JPH0334081A - Drawing reader - Google Patents

Abstract

PURPOSE:To easily and accurately read the characters regardless of the sizes and formats of these characters when the characters are read out of a document or a drawing including the characters, the segments and symbols and recognized by preparing a contour vectorization processing part, a histogram preparing part, a character string extracting part, and a character segmenting part. CONSTITUTION:A contour vectorization processing part 1 which applies a contour vector process to the binarized picture element data and obtains the square data circumscribed to the contour vector as the character candidates, a histogram preparing part 2 which obtains a one-dimensional histogram for the circumscribed square serving as the character candidates, a character string extracting part 3 which extracts a character string based on the one-dimensional histogram data, a character segmenting part 4 having a function to integrate plural character candidates included in the extracted character string are provided. In such a constitution, the characters can be easily and accurately read regardless of the sizes and formats of these characters when the characters are read out of a document or a drawing including the characters, the segments and the symbols.

Description

DETAILED DESCRIPTION OF THE INVENTION A. Industrial Applications The present invention relates to a drawing reading device that detects character data from documents and drawings containing a mixture of characters, line segments, symbols, etc. and recognizes the characters.

B1 Summary of the Invention The present invention extracts character data from binary pixel data corresponding to black and white obtained by raster scanning an object such as a document or drawing containing a mixture of characters, line segments, and symbols, and recognizes the character. A drawing reading device that performs contour vectorization processing for the binarized pixel data, and a contour vectorization processing unit that calculates circumscribed rectangle data circumscribing the contour vector as a character candidate; A histogram creation unit that creates a one-dimensional histogram, a character string extraction unit that extracts a character string based on the one-dimensional histogram data, and a character extraction unit that has a function of integrating multiple character candidates within the extracted character string. By establishing this section, when reading and recognizing characters from documents or drawings that contain a mixture of characters, line segments, and symbols, it is possible to read and recognize characters of any size or format without imposing any restrictions on size or format. This allows for easy and accurate reading.

C8 Conventional technology When reading characters, figures, etc., for example, an object such as a document or drawing is raster-scanned using an input device such as an image scanner, and converted into black and white binary image data (human cover pattern). The character data of the object is extracted from the value image data, and object recognition processing is performed. In order to recognize characters included in a drawing or the like in which characters, figures, etc. are mixed, it is necessary to first separate the text part and the figure part, and then cut out each character's worth of information for recognition.

D1 Problems to be Solved by the Invention Unlike those in documents, the characters included in the drawings do not have a constant size, and their format is also not fixed. For this reason, it has been extremely difficult to extract and recognize characters contained in documents or drawings that contain a mixture of characters, line segments, symbols, etc., regardless of their format or size.

The present invention has been made in view of the above points, and its purpose is to:
To provide a drawing reading device that can easily and accurately read characters included in documents and drawings, regardless of their size and format.

E9 Means for Solving Problems The present invention extracts character data from binarized pixel data corresponding to black and white obtained by raster scanning an object in which characters, line segments, and symbols are mixed, and extracts character data from the extracted pixel data. In a drawing reading device that recognizes characters by referring to and comparing character data and a character dictionary, two adjacent black pixels are detected based on binarized pixel data corresponding to black and white obtained by raster scanning the subject. a contour vectorization processing section that obtains contour vector data formed by combining vectors connecting pixels, and obtains circumscribed rectangular data circumscribed to the contour vector as a character candidate; and a character candidate obtained by the contour vectorization processing section. a histogram creation unit that calculates a one-dimensional histogram in each of the horizontal, vertical, and diagonal directions for a circumscribed rectangle; and a histogram creation unit that uses data that is equal to or higher than a predetermined threshold value as character string candidates among the data obtained by the histogram creation unit. At the same time, from among the extracted data, data in which the height h of a character string candidate and the distance fsd between mutually adjacent character candidates in the character string candidate are in a relationship such that d<h×α (α is a constant) are extracted. A character string extraction unit that extracts characters as a character string; and a character extraction unit that has a function of integrating a plurality of adjacent character candidates in the character string extracted by the character string extraction unit and that cuts out characters from the character string. , character recognition is performed based on the character data extracted by the character extraction section.

