JP2017010187A - Image processing device and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device for accurately extracting separations, compared with a technique for extracting as separations, parts which are erected by short distance from a lower side frame line of a field frame line, in extracting separations for dividing a region surrounded by rule marks.SOLUTION: The image processing device is configured so that: first extraction means of the image processing device extracts rule marks and non-rule marks from an image; second extraction means extracts the non-rule marks as separations for dividing a region surrounded by the rule marks, when positions of the non-rule marks are arranged in a regular state, on the rule marks.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an image processing apparatus and an image processing program.

  It is an object of Patent Document 1 to provide a character recognition device and a method for detecting a character frame line that can detect the character frame line regardless of the line type of the frame line such as the inclination of the character frame line, a solid line, or a broken line. The frame line detection unit includes a field frame line detection unit, a character delimiter line detection unit, and a whisker line detection unit, and detects a character frame line according to its property, that is, a field frame line detection unit forms a field. A field border (field separator), which is a type of character border, is detected, and the character separator detector is a type of character border that constitutes a field based on the field separator detected by the field border detector. It is disclosed that a certain character delimiter line is detected, and that the beard line detection unit similarly detects a beard line that is a type of character frame line that constitutes a field on the basis of the field delimiter line. It has been.

  In Patent Document 2, it is an object to extract elements such as ruled lines and frames as accurately as possible, suppress unextraction and overextraction, and reduce manual figure input, and extract ruled lines from read form images. After extracting the ruled line frame from the ruled line information, the image data of the portion corresponding to the extracted ruled line frame is erased from the form image, the ruled line is extracted again from the generated frame erased image, and the extracted ruled line is further extracted. It is disclosed that image data of a corresponding portion is erased and a character string is extracted from the generated ruled line erased image.

JP 2001-155113 A JP 2000-172780 A

According to the prior art, a whisker is formed which consists of a vertical line that stands short from the lower border of the field border and divides the entry area for each character in the field.
However, even a non-ruled line that is not regularly arranged is detected as a separating beard line if it is short and upright from the lower frame line of the field frame line.
The present invention, when extracting a break for dividing a region surrounded by a ruled line, is more accurate than the technique of extracting a short upright from the lower border of the field border as a break. It is an object of the present invention to provide an image processing apparatus and an image processing program that can extract a break well.

The gist of the present invention for achieving the object lies in the inventions of the following items.
According to the first aspect of the present invention, when the positions of the non-ruled lines in the ruled lines are regularly arranged, the first extraction means for extracting ruled lines and non-ruled lines from the image, the region surrounded by the ruled lines is An image processing apparatus comprising: a second extraction unit that extracts the non-ruled lines as divisions for division.

  According to a second aspect of the present invention, when the second extraction means includes the non-ruled line at a position where the ruled line is divided into equal parts with a predetermined width as a minimum width, the non-ruled line is used as the delimiter. The image processing apparatus according to claim 1, wherein the image processing apparatus is extracted.

  According to a third aspect of the present invention, in the image processing apparatus according to the first or second aspect, the second extraction unit targets the non-ruled line whose height and width are in a predetermined range. is there.

  According to a fourth aspect of the present invention, the second extracting means includes a first non-ruled line exceeding the size as a delimiter among the non-ruled lines extracted by the second extracting means and other second non-ruled lines. 4. The image processing apparatus according to claim 1, wherein the first non-ruled lines are divided using the size of the second non-ruled lines.

  According to a fifth aspect of the present invention, when the printed character is already included in the area surrounded by the ruled line, the second extracting unit divides the ruled line part that forms an area other than the character. The image processing apparatus according to claim 1, wherein a non-ruled line is extracted as a delimiter for performing the processing.

  In a sixth aspect of the present invention, when a plurality of regions surrounded by ruled lines are arranged, the second extracting means uses a non-ruled line as a delimiter in another region as a delimiter in each region. 6. The image processing apparatus according to claim 1, wherein non-ruled lines are extracted.

  In the invention of claim 7, the computer is surrounded by the first extracting means for extracting ruled lines and non-ruled lines from the image, and when the positions of the non-ruled lines in the ruled lines are regularly arranged. This is an image processing program for functioning as a second extraction means for extracting the non-ruled lines as a partition for dividing the existing area.

  According to the image processing apparatus of claim 1, in the case of extracting a partition for dividing an area surrounded by a ruled line, a short upright from the lower frame line of the field frame line is extracted as a partition Compared with the technique to do, it can extract a delimiter with high precision.

