JPH10334188A - Type face identification device, type face identification method and information storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily and accurately identify the type face of the character even for the character images of various type faces by comparing the image of the stroke tip part of an extracted character with stroke tip shape models for the respective type faces prepared beforehand and identifying the type face of the character. SOLUTION: A document to which the character which is a type face identification object is written is read by an image input part 2 and it is fetched into a memory 2 as a document image first. Then, only the character image is segmented from the document image by a character segmentation processing part 3 and a character rectangle segmentation processing for obtaining the coordinate of the circumscribing rectangle area is performed. The segmented respective character images are numbered in an order. Then, a number (i) for searching the respective character images is initialized to '1'. Thereafter, the type face of an (i)-th character is judged. After it is judged, the number (i) is increased for '1', then the presence of the (i)-th character is confirmed, and when it is present, a similar processing is performed for the next character.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a typeface identification device for identifying a typeface (font) of a character, a typeface identification method, and an information storage medium.

[0002]

2. Description of the Related Art Conventionally, for example, Japanese Unexamined Patent Publication No. 6-208649 discloses a technique of estimating the character line width in the vertical and horizontal directions of a character.
A typeface identification technique for identifying whether the typeface of a character is Mincho type or Gothic type based on the ratio of these line widths is disclosed. More specifically, this typeface identification technique estimates horizontal and vertical character line widths based on the mode (mode) of horizontal and vertical run-length histograms of a character image, and estimates these line widths. The ratio is used to identify whether the typeface of a character is Mincho or Gothic.

[0003]

However, in the above-described conventional typeface identification technology, most of the strokes constituting characters such as "middle" and "field" are horizontal or vertical straight lines, and noise is included in the image. There is a problem that a font cannot be satisfactorily identified only when there is no font. In other words, most character fonts have diagonal strokes, and in the case where diagonal strokes exist in characters, the above-described conventional typeface identification technology may cause the run-length histogram peak (mode) to be incorrect. As a result, the correct line width cannot be detected, so that the typeface of most characters cannot be accurately identified, which is not suitable for practical use.

The present invention makes it possible to easily and accurately apply character typefaces to character images containing oblique strokes, character images having noise, crushing, blurring, and character images of various fonts (fonts). It is an object of the present invention to provide a typeface identification device, a typeface identification method, and an information storage medium that can be identified by a user.

[0005]

In order to achieve the above object, according to the present invention, an image of a leading end portion of a character stroke is extracted from a character image, and a stroke leading end portion of the extracted character is extracted. Character images are identified by comparing the image of the part with a stroke tip shape model prepared for each typeface prepared in advance. This makes it possible to accurately identify the typeface (font) of the character in the character image.

[0006]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration example of a typeface identification device according to the present invention. Referring to FIG. 1, this typeface identification apparatus includes an image input unit 1 that reads a document as, for example, a binary image (black and white pixel images), and a memory 2 that stores a document image and the like read by the image input unit 1. A character cutout processing unit 3 for extracting a character image from a document image, a font type identification unit 4 for identifying a character type (font) of the character image cut out by the character cutout processing unit 3, The control unit 5 includes a control unit 5 that performs control and a result output unit 6 that outputs a result of character type identification performed by the type identification unit 4.

Here, the character cutout processing section 3 cuts out one character image from the document image, for example, as shown in FIG. That is, in the example of FIG. 2, one character image (the character “K” in the example of FIG. 2) is cut out as a circumscribed rectangular area AR of the character.

Further, the typeface identification section 4 includes a stroke leading end extracting means 8 for extracting an image of a leading end of a character stroke in a character image, and an image of a stroke leading end of the character extracted by the stroke leading end extracting means 8. Is compared with a stroke tip shape model prepared for each typeface prepared in advance to identify the typeface of the character.

FIG. 3 is a diagram showing a first configuration example of the stroke leading end extraction means 8 of the typeface identification unit 4 in FIG. In the example of FIG.
A thinning unit 11 for thinning a character image, an endpoint extracting unit 12 for extracting an endpoint from the thinned image, and an endpoint extracting unit 12
And a stroke leading portion extracting unit 13 for extracting, as a leading stroke portion, a character image within a circle having a radius r with the end point position extracted as the center.

Here, the end point extracting unit 12 and the stroke leading end part extracting unit 13 can extract the end points and leading end parts of all the strokes constituting the character as a first example of extraction.