171 Effect When a circumscribed rectangle is determined by the vectorization processing unit, character candidate data is obtained from an object (document or drawing) in which characters, line segments, and symbols are mixed. Among the histograms obtained by the histogram creation unit for the circumscribed rectangle of this character candidate, data having a predetermined value or more becomes a character string candidate.

Among these character string candidates, data for which the relationship d<h×α (d is the distance between adjacent character candidates, h is the height of the character string candidate, and α is a constant) is extracted into a character string by the character string extraction unit. is extracted as That is, line segments and symbols mixed in the subject are excluded, and only character string data is extracted. Next, the character cutting section cuts out characters from the character string. At this time, if the size of a plurality of adjacent character candidates in the character string or the distance between character candidates is a predetermined value, the plurality of character candidates are integrated, and the integrated character candidate is cut out as one character. This allows characters to be read even if the size and format of the characters are questionable.

G. Embodiment Hereinafter, an embodiment of the present invention will be described with reference to the drawings. First, as shown in FIG. 1, the drawing reading device according to the present invention is equipped with a contour vectorization processing section 11, a histogram creation section 2, a character string extraction section 3, a character cutting section 4, and a character recognition section 5, and the overall processing flow. is shown as in FIG. That is, first, in step S1, the contour vectorization processing unit 1 performs preprocessing, in step S1, the histogram creation unit 23, character string extraction unit 3, and character extraction unit 4 perform character extraction processing, and in step S, the character recognition unit 5 performs preprocessing. Performs character recognition processing. The preprocessing performed by the contour vector ratio processing section 1 is represented by a flowchart as shown in FIG. That is, in step S, image input processing is performed;

Contour vectorization processing is performed in step S.

Element separation processing is performed at . Among these, the image input processing and contour vectorization processing are performed, for example, in
The contour detection device described in the '684 patent specification is used as follows.

FIG. 4 is a diagram showing the configuration of the contour detection device, and 11 is a contour extraction section. As shown in FIG. 5, this contour extraction unit 11 imports binary pixel data corresponding to black and white obtained by raster scanning a subject such as a document or drawing, and extracts 2i horizontally and vertically! Pixel data of ii pixels x 2 pixels is sequentially extracted along the scan line, a command string for contour detection is created based on this pixel data, and the subsequent contour analysis section I2
Output to. The contour analysis section +2i updates the descriptions of the contour element table, contour management table, and contour connection table in the table storage section 13 based on the above command sequence. Additionally, the contour analysis unit 12 (macroscopic processing unit 3
In addition to 1), the microscopic processing unit 32) extracts several points based on the relationship between the maximum distance between two consecutive points extracted in this process and the threshold value that sets the points that fit together. The contour vector of the point thus determined is written in the contour vector series table 33, and the short vector among these contour vectors is removed by the short vector decentering section 34, and the description thereof is updated.

Here, the contour element corresponds to a vector connecting two adjacent black pixels, and the contour is constructed by connecting these. As shown in FIG. 6, the contour element table is a table in which a unique code is attached to each contour element and a contour made up of a group of these contour elements, and for each contour element, its coordinates and direction, and the contour element to which it belongs are assigned. This is to describe the code of the contour element and the code of other contour elements connected before and after the contour element, respectively. The contour management table is for recording the codes of the contour elements located at the tip and end of each contour as shown in FIG. This is to describe the coordinates arranged in the direction, the code of an unconnected contour element whose front end or rear end exists at the coordinate, and the distinction between the front and rear ends of the unconnected end of the contour element in correspondence with each other. The contour vector series table is used to record the code of the starting point coordinates for each contour vector series, as shown in FIG.

The command string output from the contour extraction unit 11 includes information on the connection relationship between pixels and contour pixels related to the 2 pixel x 2 pixel pixel data extracted at that time, and information on the black and white arrangement pattern of the pixel data. This is a combination of a connection/pattern code including the above pixel data and a coordinate code indicating the X coordinate of the pixel data.

