predetermined threshold, the area is determined to be a black and white inversion candidate area, and the black and white inversion candidate area is integrated in each mask scanning direction. Line segment segment creating means for creating horizontal and vertical line segment segment strings for the outline of the black-and-white inversion candidate area integrated by the second integrating means; Tone reversal region identification apparatus characterized by comprising: a tone reversal region setting means for setting a region surrounded by horizontal line segments rows and vertical line segment row in tone reversal region, the. 2. The table-format binary image data is scanned in the horizontal direction using a line extraction mask that is long in the horizontal direction, and is scanned in the vertical direction using the line extraction mask that is long in the vertical direction. Region extracting means for respectively extracting regions having a pixel density higher than a predetermined threshold; and first integrating means for integrating the regions having a high black pixel density extracted by the region extracting means in a direction orthogonal to the respective mask scanning directions. If the line width of the area integrated by the first integration means is equal to or greater than a predetermined threshold, the area is determined to be a black and white inversion candidate area, and the black and white inversion candidate area is integrated in each mask scanning direction. And black-and-white inversion area setting means for setting a black-and-white inversion area based on the overlapping state of the horizontal and black-and-white inversion candidate areas integrated by the second integration means. That black-and-white inversion region identification device. 3. A ruled line determining means for determining that the area integrated by the first integrating means is a ruled line when determining that the line width of the area integrated by the first integrating means is equal to or smaller than a predetermined threshold value. The black-and-white inversion area identification device according to claim 1 or 2, further comprising: 4. A black-and-white inversion area identifying apparatus according to claim 3, further comprising a ruled line creating means for creating a ruled line by integrating the ruled lines determined by said ruled line determining means in respective mask scanning directions. 5. When the ruled line creating means creates a ruled line in the vertical direction, an area enclosed between the horizontal line segment segment created by the line segmented string creating means and the vertical ruled line is defined. 5. The black-and-white inversion area identification device according to claim 4, wherein the black-and-white inversion area is used. 6. When the ruled line creating means creates a horizontal ruled line, an area enclosed between the vertical line segment segment created by the line segment segmented creating means and the horizontal ruled line is defined. 5. The black-and-white inversion area identification device according to claim 4, wherein the black-and-white inversion area is used. 7. The black-and-white inverting region identification apparatus according to claim 1, wherein a margin for absorbing irregularities in the outline of the black-and-white inverting region is added to the line segment row. . 8. The black-and-white inversion area identification device according to claim 1, wherein said black-and-white inversion area setting means sets a black-and-white inversion area when the number of connected components is equal to or less than a predetermined threshold value. 9. A calculating means for calculating the number of connected components,
9. The black-and-white inversion area identification device according to claim 8, further comprising identification means for identifying that the number of connected components calculated by said calculation means is equal to or less than a predetermined threshold value. 10. An isolated ruled line integrating means for extracting an isolated ruled line candidate and integrating the isolated ruled line candidate into the black-and-white inverted candidate region when the black-and-white inverted candidate region exists around the isolated ruled-line candidate. The black-and-white inversion area identification device according to claim 1 or 2, further comprising: 11. The extraction of the isolated ruled line candidate comprises means for distinguishing a character region from a ruled line region, and extracting a black-and-white inversion region after the ruled line extraction processing.
The black-and-white inversion area identification device described in the above. 12. The black-and-white inversion area identification device according to claim 1, wherein an attribute indicating that the line-segment row is a black-and-white inversion area is added to the line segment segment row. 13. The monochrome image according to claim 12, wherein the position of the line segment segment with respect to the black-and-white inversion area is added to the line segment line as positional information of upper and lower or left and right sides. Reversal area identification device. 14. When analyzing the cell structure of the table, if the position information of the cell is the same as the position information of the upper and lower or left and right sides added to the line segment segment row, it is determined that the cell is a black-and-white inverted cell area. 14. The black-and-white inversion area identification device according to claim 13, wherein if different, a normal cell area is determined. 15. When the inside of the black-and-white inversion area includes a plurality of cell areas, a ruled line connected to the black-and-white inversion area is extended to extract the cell area in the black-and-white inversion area. Or the black-and-white inversion area identification device according to 2. 16. The method according to claim 15, further comprising, when dividing the cells in the black-and-white inversion area, obtaining a black pixel density by subdividing an area in which a ruled line connected to the black-and-white inversion area is extended. Black and white reversal area identification device. 17. For binary image data in a table format,
An area extraction that scans in the horizontal direction using a line extraction mask that is long in the horizontal direction and scans in the vertical direction using a line extraction mask that is long in the vertical direction, and extracts areas where the black pixel density is higher than a predetermined threshold value Means, first integration means for integrating the areas having a high black pixel density extracted by the area extraction means in a direction orthogonal to the respective mask scanning directions, and line width of the area integrated by the first integration means Is greater than or equal to a predetermined threshold value, it is determined that the black-and-white reversal candidate area is a black-and-white reversal candidate area, and a black-and-white reversal candidate area integrated by the second integration means. Line segment segment sequence creating means for creating horizontal and vertical line segment segment sequences for the contours of, and the horizontal line segment segment sequence and the vertical line segment segment created by the line segment segment sequence creating device. And a black-and-white inversion area setting unit for setting a region surrounded by the print sequence as a black-and-white inversion area, and a black-and-white inversion processing performed on the black-and-white inversion area identified by the black-and-white inversion area recognition unit. A table recognizing device comprising: a black-and-white inversion processing unit for performing; and a recognition unit for performing a recognition process on a character string located in an area subjected to the black-and-white inversion processing unit. 18. Binary image data in a table format