Alternatively, as a second extraction example, the end point extracting unit 12 and the stroke leading end portion extracting unit 13 determine the end point and the end point of a specific (for example, one having a predetermined inclination) stroke among the strokes constituting the character. It is also possible to extract only the tip part.

Next, the processing operation of the typeface identification apparatus having the configuration shown in FIG. 1 will be described with reference to the flowcharts shown in FIGS. FIG. 4 is a flowchart for explaining the entire processing operation, and FIG. 5 is a flowchart of processing for determining a font (font) when the stroke leading end extraction means 8 of the font identification unit 4 is as shown in FIG. Fonts (fonts) in the processing operation of FIG.
9 is a flowchart illustrating an example of a process of determining ().

Referring to FIG. 4, first, at step S10
In step 1, the image input unit 1 reads a document (for example, a manuscript) in which characters to be typeface-identified are described, and takes the document as a document image into the memory 2. Then, step S1
In step 02, the character extracting unit 3 extracts only the character image from the document image, and performs character rectangle extracting processing for obtaining the coordinates of the circumscribed rectangular area. In this way, clipping is performed for each character image included in the character image,
Numbering is sequentially performed on the cut-out character images.
Next, in step S103, a number (character counter) i for searching each character image is initialized to "1".

Thereafter, in step S104, the typeface (font) of the i-th character is determined.

Typeface (font) in step S104
Is determined, for example, as shown in FIG. Note that the processing example of FIG. 5 is for extracting the leading end portions of all the strokes constituting the character according to the above-described first extraction example. Referring to FIG. 5, first, in step S201,
In step S202, the character image is thinned.
Then, endpoints are extracted from the character image (skeleton image) thinned in step S201, and all the endpoints are stored in the memory 2. At this time, the extracted end points are sequentially numbered and stored. Next, in step S203, the number j for searching for an end point is initialized to "1".

Next, in step S204, a black pixel within a circle having a radius r with the j-th end point as the center is obtained as the leading end of the stroke.

Thereafter, at step S205, the number j
Is incremented by "1". In step S206, it is determined whether or not the j-th end point exists. If the j-th end point exists, the process returns to step S204, and the same processing as described above (in the character (The process of extracting the leading end of one stroke).

In this manner, the leading end portions of the strokes corresponding to all the end points stored in the memory 2 are sequentially obtained in step S202, and when the j-th end point is no longer present (step S206). When the processing is completed), in step S207, the stroke end portion extracted in this one character image is compared with the tip shape of all the fonts prepared in advance, and the font of the i-th character is
(Font). For example, whether the font of the i-th character is Mincho or Maru Gothic,
Determine whether it is a square Gothic body.

At step S104 in FIG. 4, the typeface (font) of the i-th character is
After the determination in steps S01 to S207, in step S105 of FIG. 4, the number i is incremented by "1". Then, in step S106, it is determined whether or not the i-th character exists. If step S104
Then, the same processing as described above (processing for determining the typeface (font) of this character) is performed for the next character.

As described above, for each character image included in the document image input in step S101,
(Font) determination processing is sequentially performed, and step S
When the i-th character no longer exists in 106 (when the processing of determining the typeface has been completed for all the character images), all the processing ends.

In the example of FIG. 5, the typeface (font) is determined using the leading end of all strokes in accordance with the first extraction example. It is also possible to determine the font (font) using only the leading end of the stroke.

FIGS. 6 (a), 6 (b) and 6 (c) show the leading end of a stroke of a character whose typeface (font) may be Mincho, round Gothic and square Gothic, respectively. .
In FIGS. 6A, 6B, and 6C, a circle is a range that is extracted centering on an end point of the thinned image. FIG. 6 (a),
As can be seen from (b) and (c), the fonts (fonts) are Mincho, Maru Gothic, and Square Gothic, respectively.
The shapes of the leading ends of the strokes are different from each other, and therefore, the typeface (font) can be identified based on the difference in the shape of the leading end of the stroke.

FIG. 7 is a diagram showing a second configuration example of the stroke leading end extraction means 8 of the typeface identification unit 4 in FIG. In the example of FIG.
A contour tracing unit 21 for tracing the contour of the character image, a curvature calculating unit 22 for calculating the curvature of the contour, and a stroke tip extracting unit 23 for extracting a stroke tip based on the curvature calculated by the curvature calculating unit 22; have.

