JP3215163B2 - Ruled line identification method and area identification method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to extraction of ruled lines, table areas, and the like in a document image.

[0002]

2. Description of the Related Art In a case where data of a general document image in which figures and characters are mixed is filed on a recording medium or transmitted by a G4 facsimile, horizontal ruled lines in the document image,
It is convenient to be able to automatically identify vertical ruled lines, bounded frame regions by ruled lines, table regions by ruled lines, etc., but a technique optimal for such purpose has not yet been established.

[0003] As a related technique known so far, a black-connected rectangle is extracted from a binary image by scanning a document, and the size of the rectangle is compared with a threshold value.
An image extraction method for discriminating between a character rectangle and a line graphic rectangle (Japanese Patent Laid-Open No. 55-162177), integrating black runs into rectangles and integrating runs longer than a certain length inside a rectangle larger than a certain size. Table area identification method for extracting horizontal ruled lines and determining whether or not the rectangle is a table area based on the number of horizontal ruled lines
(JP-A-3-9489) .

[0004]

SUMMARY OF THE INVENTION
It does not identify a horizontal ruled line, a vertical ruled line, a table, an enclosing frame, and the like in detail regarding a line figure. In addition, the table area identification method also extracts only horizontal ruled lines and determines the table area based on the number of horizontal ruled lines. Therefore, the identification accuracy of the table area may be insufficient. Can not.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a technique for identifying and extracting a horizontal ruled line, a vertical ruled line, a table area, a frame area, and other areas with higher accuracy.

[0006]

According to the first or second aspect of the present invention, based on the size in the horizontal direction and the vertical direction from the circumscribed rectangle of the black connected component extracted from the document image or its reduced image. To select a horizontal ruled line rectangle candidate or a vertical ruled line rectangle candidate. By scanning the document image or its reduced image within the range of each rectangular candidate in the horizontal direction in the case of identifying a horizontal ruled line and in the vertical direction in the case of identifying a vertical ruled line, a length equal to or greater than a certain threshold value is obtained. A horizontal ruled line rectangle or a vertical ruled line rectangle consisting of black runs is extracted. Then, based on the relationship between each direction ruled line rectangle candidate and the size of each direction ruled line rectangle extracted in the horizontal direction and the vertical direction (the invention of claim 1), or based on the positional relationship in the horizontal direction and the vertical direction. It is determined whether the ruled line rectangle is a horizontal or vertical ruled line.

According to the third aspect of the present invention , a table area rectangle candidate is selected from the circumscribed rectangles of the black connected components extracted from the document image or its reduced image based on the horizontal and vertical sizes. By scanning the document image or its reduced image in the horizontal direction with respect to the range of each table region rectangle candidate, a horizontal ruled line rectangle consisting of black runs having a length equal to or more than a certain threshold value is extracted, and the horizontal ruled line rectangle is horizontally scanned. A horizontal rule is extracted by performing rule condition determination, and a document image or a reduced image thereof is vertically scanned to extract a vertical rule consisting of a black run having a length equal to or greater than a certain threshold value. On the other hand, a vertical ruled line is extracted by determining the condition of the vertical ruled line. And the table area rectangle candidate
It is determined whether or not the table area rectangle candidate is a table area, based on a determination condition including at least the number of horizontal ruled lines and the number of vertical ruled lines extracted from . Further, the determination conditions also include the presence or absence of horizontal ruled lines or vertical ruled lines in a specific range in the table region rectangular candidate, said surface region rectangular candidate enclose by the determination condition
It is determined whether the area is a frame area or another area.

[0008]

According to the first or second aspect of the present invention, accurate identification of a horizontal or vertical ruled line is performed by going through the steps of selecting a ruled line rectangle candidate, extracting a ruled line rectangle from the ruled line rectangle candidates, and determining a ruled line rectangle condition. , Extraction is possible.

According to a third aspect of the present invention , there is provided an area identifying method for selecting a table area rectangular candidate, extracting a horizontal ruled line and a vertical ruled line in the table area rectangular candidate, and extracting a horizontal ruled line and a vertical ruled line in the table area rectangular candidate.
In addition to the number, the determination of conditions including the presence of horizontal or vertical ruled line at a particular range of the table region rectangular candidates, by going through the steps of a possible accurate identification of the tablespace, be
La is also possible identification tablespace other surrounding frame region and other regions.