  According to the image processing apparatus of the second aspect, when there is a non-ruled line at a position where the ruled line is equally divided with the predetermined width as the minimum width, the non-ruled line can be extracted as a delimiter.

  According to the image processing apparatus of the third aspect, it is possible to extract a break for a non-ruled line whose height and width are in a predetermined range.

  According to the image processing apparatus of the fourth aspect, even when a non-ruled line other than a break is in contact with a non-ruled line that is a break, the break can be extracted by dividing it.

  According to the image processing apparatus of the fifth aspect, even when a printed character is already included in the area surrounded by the ruled line, the break can be extracted.

  According to the image processing device of claim 6, when a plurality of regions surrounded by ruled lines are arranged, non-ruled lines as delimiters in individual regions are extracted using non-ruled lines as delimiters in other regions. can do.

  According to the image processing program of claim 7, in the case of extracting a partition for dividing an area surrounded by a ruled line, a short upright from the lower frame line of the field frame line is extracted as a partition Compared with the technique to do, it can extract a delimiter with high precision.

It is a conceptual module block diagram about the structural example of this Embodiment. It is explanatory drawing which shows the system configuration example using this Embodiment. In this Embodiment, it is explanatory drawing which shows the example of the image made into object. It is explanatory drawing which shows the example which extracted the ruled line and the non-ruled line in this Embodiment. It is explanatory drawing which shows the example at the time of character-recognizing a non-ruled line part. It is a flowchart which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is a flowchart which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is a flowchart which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is explanatory drawing which shows the process example by this Embodiment. It is a block diagram which shows the hardware structural example of the computer which implement | achieves this Embodiment.

Hereinafter, an example of a preferred embodiment for realizing the present invention will be described with reference to the drawings.
FIG. 1 shows a conceptual module configuration diagram of a configuration example of the present embodiment.
The module generally refers to components such as software (computer program) and hardware that can be logically separated. Therefore, the module in the present embodiment indicates not only a module in a computer program but also a module in a hardware configuration. Therefore, the present embodiment is a computer program for causing these modules to function (a program for causing a computer to execute each procedure, a program for causing a computer to function as each means, and a function for each computer. This also serves as an explanation of the program and system and method for realizing the above. However, for the sake of explanation, the words “store”, “store”, and equivalents thereof are used. However, when the embodiment is a computer program, these words are stored in a storage device or stored in memory. This means that control is performed so as to be stored in the apparatus. Modules may correspond to functions one-to-one, but in mounting, one module may be configured by one program, or a plurality of modules may be configured by one program, and conversely, one module May be composed of a plurality of programs. The plurality of modules may be executed by one computer, or one module may be executed by a plurality of computers in a distributed or parallel environment. Note that one module may include other modules. Hereinafter, “connection” is used not only for physical connection but also for logical connection (data exchange, instruction, reference relationship between data, etc.). “Predetermined” means that the process is determined before the target process, and not only before the process according to this embodiment starts but also after the process according to this embodiment starts. In addition, if it is before the target processing, it is used in accordance with the situation / state at that time or with the intention to be decided according to the situation / state up to that point. When there are a plurality of “predetermined values”, they may be different values, or two or more values (of course, including all values) may be the same. In addition, the description having the meaning of “do B when it is A” is used in the meaning of “determine whether or not it is A and do B when it is judged as A”. However, the case where it is not necessary to determine whether or not A is excluded.
In addition, the system or device is configured by connecting a plurality of computers, hardware, devices, and the like by communication means such as a network (including one-to-one correspondence communication connection), etc., and one computer, hardware, device. The case where it implement | achieves by etc. is also included. “Apparatus” and “system” are used as synonymous terms. Of course, the “system” does not include a social “mechanism” (social system) that is an artificial arrangement.
In addition, when performing a plurality of processes in each module or in each module, the target information is read from the storage device for each process, and the processing result is written to the storage device after performing the processing. is there. Therefore, description of reading from the storage device before processing and writing to the storage device after processing may be omitted. Here, the storage device may include a hard disk, a RAM (Random Access Memory), an external storage medium, a storage device via a communication line, a register in a CPU (Central Processing Unit), and the like.

  The image information processing apparatus 100 according to the present embodiment extracts a break for dividing a region surrounded by ruled lines. As shown in the example of FIG. A detection module 120, a cell configuration detection module 130, a non-ruled line element connection calculation module 140, and a non-line segment break determination module 150 are included.