One row of the above connection relationships is shown in Figure 1O. When the pixel data of 2 pixels x 2 pixels surrounded by the large frame in Figure 1(A) is imported, the outline element located at the X coordinate Xn of this pixel data is This information indicates that the front end is connected. In this example, the coordinates of the pixel data are the coordinates of the pixel P located at the lower right when facing the page. Also, when you import the pixel data surrounded by the large frame in the same figure (b),
This is information that the rear end of the contour element located at the coordinate Xn-+ one position before the X coordinate of this pixel data is connected. In addition, the frame with a circle mark in Figure 1O indicates a black pixel,
A frame without a circle mark indicates a white pixel.

For example, if we focus on the broad frame in Figure 10 (a), the connection/pattern code in this case includes the above-mentioned connection information and the information on the black and white array pattern inside the thick frame, and the command string is based on this connection/pattern code. It is a combination of the code and the coordinate code indicating the X coordinate of the pixel P. The coordinate code of the mouse can also be generated on the contour analysis section 12 side in synchronization with the timing of fetching the command sequence.

Is the command sequence obtained in this way contour analysis? When imported into B12, the following processing is performed.

Assuming that a command string related to the pixel data of the large frame in Figure 11 has been imported, the contour element Cj shown by the dotted line is registered in the contour element table, its direction and coordinates are entered, and the lines connected before and after the contour element Cj are respectively entered. Enter the number of the other contour element in the connected element number column. In this case, the contour element Cj is located in front of the contour element C1, so Ci is written in the rear connection column related to the column of the contour element Cj, and Ci is written in the front connection shelf related to the column of the contour element Cj. Furthermore, in this example, the number of the contour to which the contour element Cj belongs is
Enter in the contour number field. For example, in the case of 8-connected contour elements, the directions of the contour elements are up and down, left and right, as shown in Figure 12.
a1 to a are defined corresponding to the diagonal directions, and in the case of four connections, four directions, up, down, left and right, are defined. Then, regarding the contour number Si in the contour management table, the tip contour element number column is updated from Ci to Cj, and regarding the X coordinate of the pixel data in the contour connection table, the forward connection column of the unconnected contour element number is updated to Ci. Update from Cj to Cj. By the way, in actual processing, the contour element to be connected to the contour element Ci is clarified by the contour connection table, so the description of Cj of the front connection shelf related to contour element number c3- in the contour element table is This is done by referring to the connection table. In the above, when two contours generated separately as the raster scan progresses are connected to form one contour, one contour is integrated with the other and the associated contour numbers become the same.

In the above process, the point that is the maximum from the starting point is extracted from each point that makes up one contour pixel series, and then the point that is the maximum distance between the two consecutive extracted points is extracted. . The contour vectors extracted in this way are stored in a table.

Through the above-described manual image processing and contour vectorization processing, an input image of the character rAJ of the subject, for example, is represented by a contour vector as shown in FIG.

Next, the element separation process (step S4 in FIG. 3) for extracting character candidates will be explained. First, short vectors are removed using the contour vector series table shown in FIG. 9 mentioned above. FIG. 14 is a flowchart illustrating the short vector removal means. In step S4, the circumscribed rectangle of the contour vector (shown in FIG. 15) (X + mln, y + mt) is extracted from the contour vector series table shown in FIG. 9.
n), (X va a x +y--x) is calculated. Next, in step S, dX=X+++an X+sl
n+ dV=Vu+ax 5'sin and threshold D
Compare X th and D Y th. Step S3
is the comparison judgment part of step S, and in this step S, (
dx>DXth)U (dy>DYth) and Y
If it is ES, information on a circumscribed rectangle is created as a contour vector in step S4. An example of this information is shown in FIG. 16 as a circumscribed rectangle information table.

If No in step S, then in step S5 vectors of the same series are deleted from the contour vector series table shown in FIG. 9 to remove noise.

FIG. 15 is an explanatory diagram showing a circumscribed rectangle of a contour vector, and in this figure, XmallXIllm0y□x,
ymln looks like this:

Xo+ax=Tn a X (”'+ X ++ ”
')X n1n= rn l n (''' + X 1
+...)3/ max-rn a
...) d X −X, wax X m
s.

dy=y+++□-5'aln (dx<DXth)n (dy<+)yth) However, D
X th, D Y Ih: L threshold.