Here, the stroke leading end portion extracting section 23
Can extract, as a first extraction example, the leading end portions of all strokes constituting a character.

Alternatively, the stroke leading end portion extracting section 23
As a second example of extraction, it is also possible to extract only the leading end of a specific (for example, one) stroke from each stroke constituting a character.

The calculation of the curvature in the curvature calculation section 22 is performed, for example, as follows. That is, the curvature R of the outline of the character image is given by the following equation when the outline of the character image is y = f (x).

[0027]

R = [(d ² y / dx ² ) / {1+ (dy / dx) ² } ^3/2 ]

In order to calculate the curvature by approximating this equation with a discrete broken line, the concept of averaging can be introduced. That is, as shown in FIG. 8, for the point (x ₀ , y ₀ ) of the contour for which the curvature is to be obtained, k points are respectively taken on the left and right sides, and from these, the following equation (Equation 2) is obtained. Value d
_-, d _+, to calculate such as d _±.

[0029]

(Equation 2)

Alternatively, it can be more simply obtained by the following equation.

[0031]

[Number 3] _{d - = (y 0 -y -k} ) / (x 0 -x -k) d + = (y k -y 0) / (x k -x 0) d ± = (y k / 2 −y _{-k / 2} ) / (x _{k / 2} −x _{-k / 2} )

[0032] Then, the value d obtained in this way _-,
From d ₊ and d _± , (d ² y / dx ² ), (dy
/ Dx).

[0033]

(D ² y / dx ² ) = d ₊ −d ₋ (dy / dx) = d _±

Then, (d ² y / dx ² ) obtained by Equation 4 is used.
By substituting (dy / dx) into Equation 1, the curvature R for the point (x ₀ , y ₀ ) can be obtained.

As described above, when the contour of the character image is tracked and the curvature is calculated for each point of the contour, the stroke leading end extraction unit 23 determines, for example, pixels (e.g., pixels) near the contour point having a large change in curvature. More specifically, for example, a contour pixel before and after this contour point can be extracted as a stroke leading end portion.

FIG. 9 shows a process of judging a font (font) when the stroke leading end extracting means 8 of the font identification unit 4 is as shown in FIG. 9 is a flowchart illustrating an example of a process of determining a typeface (font) in the processing operation of FIG.

That is, when the stroke leading end extraction means 8 of the font identification unit 4 is as shown in FIG. 7, the font (font) determination processing in step S104 of FIG. Done. Note that the processing example in FIG. 9 is for extracting the leading end portions of all strokes constituting a character according to the above-described first extraction example.
Referring to FIG. 9, first, in step S301, the contour of the i-th character image is tracked, and in step S302,
The curvature of each contour of the character image is calculated, and in step S303, contour points having a large change in curvature are extracted, and all extracted contour points (contour points having a large change in curvature) are stored in the memory 2. At this time, the extracted contour points are sequentially numbered and stored. Next, in step S304, the number j for searching for each contour point is initialized to "1".

Thereafter, in step S305, contour pixels before and after the j-th contour point are extracted, and these are determined as the leading end of the stroke.

Thereafter, in step S306, the number j
Is incremented by “1”, and in step S307, it is determined whether or not the j-th contour point exists. If the j-th contour point exists, the process returns to step S305, and for the next contour point,
The same processing as described above (processing for extracting the leading end of one stroke in a character) is performed.

In this manner, the leading end of each stroke corresponding to all the contour points stored in the memory 2 is sequentially obtained in step S303, and when the j-th contour point is no longer present in step S307 (all When the processing of the contour point is completed), in step S308, the tip of each stroke extracted in this one character image is compared with the tip shapes of all the fonts prepared in advance, and the typeface (font) of the i-th character is obtained. Identify. For example, it is determined whether the font of the i-th character is, for example, Mincho, round Gothic, or square Gothic.

In the configuration example of FIG. 7, step S10 of FIG.
In step S4, the typeface (font) of the i-th character is determined, for example, as in steps S301 to S308 in FIG. 9, and in step S105 in FIG.
Then, in step S106, it is determined whether or not the i-th character exists. If there is, the process returns to step S104, and the same processing as described above for the next character (the typeface of this character ( Font)).