  The data reading module 110 is connected to the line segment detection module 120. The data reading module 110 reads an image and passes the image to the line segment detection module 120. Reading an image means, for example, reading an image with a scanner, a camera, etc., receiving an image from an external device via a communication line by fax or the like, a hard disk (in addition to what is built in a computer, via a network). And the like, and the like read out the images stored in the device etc.). The image may be a binary image or a multi-value image (including a color image). One image may be read or a plurality of images may be read. Further, the content of the image may be, for example, a document used for business like a form, a pamphlet for advertisement, or the like. Note that character frame lines are printed in the original from which the image is read. The character frame line is used for entering characters by printing or handwriting so as to fit within the character frame line. In addition, characters may be printed or entered (hereinafter referred to as entry) in the character frame line of the read image, or may be left blank without being entered. When a character is entered, for example, the character is recognized in post-processing of the image information processing apparatus 100. If no character is entered, it is used for template creation in post-processing of the image information processing apparatus 100, for example.

An area surrounded by character frame lines is an area surrounded by ruled lines. In many cases, a partition for dividing the area is added to define the position of the character. When this delimiter is a solid line, it is possible to separate a ruled line (straight line segment) including the delimiter and a pixel block other than the ruled line (non-ruled line, for example, a character) by a conventional technique. Note that the pixel block is a pixel region that is continuous in four or eight connections.
However, as shown in the example of FIG. 3, there is a delimiter (also referred to as a cell delimiter or a whisker) that is a short line extending in the opposite direction (lower or upper) from the ruled line on the upper side or the lower side. In the above-mentioned Patent Document 1, it is expressed as “short upright from the lower frame line of the field frame line”. In the case of horizontal writing, the separation extends in the opposite direction from the ruled line on the upper side or the lower side, and in the case of vertical writing, the separation extends in the opposite direction (right or left) from the ruled line on the left or right side.
The image information processing apparatus 100 mainly targets such a break.

The line segment detection module 120 is connected to the data reading module 110, the cell configuration detection module 130, and the non-ruled line element connection calculation module 140. The line segment detection module 120 extracts a line segment. This line segment is a candidate for a ruled line that becomes a character frame line. For example, a conventional technique for extracting a straight line segment may be used.
The cell configuration detection module 130 is connected to the line segment detection module 120, the non-ruled line element connection calculation module 140, and the non-line segment break determination module 150. The cell configuration detection module 130 detects the configuration of ruled lines, which are character frame lines, for the line segments extracted by the line segment detection module 120. For example, a ruled line structure constituted by horizontal and vertical line segments is analyzed, and cell components of a character frame are detected. A table that is a set of character frames may be included as an analysis target. The analysis includes extracting the coordinates of the start point and end point (end point of the line segment) of the ruled line. Further, the thickness, color, etc. of the line segment may be extracted.

The non-ruled line element connection calculation module 140 is connected to the line segment detection module 120, the cell configuration detection module 130, and the non-line segment separation determination module 150. The non-ruled line element connection calculation module 140 separates ruled lines that are character frame lines detected by the cell configuration detection module 130 and non-ruled lines that are pixel blocks other than the ruled lines. Note that the above-described breaks in the character frame are extracted as non-ruled lines.
Then, the position of the non-ruled line in the ruled line is calculated. That is, the position of the non-ruled line is extracted using the ruled line as a coordinate axis. For example, the position (X coordinate) of the non-ruled line may be extracted with the ruled line on the lower side of the character frame line as the X coordinate axis (for example, the left end is 0). In this case, a non-ruled line that is a block of pixels such as characters may be included in addition to the delimiters. Alternatively, only the coordinates of the non-ruled lines that are in contact with the ruled lines may be extracted. Also in this case, since characters or the like may be in contact with the ruled lines, non-ruled lines that are pixel blocks of characters or the like may be included in addition to the delimiters.

The non-line segment break determination module 150 is connected to the cell configuration detection module 130 and the non-ruled line element connection calculation module 140. When the positions of the non-ruled lines in the ruled lines are regularly arranged, the non-segment delimiter determining module 150 extracts the non-ruled lines as a break for dividing the area surrounded by the ruled lines. The position of the non-ruled line in the ruled line calculated by the non-ruled line element connection calculating module 140 is used. Among the positions, the regularly arranged ones are extracted as non-ruled lines that are separated.
The non-line segment break determination module 150 may extract a non-ruled line as a delimiter when there is a non-ruled line at a position where the ruled line is divided into equal parts with a predetermined width as the minimum width.

