KR20150137752A - Character recognition method and apparatus thereof - Google Patents

Abstract

The present invention relates to a method and a device for recognizing characters from a low-quality input image including the characters. According to an embodiment of the present invention, the character recognition method includes the steps of: extracting feature values from the input image; using font shape information for each font and the feature value to determine the application font with respect to the characters displayed in the input image; using the font feature information of the application font to divide the input image into respective character area images; comparing the character area images with each character template of the application font to calculate the similarity between the respective character area images and the respective character templates; and using the similarity to determine the characters corresponding to the character area images.

Description

[0001] The present invention relates to a character recognition method and apparatus,

The present invention relates to a character recognition method and apparatus therefor. More specifically, the present invention relates to a method of recognizing an input image having a low quality including characters, that is, an input image having an irregular shape in which a character portion is contaminated by an external factor or a character configuration is uneven, and an apparatus to which the method is applied .

The character recognition method can be largely divided into pattern matching and structure analysis. The former is mainly used for recognition of printed characters, and the latter is mainly used for recognition of handwritten characters. There are feature matching methods and stroke analysis methods as intermediate character recognition methods of both, and they are used in accordance with the purpose or purpose of each.

On the other hand, when characters displayed on the surface of a material such as steel or wood are to be recognized, the conventional character recognition method is difficult to apply. This is because the surface of the material is not unlike the paper, or the characters of the process are corroded or contaminated or damaged. In addition, the number of dot embossing characters also occupies a considerable number, and the shape of characters varies. Therefore, it is required to provide a method of recognizing a character from an image in which the character display state is poor, such as an image including a character displayed on the surface of the material.

Korea Patent Publication No. 2014-0007764

SUMMARY OF THE INVENTION The present invention provides a method and apparatus for recognizing characters from a low-quality input image including characters.

According to another aspect of the present invention, there is provided an image processing method comprising the steps of: determining a font of a character using characteristic values of an input image including characters; dividing the input image into individual character area images; And a method and apparatus for recognizing a character using the comparison result.

It is another object of the present invention to provide a method and apparatus for recognizing a character in an image even when the character is displayed as a dot type.

The technical objects of the present invention are not limited to the above-mentioned technical problems, and other technical subjects not mentioned can be clearly understood by those skilled in the art from the following description.

According to another aspect of the present invention, there is provided a data checking method including extracting a feature value from an input image, extracting feature values from a character displayed on the input image using the feature value and font shape information of each font, Determining an applied font for the application font, dividing the input image into respective character area images using the font feature information of the applied font, comparing the character area image with each character template of the applied font, Computing the similarity for each character template in the character area image, and determining the character corresponding to the character area image using the similarity.

Wherein extracting the feature value comprises applying a predetermined correction routine to the input image to extract a high quality edge from the input image and extracting an edge from the input image . ≪ / RTI >

Wherein the step of determining the applied font comprises the steps of: determining the size of each character displayed on the input image using the extracted edge and the thickness of a stroke constituting the character; And determining the font having the font shape information as the applied font.

The method of claim 1, wherein determining the applied font comprises: determining a dot type font if there are more than a predetermined number of round edges of the extracted edges; grouping adjacent circular edges; Determining the size of each character using the coordinates of pixels constituting the edge, determining the thickness of the dot by using the diameter of each circular edge, and determining the font size corresponding to the size of the character and the thickness of the dot Determining a dot type font having information as the applied font.

The calculating of the similarity may include matching the size of the character area image with the size of the image of each character template of the applied font.

Calculating the degree of similarity by calculating a pixel value matching ratio between the character area image and the image of each character template of the applied font to calculate a degree of similarity, And adjusting the degree of similarity based on whether or not the characteristic requirement of the first embodiment is satisfied.

Wherein the dividing step includes setting a character separation guideline on the input image by using a character spacing, a long pillar, a font image width, and a font image height in the font characteristic information of the applied font, And dividing the input image into respective character area images.

According to another aspect of the present invention, there is provided a data checking method comprising: dividing an input image into individual character area images using font characteristic information; comparing the character area image with each character template; Calculating a similarity for each character template in the character area image, and determining the character corresponding to the character area image using the similarity. Wherein calculating the similarity comprises reflecting whether the character area image satisfies a feature requirement of each character template.

According to another aspect of the present invention, there is provided a character recognition apparatus including an image feature extraction unit for extracting a feature value from an input image, A character region dividing unit for dividing the input image into respective character region images by using the font characteristic information of the applied font; And a character determination unit for calculating a similarity to each character template of the character area image by comparing the character template of each of the applied fonts and determining a character corresponding to the character area image using the similarity.

According to one embodiment, the character recognition apparatus performs predetermined image quality correction for the character area image, performs binarization of the character area image on which the image processing is completed, and outputs the binarized character area image To the character determination unit.