As described above, for each character image included in the document image input in step S101,
(Font) determination processing is sequentially performed, and step S
When the i-th character no longer exists in 106 (when the processing of determining the typeface has been completed for all the character images), all the processing ends.

In the example shown in FIG. 9, the typeface (font) is determined by using the leading end of all strokes according to the first extraction example. It is also possible to determine the font (font) using only the leading end of the stroke.

FIGS. 10A, 10B and 10C show fonts (fonts).
Is a diagram showing a leading end of a stroke of a character which may be a Mincho, a round Gothic, and a square Gothic, respectively. As can be seen from FIGS. 10 (a), (b) and (c),
When the typeface (font) is Mincho, Round Gothic, and Square Gothic, respectively, the shape of the tip of the stroke
(Shape formed by contour pixels before and after the contour point with a large change in curvature) are different from each other, and therefore, it is possible to identify a font based on the difference in shape at the tip of the stroke. it can.

As described above, in the present invention, the image of the leading end of the stroke of the character in the character image is extracted, and the extracted image of the leading end of the stroke of the character is compared with the stroke leading end shape model prepared for each typeface. Since the font of the character is identified by using a small program size, the font of the character can be easily and accurately identified with a small program size.

That is, since the shape of the tip of the stroke of the character is used as the characteristic amount, the typeface (font) can be identified with extremely high accuracy and efficiency. For example, the characteristics of a typeface (font) are markedly represented by the shape of the tip of a stroke, so that a font that cannot be identified by a conventional method, such as a Gothic font and a round Gothic font, can be identified.

Further, in the present invention, various fonts can be identified for various characters by pre-registering stroke tip shape models of various characters of various fonts. In other words, a shape model of the tip part of the stroke of all the typefaces (fonts) to be identified is prepared in advance, and the typeface (font) of the target character is identified by measuring the degree of matching (difference) of the shape with that. it can.

In the example of FIG. 3, to extract the leading end of the stroke, the character image is first thinned, its end point is detected, and an image within a predetermined radius r centered on the detected end point is set. By extracting the leading end of the stroke, the leading end of the stroke can be stably extracted.

In the example of FIG. 7, to extract the leading end of the stroke, the contour of the character image is first traced, a portion where the curvature of the contour changes greatly is regarded as the leading end of the stroke, and the image of that portion is regarded as the leading end of the stroke. By extracting the leading end of the stroke, the leading end of the stroke can be stably extracted.

In the above-described example, the case where any one of the Mincho font, the round Gothic font, and the square Gothic font is identified as the font is shown. It is of course possible to identify other fonts in Gothic and square Gothic fonts, and it is also possible to identify other fonts in addition to Mincho, Round Gothic and Square Gothic fonts. is there.

As described above, according to the present invention, it is possible to accurately identify the character typeface (font) of a character in a character image, and based on the identification result of the character typeface (font) thus obtained, For example, it is useful for reproducing a document image.

FIG. 11 is a diagram showing an example of a hardware configuration of the typeface identification apparatus of FIG. Referring to FIG. 11, this typeface identification device is realized by, for example, a personal computer or the like, and controls a CPU 31 that controls the whole, a ROM 32 that stores a control program of the CPU 31, and the like.
A scanner 33 for reading a document as a document image; a scanner 33 for reading a document as a document image;
4, a document image file 35 in which the document image is stored in, for example, a page unit, and a result output device (e.g., a device for outputting information on the result of performing typeface identification on each character image included in the document image) (Display and printer) 36
And

Here, the scanner 34, the document image file 35, and the result output device 36 correspond to the image input unit 1, the memory 2, and the result output unit 6 in FIG. Also, C
The PU 31 includes a control unit 5, a character cutout processing unit 3,
It has the function of the typeface identification unit 4.

The functions of the control unit 5, character cutout processing unit 3, typeface identification unit 4, etc. in the CPU 31 are, for example, software packages (specifically,
The information storage medium can be provided in the form of an information storage medium such as a CD-ROM). For this reason, in the example of FIG. 11, when the information storage medium 40 is set, a medium drive device 41 for driving the information storage medium 40 is provided.