[0010]

Embodiment 1 In this embodiment, as shown in FIG. 1, a document image memory 10 for storing document image data input by a scanner or the like.
0, a rectangle extracting unit 102 for extracting a circumscribed rectangle of a black connected component of the document image data in the memory 100, and a rectangular memory 10 for recording information of the rectangle extracted by the rectangle extracting unit 102
4. Based on the rectangle information in the rectangle memory, the rectangle classification unit 106 classifies the rectangle by comparing the width and height of the extracted rectangle with a predetermined threshold value. The rectangular image (stored in the rectangular memory 104) determined as a horizontal separator (horizontal ruled line) is referred to, and the rectangular image (the document image memory 100) is referred to.
, A horizontal ruled line extracting unit 10 that performs a horizontal scan on the horizontal rule to extract a black run in the horizontal direction and generates a rectangle from a black run having a length equal to or greater than a predetermined threshold.
8. A horizontal ruled line rectangle memory 110 for storing the information of the rectangles extracted by this, based on the information of the rectangles stored in the memory 110, a horizontal line for making a final determination as to whether or not the rectangle is a horizontal separator (horizontal ruled line) Ruled line verification unit 11
2. It has a functional configuration including a control unit 114 that controls each of the processing units 102, 106, 108, and 112.

Note that an image reduction unit may be provided in a stage preceding the rectangle extraction unit 102, and the following processes may be performed on a reduced image of the input document image. Each processing unit including the image reduction unit is realized as hardware or software on a computer system.

FIG. 2 shows the flow of the horizontal parameter (horizontal ruled line) determination processing in this embodiment. Less than,
The details of this process will be described. FIG. 3 is an explanatory diagram of the process which is appropriately referred to in the description.

First, the rectangle extracting unit 102 extracts a circumscribed rectangle of the black connected component from the input document image (step 200). This processing can be performed by, for example, repeating the operation of extracting the circumscribed rectangle of the connected black run while scanning the image, and integrating the extracted rectangles within a certain distance.

For the extracted rectangle, a rectangle classification unit 1
At 06, a comparison is made between the horizontal size W and the vertical size H of the rectangle and the respective thresholds RLHTH and RLVTH, and a rectangle satisfying the conditions of W> RLHTH and H <RLVTH is determined as a candidate for a horizontal ruled line rectangle. Barrel rectangle (1)
Is determined (steps 205 and 210 ).

In FIG . 3A, reference numeral 300 denotes a rectangle (1).
Is shown, and the hatched portion inside is a black connected component such as a character or a ruled line. Note that the rectangle extracting unit 102 extracts a rectangle by integrating such black connected components such as characters and ruled lines, but such processing is known and will not be described in detail.

In this embodiment, the threshold values RLHTH, RL
Although VTH is fixed, it may be adaptively and automatically set. As an example, a histogram of the vertical size (in the case of a horizontally written document) of a black connected component that can be regarded as a character in the rectangle extraction process 200 is created, and a standard character size is determined based on the histogram. By multiplying this standard character size by an appropriate coefficient, the threshold R
LHTH and RLVTH are calculated. However, this is only an example.

Next, the horizontal ruled line extracting unit 108 scans the image in the range of the rectangle (1) in the horizontal direction to extract a black run having a length equal to or longer than the threshold value RUNHTH. A rectangle (rectangle (2)) circumscribing to is extracted as a ruled line rectangle (step 21).
5). In FIG. 3B, reference numeral 302 denotes a rectangle (1) 300
Is a rectangle (2) generated only from the long black run extracted from, and H1 and W1 are the sizes in the vertical and horizontal directions. Although the threshold value RUNHTH is given in advance as a fixed value, it may be adaptively and automatically set in the same manner as RLHTH and RLVTH.

For such a rectangle (2), W1 / W> threshold (for example, 0.8) in the horizontal ruled line verification section 112.
In addition, a condition determination of H1 / H < threshold (for example, 0.8) is performed (steps 220 and 225), and when this condition is satisfied, the rectangle (2) is finally determined to be a horizontal ruled line (step 230). The rectangle (2) 3 shown in FIG.
02 is finally determined to be a horizontal ruled line.

The control unit 114 determines the end of the processing on the rectangle extracted by the rectangle extraction unit 200 (step 23).
5) is performed, and if an unprocessed rectangle remains, the processing from step 205 onward is restarted.

Embodiment 2 This embodiment has the functional configuration shown in FIG. 1 as in Embodiment 1. The processing content is the same as that of the first embodiment except for the processing of the horizontal ruled line verification unit 112.