Further, the non-line segment break determination module 150 may target non-ruled lines whose height and width are in a predetermined range. Detailed processing will be described later using the example of FIG.
Further, the non-line segment delimitation determination module 150 classifies the non-ruled lines extracted by the above processing into a first non-ruled line exceeding the size as a delimiter and other second non-ruled lines, and the second The first non-ruled line may be divided using the size of the non-ruled line. Detailed processing will be described later using the examples in FIGS. 13 and 14.
In addition, the non-line segment delimiter determining module 150, when an already printed character is included in an area surrounded by a ruled line, is a delimiter for dividing a ruled line part that forms an area other than the character. As such, a non-ruled line may be extracted. Detailed processing will be described later using the example of FIG.
In addition, when there are a plurality of regions surrounded by ruled lines, the non-segment delimiter determination module 150 extracts non-ruled lines as delimiters in individual regions using non-ruled lines as delimiters in other regions. You may make it do. Here, “when a plurality of regions surrounded by ruled lines are arranged” may be a case where these regions are in contact (for example, when a table is formed by character frame lines), It may be a case where these areas are separated. Moreover, you may make it apply only when those area | regions are contacting. If these areas are separated from each other, conversely, a non-ruled line serving as a break may be extracted independently of the position of the break in other areas (irrelevant). Detailed processing will be described later using the example of FIG.

FIG. 2 is an explanatory diagram showing a system configuration example using the present embodiment.
The example shown in FIG. 2A is configured by the image reading device 210, the image information processing device 100, and the form processing device 230.
The image reading device 210 is connected to the image information processing device 100. The image reading device 210 is a scanner that reads a form in which character frames with delimiters are described.
The image information processing apparatus 100 is connected to an image reading apparatus 210 and a form processing apparatus 230. The image information processing apparatus 100 accepts an image read by the image reading apparatus 210, extracts a non-ruled line from the character frame, and makes it possible to identify the character portion in the character frame.
The form processing device 230 is connected to the image information processing device 100. The form processing device 230 recognizes characters in the character frame that can be identified by the image information processing apparatus 100, and performs various forms processing using the results.
Note that the image information processing apparatus 100 may be configured integrally with either the image reading apparatus 210 or the form processing apparatus 230, or the image reading apparatus 210, the image information processing apparatus 100, and the form processing apparatus 230 may be configured integrally. May be.

In the example shown in FIG. 2B, the image information processing apparatus 100, the image reading apparatus 210A, the image reading apparatus 210B, the image reading apparatus 210C, the form processing apparatus 230A, and the form processing apparatus 230B are connected via a communication line 290, respectively. Has been. The communication line 290 may be wireless, wired, or a combination thereof, and may be, for example, the Internet or an intranet as a communication infrastructure.
Here, the image reading apparatus 210 may be a scanner or a fax machine. The function of the image information processing apparatus 100 may be realized as a cloud service. For example, a form image is transmitted from the image reading apparatuses 210 installed at a plurality of bases, the image information processing apparatus 100 performs processing, and the form processing apparatus 230 performs integrated form processing on these forms. Also good.

FIG. 3 is an explanatory diagram illustrating an example of a target image in the present embodiment. An example of an image read by the data reading module 110 is shown. In this example, there are three delimiters on the upper and lower ruled lines so that four characters are written in the character frame. Note that the characters are handwritten numbers, and some of them are connected to the ruled lines or breaks on the lower side, or noise is generated.
FIG. 4 is an explanatory diagram showing an example in which ruled lines and non-ruled lines are extracted in the present embodiment. The processing result by the line segment detection module 120 is shown. In the example of FIG. 4A, only ruled lines are extracted, and in the example of FIG. 4B, only pixel blocks other than ruled lines (non-ruled lines) are extracted.
When the image shown in FIG. 4B is recognized as it is (the processing by the cell configuration detection module 130, the non-ruled line element connection calculation module 140, and the non-segment delimiter determination module 150 of this embodiment is not performed). The character recognition result is as shown in the example of FIG. That is, “.3.'7.'2” '9. . ”Is recognized. For example, the delimiter is “. '”, The noise is [. ”], And the number that overlaps the delimiter (originally [8]) is misrecognized as [9].

FIG. 6 is a flowchart showing an example of processing according to this embodiment. As a specific example, a processing example when the image shown in the example of FIG. 3 is targeted will be described.
In step S602, the data reading module 110 reads image data. An image as shown in the example of FIG. 3 is read.
In step S604, the line segment detection module 120 extracts a straight line segment. Ruled lines are extracted as shown in the example of FIG. Of course, as a result of extracting ruled lines, non-ruled lines are also extracted as shown in the example of FIG.