Based on the circumscribed rectangle information (FIG. 13) obtained as described above, character candidates are extracted under the following conditions.

Condition ■ Part 3 < threshold and wy threshold (wx is the length of the circumscribed rectangle in the X direction, Wy is the length of the circumscribed rectangle in the X direction) Condition ■ Completely included in the outer contour vector and the circumscribed rectangle of the character candidate The outer and inner contour vectors are linked to contour vectors that are character candidates.

As mentioned above, preprocessing is performed as shown in FIGS. 3 to 16, but
Next, the character extraction process (step S in FIG. 2) will be explained.

First, this character cutting process is performed on the character candidates obtained by the contour vector processing section 1 in the horizontal direction (step S, ), vertical direction (step S, ), and diagonal direction (step S3) as shown in FIG. It will be held in The processing in each direction is the same, and as shown in Fig. 18, after performing character string candidate extraction processing (step S,), character string extraction processing (step S,), character extraction process(
Step S3) is performed. The processing of each step is performed as follows. That is, the histogram creation unit 2 applies Hough transformation to create a circumscribed rectangle (
A one-dimensional histogram regarding R (in each direction) is taken for (Fig. 13). At this time, the state of the histogram for all the stand-up characters of the character candidates is shown as shown in FIG. 1st
Figure 9 is a (R=Y) histogram in the Y direction,
He in the figure indicates the height of the character string candidate. Furthermore, when a histogram in the X direction is taken, he indicates the width of the character string candidate. Then, the character string extracting unit 3 extracts characters whose peaks in the histogram are equal to or higher than a predetermined threshold value as character string candidates. Incidentally, a flowchart of these processes is shown in FIG.

Furthermore, the character string extraction unit 3 extracts a character string that meets the following conditions from among the character string candidates. In other words, as shown in FIG. 21, when the height he of the character string candidate and the character candidate unit Md (i, j) < hc i and j are character strings. Next, the character cutting section 4 cuts out characters from the extracted character string in units of character candidates. At this time, if the conditions shown in the following equation (2) are met, the separated characters in the height direction are integrated as in step S3 in FIG. 22, and the conditions shown in equations (3) and (4) are If it matches, width direction separated character integration is performed as in step S of FIG. For example, as shown in Figure 23, the characters Matsuura i, j are printed in the height direction.
If the following condition (2) is satisfied when the character candidates i and j are separated, the character candidates i and j are integrated as shown.

l1ax (x + wtn(i), x −+n(j)
≦(X c, n(+) Or X cen(j)≦ma
n (x + max (1), x 5ax (j)
+++ +++ (2) (Xm1n: Minimum coordinate in the X direction of the circumscribed rectangle Xwma*: Maximum coordinate in the X direction of the circumscribed rectangle
c*n: X-direction center coordinate of circumscribed rectangle) Also, for example, the second
4 Character candidates j in the width direction as shown in Figure 4.

If the conditions of equations (3) and m(4) below are satisfied when k is separated, the character candidates j are integrated as shown in the figure.

d(j,k)<(average within string)×constant・・・(
3) And w(j)4w(k)+d(j,k)≦h×constant...(4) (d: Distance between character candidates W: Character candidate width h: Character string height) As mentioned above In the process of character extraction processing (FIGS. 17 and 18), each time a character candidate is determined as a character string, a processed mark is attached and the number of objects to be processed decreases. Further, each process in FIG. 18 has been described in the horizontal direction, but in the vertical direction, the x and y directions are reversed. For the diagonal direction, a predetermined inclination angle is assumed, and the coordinate conversion processing at that angle is the character string extraction processing and character cutting processing (step S in Fig. 18).
.

Included in Part 3).

The character recognition process (step S in FIG. 2) is performed as follows. The character recognition unit 5 references the characters extracted by the character extraction unit 4 and a dictionary in which characters are registered in advance,
Compare and recognize -characters one by one.