In other words, the typeface identification apparatus of the present invention causes a general-purpose computer system having an image scanner, a display, and the like to read a program code recorded on an information storage medium such as a CD-ROM, and The present invention can also be implemented in an apparatus configuration in which a typeface identification process is executed by a microprocessor of the system. In this case, the information storage medium for storing the typeface identification processing program of the present invention is not limited to a CD-ROM, but may be a ROM, a RAM, an FD, or the like.

[0056]

As described above, according to the first, fourth and fifth aspects of the present invention, an image of a leading end portion of a character stroke is extracted from a character image, and the stroke of the extracted character is extracted. Since the image of the tip portion is compared with the stroke tip shape model prepared for each typeface prepared in advance, the character typeface is identified, so that the character typeface (font) of the character in the character image can be easily and accurately identified with high accuracy. .

According to the second aspect of the present invention, in order to extract the leading end of the stroke, the character image is first thinned, its end point is detected, and the end of the stroke is detected within a predetermined radius r around the detected end point. By extracting the image as the leading end of the stroke, the leading end of the stroke can be stably extracted.

According to the third aspect of the present invention, in extracting the leading end of a stroke, the outline of a character image is first traced, and the part where the curvature of the outline changes greatly is regarded as the leading end of the stroke, and the image of that part is determined. Is extracted as the leading end of the stroke, the leading end of the stroke can be stably extracted.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration example of a typeface identification device according to the present invention.

FIG. 2 is a diagram illustrating an example of one character image.

FIG. 3 is a diagram showing an example of a configuration of a stroke leading end portion extracting means of the typeface identification unit in FIG. 1;

FIG. 4 is a flowchart for explaining a processing operation of the typeface identification device of FIG. 1;

FIG. 5 is a flowchart illustrating a process of determining a typeface (font) when the stroke leading end extraction unit of the typeface identification unit 4 is configured as shown in FIG. 3 in the typeface identification device having the configuration of FIG. 9 is a flowchart illustrating an example of a process of determining a typeface (font).

FIG. 6 is a diagram illustrating a leading end of a stroke of a character in which a font (font) may be a Mincho font, a round Gothic font, or a square Gothic font.

FIG. 7 is a diagram showing another configuration example of the stroke leading end portion extracting means of the typeface identification unit in FIG. 1;

FIG. 8 is a diagram illustrating an example of calculating a curvature.

FIG. 9 is a flowchart illustrating a process of determining a typeface (font) when the stroke leading end extraction unit of the typeface identification unit 4 is configured as shown in FIG. 9 is a flowchart illustrating an example of a process of determining a typeface (font).

FIG. 10 is a diagram showing a leading end portion of a stroke of a character in which a font (font) may be a Mincho font, a round Gothic font, or a square Gothic font.

11 is a diagram illustrating an example of a hardware configuration of the typeface identification device in FIG. 1;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image input part 2 Memory 3 Character cutout processing part 4 Typeface identification part 5 Control part 6 Result output part 8 Stroke tip part extraction means 9 Identification means 11 Thin line thinning part 12 Endpoint extraction part 13, 23 Stroke tip part extraction part 21 Outline tracking Unit 22 curvature calculating unit 31 CPU 32 ROM 33 RAM 34 scanner 35 document image file 36 result output device 40 information storage medium 41 medium drive device

Claims (5)

[Claims] 1. A stroke tip extracting means for extracting an image of a tip of a stroke of a character in a character image, and an image of a stroke tip of the character extracted by the stroke tip extracting means is prepared for each typeface. A type identification device for identifying a typeface of a character as compared with a stroke tip shape model. 2. The typeface identification device according to claim 1, wherein
A typeface identification device, wherein the stroke leading end extracting means thins a character image, extracts an end point from the thinned image, and extracts an image of a leading end of the stroke from the character image based on the extracted end point position. . 3. The typeface identification device according to claim 1, wherein
The typeface identification apparatus, wherein the stroke tip extracting means traces the outline of the character image and extracts an image of the tip of the stroke from the character image based on the curvature. 4. An image of a leading end of a stroke of a character in a character image is extracted, and the extracted image of the leading end of a stroke of the character is compared with a prepared stroke leading end shape model for each typeface to identify a character typeface. A typeface identification method. 5. An image of a leading end of a stroke of a character in a character image is extracted, and the image of the leading end of a stroke of the extracted character is compared with a stroke leading end shape model prepared for each typeface in advance to identify a character typeface. An information storage medium characterized by storing a program for performing the operation.