An area extraction that scans in the horizontal direction using a line extraction mask that is long in the horizontal direction and scans in the vertical direction using a line extraction mask that is long in the vertical direction, and extracts areas where the black pixel density is higher than a predetermined threshold value Means, first integration means for integrating the areas having a high black pixel density extracted by the area extraction means in a direction orthogonal to the respective mask scanning directions, and line width of the area integrated by the first integration means Is greater than or equal to a predetermined threshold, it is determined that the black-and-white inversion candidate area is a black-and-white inversion candidate area, and the horizontal integration of black and white in the respective mask scanning directions is performed by the second integration means. A black-and-white inversion area identification unit including black-and-white inversion area setting means for setting a black-and-white inversion area according to the overlapping state of the black-and-white inversion area and the black-and-white inversion area recognition unit; Black-and-white inversion processing means for performing black-and-white inversion processing on the inverted area, and recognition means for performing recognition processing on a character string located in the area inverted by the black-and-white inversion processing means. Table recognition device. 19. The black-and-white inversion processing unit includes a filling unit that fills an outline segment of the black-and-white inversion region, and a processing unit that performs black-and-white inversion of a black-and-white inversion region other than the outline segment row. 17
Table recognition device as described. 20. The table recognition apparatus according to claim 19, wherein said black-and-white inversion processing means has a cell area filling means for filling an area divided by cells. 21. A line width of a character string located in a black-and-white inversion area subjected to the black-and-white inversion processing means is calculated, and the character string is expanded or contracted in accordance with the calculated line width. The table recognition device according to claim 17, wherein the table recognition is performed. 22. The table recognition apparatus according to claim 17, wherein said black-and-white inversion area setting means sets a black-and-white inversion area when the number of connected components is equal to or less than a predetermined threshold value. 23. The area extracting means extracts an isolated ruled line candidate, and when the black-and-white inversion candidate area exists around the isolated ruled-line candidate, the isolated ruled-line candidate is also integrated into the black-and-white inversion candidate area. 19. The table recognizing device according to claim 17, further comprising ruled line integrating means. 24. The table recognition apparatus according to claim 17, wherein an attribute indicating that the line segment segment is a black-and-white inversion area is added to the line segment segment. 25. When the inside of the black-and-white inversion area is composed of a plurality of cell areas, a ruled line connected to the black-and-white inversion area is extended to extract the cell area in the black-and-white inversion area. Or the table recognition device according to 18. 26. The apparatus according to claim 25, wherein, when dividing the cells in the black-and-white inversion area, means for obtaining a black pixel density by subdividing an area obtained by extending a ruled line connected to the black-and-white inversion area is provided. Table recognition device. 27. For binary image data in table format,
A function of scanning in the horizontal direction using a long line segment extraction mask in the horizontal direction and a vertical direction using a long line segment extraction mask in the vertical direction, and extracting a region where the black pixel density is higher than a predetermined threshold value; A function of integrating the areas having a high black pixel density extracted by the function in a direction orthogonal to the respective mask scanning directions; and a black-and-white inversion candidate area if the line width of the area integrated by the function is equal to or more than a predetermined threshold. A function of integrating the black-and-white inversion candidate areas in the respective mask scanning directions; and a function of creating horizontal and vertical line segment segments for the outline of the black-and-white inversion candidate areas integrated by the function. And a function of setting a region surrounded by the horizontal line segment segments and the vertical line segment segments created by the function as a black-and-white inversion region, and an instruction for causing a computer to execute The computer-readable storage medium storing the program. 28. For binary image data in a table format,
A function of scanning in the horizontal direction using a long line segment extraction mask in the horizontal direction and a vertical direction using a long line segment extraction mask in the vertical direction, and extracting a region where the black pixel density is higher than a predetermined threshold value; A function of integrating the areas having a high black pixel density extracted by the function in a direction orthogonal to the respective mask scanning directions; and a black-and-white inversion candidate area if the line width of the area integrated by the function is equal to or more than a predetermined threshold. And a function of integrating the black-and-white inversion candidate areas in the respective mask scanning directions, and setting of a black-and-white inversion area based on an overlapping state of the horizontal and black-and-white inversion candidate areas integrated by the function. And a computer-readable storage medium storing a program including an instruction to cause a computer to execute the function. 29. The black-and-white inverted area discriminating apparatus according to claim 1, wherein the black and white shown in the black-and-white inverted area includes any two colors created by three elements of colors. 30. The table recognition apparatus according to claim 17, wherein the black and white shown in the black and white reversal area includes any two colors created by three elements of colors.