H1 Effects of the Invention As described above, according to the present invention, the contour vectorization processing section,
Since we have provided a histogram creation section, a character string extraction section, and a character extraction section, when reading and recognizing characters from documents or drawings that contain a mixture of characters, line segments, and symbols, it is possible to set limitations on size and format. Characters of any size or format can be read easily and accurately.

[Brief explanation of drawings]

1 to 24 show an embodiment of the present invention, in which FIG. 1 is a block diagram of the overall configuration, FIG. 2 is an overall flowchart, FIG. 3 is a flowchart of the contour vectorization processing section, and FIG. 4 is a block diagram of a contour detection device, FIG. 5 is an explanatory diagram showing the state of raster scanning, FIG. 6 is an explanatory diagram showing a contour element table, FIG. 7 is an explanatory diagram showing a contour management table,
Fig. 8 is an explanatory diagram showing the contour connection table, Fig. 9 is an explanatory diagram showing the contour vector series table, Fig. 1O (a),
(B) is an explanatory diagram showing the relationship between the square element data and the contour elements, and FIG. 11 is an explanatory diagram showing the connection state between the contour elements.
FIG. 12 is an explanatory diagram showing the direction of contour elements, FIG. 13 is an explanatory diagram of the contour vectorization process, FIG. 14 is a flowchart of the contour vector removal means, FIG. 15 is an explanatory diagram of the circumscribed rectangle, and FIG. 16 is an explanatory diagram of the contour vectorization process. is an explanatory diagram showing a circumscribed rectangle information table, FIGS. 17 and 18 are both flowcharts of character extraction processing, FIG. 19 is an explanatory diagram of the histogram creation section, FIG. 20 is a flowchart of character string candidate extraction processing, and FIG. Figure 22 is an explanatory diagram of character string extraction processing, Figure 22 is a flowchart of character extraction processing, Figure 23 is an explanatory diagram of height direction separated character integration processing, and Figure 24 is an explanatory diagram of width direction separated character integration processing. . l...Contour vectorization processing unit, 2...Histogram creation unit, 3...Character single string extraction unit, 4...Character cutting unit, 5...Character recognition unit, 11...Contour extraction Part 1
2... Contour analysis section, 13... Table storage section, 31
... Macroscopic processing section, 32... Microscopic processing section, 33.
. . . Contour vector series table, 34 . . . Short vector removal unit. Figure 1 Block diagram of the embodiment Figure 2 Flowchart of the entire system Figure 3 Flowchart of the contour vector turtle processing section (pre-processing) Figure 5
Figure 7: Illustration of raster scan Figure 7: Illustration of contour management table Figure 8: Contour connection table Figure 9: Contour vector series table Figure 10: Pixel data and contour I! ! Figure 14 Flowchart of sit vector removal means Figure 15 Figure 17 Flowchart of character extraction processing Character Figure 22: Flowchart of character extraction processing Figure 23: Explanation of height separation character integration processing Figure 24: Explanation of widthwise fractional numeric integration processing Figure integration process

Claims (1)

[Claims] (1) Extract character data from binary pixel data corresponding to black and white obtained by raster scanning an object containing a mixture of characters, line segments, and symbols, and refer to and compare the extracted character data with a character dictionary. In a drawing reading device that recognizes characters by scanning, a vector connecting two adjacent black pixels is combined based on binary pixel data corresponding to black and white obtained by raster scanning the subject. A contour vectorization processing unit that obtains contour vector data and obtains circumscribed rectangular data circumscribing the contour vector as a character candidate, and a horizontal, a histogram creation unit that calculates one-dimensional histograms in each of the vertical and diagonal directions; and a histogram creation unit that extracts as character string candidates data that is equal to or higher than a predetermined threshold value from among the data found in the histogram creation unit; Therefore, the height h of a character string candidate and the distance d between adjacent character candidates in the character string candidate are d
A character string extraction unit that extracts data with the relationship <h×α (α is a constant) as a character string, and a function that integrates multiple adjacent character candidates in the character string extracted by the character string extraction unit. What is claimed is: 1. A drawing reading device comprising: a character cutting section for cutting out characters from a character string; and character recognition is performed based on character data cut out by the character cutting section.