That is, in the first embodiment, the horizontal ruled line test unit 112 tests the horizontal ruled line based on the ratio between the horizontal and vertical sizes of the rectangle (1) and the rectangle (2) (see FIG. 2). Steps 220 to 230). On the other hand, in the present embodiment, as shown in FIG. 3C, the difference ΔW1, ΔW2 in the horizontal position between the start and end points of the rectangle (1) 300 and the rectangle (2) 302, and the difference in the vertical position ΔH1,
When ΔH2 is ΔW1 <threshold and ΔW2 <threshold and ΔH1 <threshold and ΔH2 <threshold, the rectangle (2) 302 is finally determined to be a horizontal ruled line.

Embodiment 3 This embodiment has a functional configuration as shown in FIG. This embodiment is different from the first embodiment in the following point from the relation of extracting a vertical ruled line.

In the rectangle classifying unit 106, the horizontal size of the rectangle is W and the vertical size is H, and W <RL
When the condition of HTH and H> RLVTH is satisfied,
The rectangle is classified as rectangle (1) which is a candidate for a vertical ruled line rectangle. Since the vertical ruled line is targeted, the threshold values RLHTH and RLVT are different from those of the first and second embodiments targeted at the horizontal ruled line.
The magnitude relationship of H is reversed.

The vertical ruled line extraction unit 408 vertically scans the document image in the range of the rectangle (1) to extract black runs, and extracts from the connected components of only black runs having a length equal to or longer than a predetermined threshold value RUNHTH. The rectangle (2) is extracted and the information is stored in the vertical ruled line rectangle memory 410.

The vertical ruled line verification unit 412 calculates the horizontal size W1 of the rectangle (2), the vertical size H1 and the horizontal size W and the vertical size H of the rectangle (1). Where W1 / W <threshold (for example, 0.8) and H1 /
If the condition of H> threshold (for example, 0.8) is satisfied, the rectangle (2) is finally determined to be a vertical ruled line.

In this embodiment, vertical ruled line extraction may be performed on an image obtained by reducing the input document image.

Embodiment 4 This embodiment has a functional configuration shown in FIG. The processing content is the same as that of the third embodiment except for the processing of the vertical ruled line verification unit 412.

That is, in the present embodiment, the vertical ruled line verification unit 412 determines the difference ΔW1, ΔW2 in the horizontal position between the start and end points of the rectangle (1) and the rectangle (2), and the difference ΔH1, ΔH2 in the vertical direction. When ΔW1 <threshold and ΔW2 <threshold and ΔH1 <threshold and ΔH2 <threshold, the rectangle (2) is finally determined to be a vertical ruled line.

In this embodiment, the processing can be performed on a reduced image of the input document image.

Embodiment 5 FIG. 5 is a functional block diagram of this embodiment. However, FIG.
Alternatively, the equivalent parts in FIG. 4 are indicated by the same reference numerals. 5
Reference numeral 06 denotes a rectangle classification unit, which is different from the rectangle classification unit 106 of FIG. 1 or 4 in classification conditions. Reference numeral 514 denotes an added area determination unit. FIG. 6 is a process flowchart of the present embodiment, and FIG. 7 is a diagram for explaining the process. Hereinafter, the processing content will be described.

The rectangle extracting unit 102 extracts a circumscribed rectangle of the black connected component from the document image, and stores information on the rectangle in the rectangle memory 104 (step 600). Note that it is also possible to perform a reduction process on the input document image, and execute the processes following the rectangle extraction for the reduction process.

In the rectangle classifying unit 506, a comparison is made between the horizontal size W and the vertical size H of the rectangle and the horizontal threshold value LARGEHTH and the vertical threshold value LARGEVTH, and W> LARGEHTH and H> A rectangle that satisfies the condition of LARGEVTH is determined to be a rectangle (1) that is a table region rectangle candidate (steps 605 and 610).

Note that, since the purpose is to identify a table area having a certain size or more, which is composed of horizontal ruled lines and vertical ruled lines, the threshold values LARGEHTH, LA
RGEVTH is predetermined in consideration of the minimum size of such a table area to be identified. However, these thresholds may be adaptively automatically determined based on, for example, a histogram of the height of the rectangle.

Next, horizontal ruled lines are extracted from the rectangle (1) by the horizontal ruled line extraction unit 108 and the horizontal ruled line verification unit 112 (step 615). As in the first embodiment, the horizontal ruled line extraction unit 108 horizontally scans an image within the range of the rectangle (1), which is a table region rectangle candidate, and forms a horizontal ruled line rectangle generated only from black runs having a length equal to or longer than a predetermined value. Extract rectangle (2). In the horizontal ruled line verification unit 112, regarding the horizontal size W1, the vertical size H1, and the horizontal size W of the rectangle (1), W1 / W> threshold (for example, 0.8) When the condition of H1 <RLHeightTH is satisfied, the rectangle (2) is extracted as a horizontal ruled line. Note that the threshold value RLHeight may be a fixed value or a variable value determined adaptively.