In step S606, the cell configuration detection module 130 extracts cell components. As shown in the example of FIG. 4A, ruled lines 710, ruled lines 720, ruled lines 730, and ruled lines 740 constituting a rectangle are extracted as shown in the example of FIG. For example, the coordinates of the start point and end point of each ruled line are extracted. And this example extracts that it consists of one character frame.
In step S608, the non-ruled line element connection calculation module 140 extracts a ruled line and a non-ruled line element and a connection relationship. In the figure, the relationship between the ruled line 720 (upper ruled line) and the ruled line 740 (lower ruled line) shown in the example of FIG. 7 and the non-ruled line shown in the example of FIG. 4B is extracted. As the non-ruled lines (non-ruled line elements 810 to 836), characters, delimiters, noises, and the like are mixed. The non-ruled lines that are in contact with the ruled line 720 are non-ruled line elements 810 to 818, and the lines that are in contact with the ruled line 740 are non-ruled line elements 820 to 828 and non-ruled line elements 832 to 836. No. “2”) is not in contact.
In step S610, the non-line segment delimiter determination module 150 extracts delimiter elements. The processing content of step S610 will be described using the flowchart shown in the example of FIG.

FIG. 9 is a flowchart illustrating a processing example according to the present exemplary embodiment (non-line segment break determination module 150).
In step S902, the coordinates of the non-ruled line elements in the ruled line section are extracted. An example is shown in FIG. The example shown in FIG. 10A is the same as that shown in FIG. 8, and shows an example of the relationship between the ruled line 720 and the non-ruled line element 810, and the relationship between the ruled line 740 and the non-ruled line element 820. The example shown in FIG. 10B shows the position of the non-ruled line element 820 and the like on the ruled line 740. For example, the position of each non-ruled line (non-ruled line element 820, etc.) is shown with the ruled line 740 as a coordinate axis (the right end is [600], the width W of the ruled line 740: 600) with the left end set to [0]. is there. Specifically, the non-ruled line element 820 is 50, the non-ruled line element 822 is 80, the non-ruled line element 824 is 150, the non-ruled line element 826 is 225, the non-ruled line element 828 is 300, the non-ruled line element 832 is 500, and the non-ruled line element 834 is at 570, and the non-ruled line element 836 is at 570. This numerical sequence is {Yj} j = 1 to N (8 in this example). The unit may be a pixel (number of pixels), mm or the like. Of course, the same is done for the ruled line 720 which is the upper side.

  In step S904, the non-ruled line elements regularly arranged are extracted. An example is shown in FIG. Let Tmin be the minimum character width. Tmin is a predetermined numerical value, for example, Tmin: = 40. The maximum value Smax that can divide the character frame (cell or ruled line 720) can be calculated based on Tmin and the length of the ruled line 740. In this example, Smax: = 15 (= 600/40). The minimum value of the number of divisions is Smin. Smin is a predetermined numerical value, for example, Smin: = 3. Smin can be set to 2 or more. Note that the minimum value of the number of divisions may be determined after the maximum value of the character width is determined. In this example, the following search process is performed in the range of 3 (Smin) to 15 (Smax).

The degree of fit is inspected from the minimum value (Smin) of the number of divisions for the coordinate sequence extracted in step S902.
With respect to the coordinate value sequence ({Xi (: = i * W / S)} i = 1 to S-1) assumed for the division number S, the coordinate value sequence of the actual data is within a predetermined range of neighborhood δ. Count how many points match with {Yj}. If the count number matches S-1, it is adopted. Note that {Xi} indicates a delimiter position in the case of the division number S. The actual coordinate value sequence {Yj} should include those that match {Xi} in various division numbers S, and a non-ruled line at the matching position may be extracted as a delimiter.
In the example of FIG. 11, since {Yj} does not exist at the position of {Xi} (range of neighborhood δ) when it is divided into three, it is determined that the ruled line 740 is not divided into three. Next, since there are all {Yj} at the position of {Xi} (range of neighborhood δ) when dividing into four, the ruled line 740 is divided into four parts, and the non-ruled line at that position is extracted as a part To do.

In the above-described example, the search order is performed in order from the smallest number of divisions, but may be performed in order from the largest number of divisions, or may be performed in order from the largest number of divisions (determined based on the length of ruled lines). The search may be started from the division number (specifically, a value obtained by dividing the length of the ruled line by the general character width).
Further, in the above-described example, the one with the count number matching S-1 is adopted, but after searching all, the one with the largest number of matches may be adopted.
Further, the ruled lines used here may be the ruled lines on the upper side or the lower side in the case of horizontal writing, and the ruled lines on the left side or the right side in the case of vertical writing.

FIG. 12 is an explanatory diagram showing a processing example according to the present embodiment. The size of the non-ruled lines connected to the ruled lines constituting the character frame (ruled lines 710 to 740 in the above example) is checked, and those that do not satisfy the conditions are deleted. Here, the width in the size check of the non-ruled lines refers to the direction of the ruled lines (horizontal direction in the example of FIG. 11), and the height refers to the connection direction with the ruled lines of the non-ruled lines (upward in the example of FIG. 11). Shall.
Elements that are not recognized as delimiters (that is, bearded protrusions) are deleted in the following processing example.
As shown in the example of FIG. 12, if the range in the two-dimensional space relating to the width and height is satisfied (in the example of FIG. 12, the region is a dark portion), it is deleted. In this example, a non-ruled line with a small width is basically left as a delimiter candidate, and it is determined that a larger width is not a delimiter candidate. For example, it can be expected that the extended noise run length generated during scanning or fax transmission or the like is not included as a candidate for delimiter.

Further, when the positions of the non-ruled lines are regularly arranged, it may be determined whether the non-ruled lines have the same size or the like. In other words, when the size is uniform, it is adopted as a delimiter. Then, the deviating non-ruled line may be corrected. This will be described with reference to the examples of FIGS.
FIG. 14 is an explanatory diagram showing a processing example according to the present embodiment. As described above, in this example, there are four divisions, and the non-ruled line element 824, the non-ruled line element 828, and the non-ruled line element 832 are extracted as delimiters. However, the delimiter of the non-ruled line element 832 is in contact with the number “8”, and if the non-ruled line element 832 itself is extracted as a delimiter, the number “8” is recognized as a recognition target in subsequent processing such as character recognition. Inconveniences such as not becoming will occur. Further, if the non-ruled line element 832 itself is recognized as a character, it may be erroneously recognized because the number “8” is in contact with the break.
Note that the circumscribed rectangles (circumscribed rectangle 1424, circumscribed rectangle 1428, circumscribed rectangle 1432) in FIG. 14 are drawn larger than the actual circumscribed rectangle for easy viewing.

FIG. 13 is a flowchart illustrating a processing example according to the present exemplary embodiment (non-line segment break determination module 150).
In step S1302, the size of the partial image recognized as a break is measured, and the isomorphism of the size, the original point or the line segmentality is calculated. For example, as shown in FIG. 14, the sizes of the circumscribed rectangle 1424 of the non-ruled line element 824, the circumscribed rectangle 1428 of the non-ruled line element 828, and the circumscribed rectangle 1432 of the non-ruled line element 832 are measured. As the isomorphism of the size, for example, a difference such as a width of the circumscribed rectangle, a height of the circumscribed rectangle, or a ratio of the width and the height of the circumscribed rectangle is within a predetermined range. Originally, the point or line segmentation means that the width of the circumscribed rectangle, the height of the circumscribed rectangle, the ratio of the width and height of the circumscribed rectangle, and the like are within a predetermined range.
In step S1304, an unexpected size is classified. That is, a non-ruled line that does not satisfy the isomorphism of the size, or a non-ruled line that does not satisfy the original point or line segmentation is extracted.

In step S1306, clustering is performed within a predetermined threshold range. For example, clustering is performed in a two-dimensional space with the width and height of a circumscribed rectangle of a non-ruled line within a predetermined range. Further, clustering by size difference may be performed.
In step S1308, a small-sized cluster is determined as a delimiter candidate.
In step S1310, the elements belonging to other clusters are decomposed so as to be equal to the size of the delimiter candidates. That is, it is divided into a separator and other character parts. Specifically, it is assumed using the size and positional relationship of non-ruled lines in a cluster as a delimiter candidate (difference between the position (Yi) where the non-ruled line is actually present and the position (Xi) where the delimiter is actually present). A process of excluding the pixel blocks in the positional relationship by the size of the candidate for separation from the non-ruled lines of the outer size may be performed. In the example of FIG. 14, the positional relationship between X1 and Y3 (the position of the non-ruled line element 824) and the positional relationship between X2 and Y5 (the position of the non-ruled line element 828) are the same as X3 and Y6 ( The pixel block of the original number “8” is extracted by deleting a portion having the same positional relationship as the non-ruled line element 832 and having the same size as the circumscribed rectangle 1424 from the circumscribed rectangle 1432.