The vertical ruled line extraction unit 408 and the vertical ruled line verification unit 412 extract a vertical ruled line from the rectangle (1) which is a table region rectangle candidate (step 620). As in the third embodiment, the vertical ruled line extraction unit 408 vertically scans the image in the range of the rectangle (1) and determines the rectangle (2) as a vertical ruled line rectangle generated only from black runs having a length equal to or more than a predetermined value. Extract,
By the vertical ruled line verification unit 412, regarding the horizontal size W1, the vertical size H1, and the vertical size H of the rectangle (1), H1 / H> threshold (for example, 0.8) When the condition of W1 <RLWidthTH is satisfied, the rectangle (2) is extracted as a vertical ruled line. Note that the threshold value RLwidthTH may be a fixed value or a variable value determined adaptively.

Next, in the area determination section 517, the rectangle (1) which is a table area rectangle candidate classified by the rectangle classification section 506 is used.
Each time, the number of horizontal ruled lines and vertical ruled lines extracted therefrom is compared with a threshold value (steps 625 and 630). Here, assuming that 3 is used as the threshold value, the rectangle (1) from which both three or more horizontal ruled lines and three or more vertical ruled lines are extracted is determined as a table area (step 635). In addition,
The position coordinates of the ruled lines extracted from the direction ruled line verification units 112 and 412 are output, and the area determination unit 517 compares the position coordinates with the information of the rectangle (1) stored in the rectangle memory 104 to determine each ruled line. It recognizes which rectangle (1) it belongs to.

In FIG. 7, reference numeral 700 denotes an example of a rectangle (1) which is a table area rectangle candidate. In this example, the rectangle (1) 70
0 is a table area including three horizontal ruled lines and three vertical ruled lines. The horizontal ruled lines are extracted as rectangles (2) 702a to 702c, and the vertical ruled lines are extracted as rectangles (2) 704a to 704c. Each of them is finally determined as a horizontal ruled line or a vertical ruled line. The condition of the number of each ruled line is satisfied, and the area is determined as a table area.

Embodiment 6 This embodiment has the same functional configuration as Embodiment 5. The processing content of the present embodiment is the same as that of the fifth embodiment except that the determination conditions (corresponding to steps 625 and 630 in FIG. 6) of the area determination unit 517 are different.

FIG. 7 shows the processing of the area determination unit 517.
This will be described with reference to FIG. The uppermost horizontal ruled line (70) in a range RangeUTH from the upper side of the rectangle (1) 700
2a) is an upper, lowermost horizontal ruled line (702) in a range RangeLTH from the lower side of the rectangle (1) 700.
Let c) be lower. Such upper and lo
If there is a wer and there are three or more horizontal ruled lines and vertical ruled lines as in the fifth embodiment, the rectangle (1) is determined as a table area.

Embodiment 7 This embodiment has the same functional configuration as Embodiment 5. The processing content of this embodiment is the same as that of the fifth embodiment except that the determination condition of the area determination unit 517 is different from that of the fifth embodiment.

The processing of the area determination unit 517 will be described with reference to FIG. As in the case of the sixth embodiment, upper and lower are obtained. Further, the vertical ruled line (704b) in the middle range excluding the range of RangeLFTHk from the left side of the rectangle (1) 700 and the range of RangeRTH from the right side of the rectangle (1) is set as middle.

Then, there are three or more horizontal ruled lines, and
If there is an upper and a lower (see Embodiment 6) and there is at least one middle, the rectangle (1) is determined as a table area.

Embodiment 8 This embodiment has the same functional configuration as Embodiment 5. The processing content of this embodiment is the same as that of the fifth embodiment except that the determination condition of the area determination unit 517 is different from that of the fifth embodiment.

The processing of the table area determination unit 517 will be described with reference to FIG. As in the case of the sixth embodiment, upper and lo
Find the wer. Furthermore, the leftmost vertical ruled line (7) in the range of RangeLFTH from the left side of the rectangle (1) 700
04a) is left and Range is from the right side of rectangle (1).
The rightmost vertical ruled line (704c) in the range of RTH is r
and the vertical ruled line (704b) in the middle range other than those ranges is determined as middle.