FIG. 15 is an explanatory diagram showing a processing example according to the present embodiment. As shown in the example of FIG. 15A, a preprint character string (for example, “price”, “birth date”, “name”, etc.) is written in the character frame name or character frame. Name), it is assumed that the space excluding the character string area (the area where the character is described) is equally divided and whether it is a non-ruled line or not. May be.
When a preprint character string is included in the character frame, the character string may be implicitly expressed as an unfillable area. In these cases, the above-described processing cannot cope with them.
Therefore, when a preprint character string is detected, the following processing is performed.
(1) The preprint portion is detected.
An area having a non-ruled line having a size larger than a predetermined size in the character frame is extracted as a preprint portion. As shown in the example of FIG. 15B, the preprint unit 1510 is extracted.
(2) Set the right edge with a margin on the right side of the preprint section.
The preprint portion is expanded by adding a margin of a predetermined width to the right side of the preprint portion. As shown in the example of FIG. 15B, the right end 1520 of the preprint frame is set. It is assumed that there is a vertical ruled line at the position of the right end 1520 of the preprint frame.
(3) The separation extraction process is performed in the remaining area.
In the example of FIG. 15B, the separation extraction process is performed on the area 1530 other than the preprint.
Note that this processing example assumes that the layout of a blank form has been registered. If there is already a handwritten description, a process for distinguishing between the preprint portion and the handwritten description may be performed. For example, a distinction process between printed characters and handwritten characters may be performed. Specifically, when a preprint character is known, the preprint portion may be distinguished by pattern matching with the character.

FIG. 16 is an explanatory diagram illustrating a processing example according to the present embodiment. In the case of the tabular format, when a candidate for a break can be detected in a plurality of rows or columns, the break that can be detected in each row / column may be comprehensively determined.
In the case of a tabular document, as shown in the example of FIG. 16, there are cases where a multi-stage delimited line is designed. In this case, there are many entry fields with similar divisions in the upper and lower related cells.
Therefore, the result of segmentation extraction within each character frame is not used as it is, but the preferred number of divisions is comprehensively determined from the result of the character frame of many lines. This increases the accuracy.
The above-described delimiter extraction is performed on each line. However, the extraction result here is a delimiter candidate and not the final delimiter extraction result. For example, row 1610 is a candidate for n1 division, row 1620 is a candidate for n2 division, row 1630 is a candidate for ni division, and row 1640 is a candidate for nN division.
Then, the division number may be determined by majority decision of the division number. Further, as a method for determining the number of divisions, probability calculation using a ratio or the like may be performed.
Further, not only the number of divisions, but also the division position (separation position) and the division size may be unified for all character frames.

  A hardware configuration example of the image processing apparatus according to the present embodiment will be described with reference to FIG. The configuration shown in FIG. 17 is configured by a personal computer (PC), for example, and shows a hardware configuration example including a data reading unit 1717 such as a scanner and a data output unit 1718 such as a printer.

  A CPU (Central Processing Unit) 1701 includes various modules described in the above-described embodiments, that is, a data reading module 110, a line segment detection module 120, a cell configuration detection module 130, a non-ruled line element connection calculation module 140, a non-line It is a control part which performs the process according to the computer program which described the execution sequence of each module, such as the division | segmentation judgment module 150. FIG.

  A ROM (Read Only Memory) 1702 stores programs used by the CPU 1701, calculation parameters, and the like. A RAM (Random Access Memory) 1703 stores programs used in the execution of the CPU 1701, parameters that change as appropriate in the execution, and the like. These are connected to each other by a host bus 1704 including a CPU bus or the like.

  The host bus 1704 is connected to an external bus 1706 such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge 1705.

  A keyboard 1708 and a pointing device 1709 such as a mouse are input devices operated by an operator. The display 1710 includes a liquid crystal display device or a CRT (Cathode Ray Tube), and displays various types of information as text or image information.

  An HDD (Hard Disk Drive) 1711 includes a hard disk (may be a flash memory or the like), drives the hard disk, and records or reproduces a program executed by the CPU 1701 and information. The hard disk stores read image data, data indicating ruled lines, data indicating non-ruled lines, and the like. Further, various other data, various computer programs, and the like are stored.

  The drive 1712 reads out data or a program recorded on a removable recording medium 1713 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and the data or program is read out as an interface 1707 and an external bus 1706. , The bridge 1705, and the RAM 1703 connected via the host bus 1704. The removable recording medium 1713 can also be used as a data recording area similar to the hard disk.

  The connection port 1714 is a port for connecting the external connection device 1715 and has a connection unit such as USB or IEEE1394. The connection port 1714 is connected to the CPU 1701 and the like via an interface 1707, an external bus 1706, a bridge 1705, a host bus 1704, and the like. The communication unit 1716 is connected to a communication line and executes data communication processing with the outside. The data reading unit 1717 is a scanner, for example, and executes document reading processing. The data output unit 1718 is a printer, for example, and executes document data output processing.