Then, in region (1), in Example 5,
If any of the table area determination conditions 6 and 7 is satisfied, the area (1) is determined as a table area. If that is not the case, upper, lower, left, ri
When one or more of the ghts exist, the area (1) is determined to be an enclosing area, but upper, lower, left,
When neither of the right does not exist, the area (1) is determined to be neither the table area nor the surrounding frame area, but another area (figure, photograph, etc.).

[0046]

As is apparent from the above detailed description , according to the first or second aspect of the present invention , the effect that the horizontal or vertical ruled line can be accurately identified is obtained, and the third aspect of the present invention is achieved.
According to the above, there is an effect that accurate identification of the table region and identification of the surrounding frame region and other regions are enabled.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a first embodiment and a second embodiment.

FIG. 2 is a processing flowchart of the first embodiment.

FIG. 3A is a diagram showing an example of extracting horizontal ruled line rectangle candidates; (B) It is explanatory drawing of a horizontal ruled line determination in Example 1. FIG. (C) It is explanatory drawing of a horizontal ruled line determination in Example 2.

FIG. 4 is a functional block diagram of a third embodiment and a fourth embodiment.

FIG. 5 is a functional block diagram of embodiments 5, 6, 7 and 8.

FIG. 6 is a processing flowchart of a fifth embodiment.

FIG. 7A illustrates an example of a table area rectangle candidate. (B) It is explanatory drawing regarding a horizontal ruled line. (C) It is explanatory drawing regarding a vertical ruled line.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 document image memory 102 rectangle extraction unit 104 rectangle memory 106 rectangle classification unit 108 horizontal ruled line extraction unit 110 horizontal ruled line rectangle memory 112 horizontal ruled line verification unit 114 control unit 300 horizontal ruled line rectangle candidate (rectangle (1)) 302 horizontal ruled line rectangle (rectangle) (2)) 408 vertical ruled line extraction unit 410 vertical ruled line rectangle memory 412 vertical ruled line verification unit 506 rectangle classification unit 517 area determination unit 700 table area rectangle candidate (rectangle (1)) 702a to 702c horizontal ruled line rectangle (rectangle (2)) 704a to 704c Vertical ruled line rectangle (rectangle (2))

Claims (3)

(57) [Claims] 1. A circumscribed rectangle of a black connected component is extracted from a document image or a reduced image thereof. (A) The rectangle extracted in the above (a) is extracted based on the horizontal and vertical sizes thereof. Selecting a ruled line rectangle candidate in a specific direction, either the horizontal direction or the vertical direction, and (c) scanning the document image or its reduced image in the specified direction for the range of the ruled line rectangle candidate selected in (b) above, generates a ruled line rectangle from black runs of longer than, (d) and ruled line rectangle candidates selected in (b), the
A ruled line identification method characterized by determining whether or not the ruled line rectangle is a ruled line in a specific direction based on the relationship between the horizontal and vertical sizes of the ruled line rectangle generated in (c) . (A) extracting a circumscribed rectangle of a black connected component from a document image or a reduced image thereof; and (a) extracting a rectangle extracted in (a) based on the horizontal and vertical sizes thereof. Selecting a ruled line rectangle candidate in a specific direction, either the horizontal direction or the vertical direction, and (c) scanning the document image or its reduced image in the specified direction for the range of the ruled line rectangle candidate selected in (b) above, generates a ruled line rectangle from black runs of longer than, (d) and ruled line rectangle candidates selected in (b), the
A ruled line identification method characterized by determining whether or not the ruled line rectangle is a ruled line in a specific direction based on the horizontal and vertical positional relationships of the ruled line rectangle generated in (c) . 3. A circumscribing rectangle of a black connected component is extracted from a document image or a reduced image thereof. A. Based on the horizontal and vertical sizes of the rectangle extracted in the above step A. (C) by scanning a document image or its reduced image in the horizontal direction over the range of the table region rectangle candidate selected in (b) above, thereby obtaining a black run having a length equal to or longer than a certain threshold value. Is extracted by performing a horizontal ruled condition determination for the horizontal ruled line rectangle, and a horizontal ruled line is extracted. (D) The document for the range of the table area rectangle candidate selected in (b) above The image or its reduced image is scanned in the vertical direction, a vertical ruled line consisting of a black run having a length equal to or more than a certain threshold is extracted, and the vertical ruled line is extracted by performing a vertical ruled condition determination on the vertical ruled line rectangle. (E) From the table area rectangle candidate selected in (a) above,
With horizontal and vertical ruled lines extracted in (c) and (d)
None, depending on the number, the table area rectangle candidate is
Area or other area
An area identification method characterized by the following .