  Note that the hardware configuration of the image processing apparatus illustrated in FIG. 17 illustrates one configuration example, and the present embodiment is not limited to the configuration illustrated in FIG. 17, and the modules described in the present embodiment are executed. Any configuration is possible. For example, some modules may be configured with dedicated hardware (for example, Application Specific Integrated Circuit (ASIC), etc.), and some modules are in an external system and connected via a communication line Alternatively, a plurality of the systems shown in FIG. 17 may be connected to each other via communication lines so as to cooperate with each other. In particular, in addition to personal computers, portable information communication devices (including mobile phones, smartphones, mobile devices, wearable computers, etc.), information appliances, robots, copiers, fax machines, scanners, printers, multifunction devices (scanners, printers, An image processing apparatus having two or more functions such as a copying machine and a fax machine) may be incorporated.

  Further, in the description of the above-described embodiment, “more than”, “less than”, “greater than”, and “less than (less than)” in a comparison with a predetermined value contradicts the combination. As long as the above does not occur, “larger”, “smaller (less than)”, “more than”, and “less than” may be used.

The program described above may be provided by being stored in a recording medium, or the program may be provided by communication means. In that case, for example, the above-described program may be regarded as an invention of a “computer-readable recording medium recording the program”.
The “computer-readable recording medium on which a program is recorded” refers to a computer-readable recording medium on which a program is recorded, which is used for program installation, execution, program distribution, and the like.
The recording medium is, for example, a digital versatile disc (DVD), which is a standard established by the DVD Forum, such as “DVD-R, DVD-RW, DVD-RAM,” and DVD + RW. Standard “DVD + R, DVD + RW, etc.”, compact disc (CD), read-only memory (CD-ROM), CD recordable (CD-R), CD rewritable (CD-RW), Blu-ray disc ( Blu-ray (registered trademark) Disc), magneto-optical disk (MO), flexible disk (FD), magnetic tape, hard disk, read-only memory (ROM), electrically erasable and rewritable read-only memory (EEPROM (registered trademark)) )), Flash memory, Random access memory (RAM) SD (Secure Digital) memory card and the like.
The program or a part of the program may be recorded on the recording medium for storage or distribution. Also, by communication, for example, a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a wired network used for the Internet, an intranet, an extranet, or a wireless communication It may be transmitted using a transmission medium such as a network or a combination of these, or may be carried on a carrier wave.
Furthermore, the program may be a part of another program, or may be recorded on a recording medium together with a separate program. Moreover, it may be divided and recorded on a plurality of recording media. Further, it may be recorded in any manner as long as it can be restored, such as compression or encryption.

DESCRIPTION OF SYMBOLS 100 ... Image information processing apparatus 110 ... Data reading module 120 ... Line segment detection module 130 ... Cell structure detection module 140 ... Non-ruled line element connection calculation module 150 ... Non-line segment delimitation judgment module 210 ... Image reading apparatus 230 ... Form processing apparatus 290 ... communication line

Claims (7)

First extraction means for extracting ruled lines and non-ruled lines from the image;
When the positions of the non-ruled lines in the ruled lines are regularly arranged, second extraction means for extracting the non-ruled lines is provided as a partition for dividing the area surrounded by the ruled lines. An image processing apparatus. The second extracting means extracts the non-ruled line as the delimiter when the non-ruled line is present at a position where the ruled line is equally divided with a predetermined width as a minimum width. The image processing apparatus according to claim 1. The image processing apparatus according to claim 1, wherein the second extraction unit targets the non-ruled line whose height and width are in a predetermined range. The second extracting means classifies the non-ruled lines extracted by the second extracting means into a first non-ruled line exceeding the size as a delimiter and other second non-ruled lines, and the second The image processing apparatus according to any one of claims 1 to 3, wherein the first non-ruled line is divided by using the size of the non-ruled line. The second extracting means, when a printed character is already included in the area surrounded by the ruled line, is used as a delimiter for dividing the ruled line part forming the area other than the character. The image processing apparatus according to claim 1, wherein a ruled line is extracted. The second extracting means extracts a non-ruled line as a delimiter in each area using a non-ruled line as a delimiter in another area when a plurality of areas surrounded by the ruled line are arranged. The image processing apparatus according to claim 1, wherein the image processing apparatus is characterized. Computer
First extraction means for extracting ruled lines and non-ruled lines from the image;
When the positions of the non-ruled lines in the ruled lines are regularly arranged, an image for functioning as a second extracting unit for extracting the non-ruled lines as a partition for dividing the area surrounded by the ruled lines Processing program.

Images