According to the present invention as described above, characters can be recognized from an image having a poor character display state, such as an image including characters displayed on the surface of a material. Particularly, even when there is contamination in a character display portion or in an irregular character whose configuration is not uniform, the effect can be effectively recognized.

Further, there is an effect that a character can be recognized from an image including a character represented by a dot type.

1 is a diagram for explaining the effect of the present invention.
2 is a flowchart of a character recognition method according to an embodiment of the present invention.
FIG. 3 is a diagram showing the results of image classification performed at some stages in the character recognition method shown in FIG. 2. FIG.
FIG. 4 is a detailed flowchart for explaining some steps of the character recognition method shown in FIG. 2 in detail.
FIG. 5 is a conceptual diagram for explaining some steps of the character recognition method shown in FIG. 2. FIG.
FIG. 6 is an example of a user interface displayed as a result of performing the method described with reference to FIGS.
7 is a diagram illustrating a character recognition apparatus according to another embodiment of the present invention.
8 is a block diagram of a character recognition apparatus according to another embodiment of the present invention.
FIG. 9 is a detailed block diagram of some blocks of the character recognition apparatus shown in FIG. 8; FIG.
10 is a diagram illustrating a hardware configuration of a character recognition apparatus according to another embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

Unless defined otherwise, all terms (including technical and scientific terms) used herein may be used in a sense commonly understood by one of ordinary skill in the art to which this invention belongs. Also, commonly used predefined terms are not ideally or excessively interpreted unless explicitly defined otherwise. The terminology used herein is for the purpose of illustrating embodiments and is not intended to be limiting of the present invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. It is noted that the terms "comprises" and / or "comprising" used in the specification are intended to be inclusive in a manner similar to the components, steps, operations, and / Or additions.

In this specification, a logic or a routine refers to a series of instructions that can be executed by a processor, and is not limited to being written by a specific programming language.

First, referring to FIG. 1, a character recognition method according to embodiments of the present invention and an image including characters and recognition results thereof that can be recognized by the apparatus are presented.

As already mentioned, related information, such as the serial number of the material, can be embossed on the surface of the material, such as wood, plate, or steel. A character to be embossed may be a general character composed of linked strokes, but may be a dot type character for efficiency of embossing. Also, depending on various situations occurring in the process, the embossed characters may have an irregular shape. In addition, the surface of the material may be uneven from the paper, or character portions of the material may be corroded, contaminated, or damaged due to the nature of the process. Fig. 1 shows an image 10 of a dot-type image of a surface of a material embossed with characters.

Embodiments of the present invention may recognize (20) characters from an image 10 as shown in FIG. Hereinafter, embodiments of the present invention will be described. However, the embodiments of the present invention are not limited to recognizing embossed characters such as shown in FIG. 1, or recognizing dot type characters, and embodiments of the present invention are not limited to those embodied in a method of recognizing characters But should be understood on the basis of the configuration of a device that recognizes each operation or character.

Referring to FIG. 2, a character recognition method according to an embodiment of the present invention will be described. Although FIG. 2 shows a flowchart for the case where an input image has already been generated, this embodiment may further include a step of photographing a subject in which characters are printed or characters are embossed to generate the input image . The present embodiment may further include a step of displaying the photographed image and a step of designating a character display area on the displayed image from the user. When a character display area is designated by a user, it is possible to reduce the possibility of error in character recognition by recognizing the character only in the character display area.

Feature values are extracted from the input image (S100). The extracted feature value may be used as a basic data for determining a font of a character in the input image (S102). A pre-process may be performed on the input image so that the feature value can be extracted well in the input image before extracting the feature value. For example, the pre-processing may include level adjusting. For example, you can lighten the hue and adjust the contrast.

In this embodiment, various elements used as feature values of an image can be extracted. For example, an edge or contour of the input image may be extracted. In addition, a histogram of color values of pixels constituting the input image may be calculated.

The extracted edges of the input image may be used to determine the size of each character included in the input image. Generally, since the outline of a character constitutes a remarkable color change compared to the surroundings, when the predetermined edge extraction logic is applied to an image including a character, the outline of the character forms an edge. Therefore, for an alphabet except a digit (0 to 9) and a lowercase letter i, j in which a stroke constitutes one character, the area occupied by one edge is regarded as a character area. (Also, characters composed of strokes including dot-shaped characters can be separated according to the input image quality, so if the interval of the edge is within the reference value, it is regarded as the same character area.)

In this specification, an edge refers to a set of pixels that are made up of adjacent pixels and that are not less than the reference number and size.

Position correction can be performed on the extracted edge. For example, edge outline removal and parallel or slope correction may be performed.

The size of a character can be defined as width and height.

For example, if the x coordinate minimum coordinate of one edge is 30, the x coordinate maximum coordinate is 100, the y coordinate minimum coordinate is 100, and the y coordinate maximum coordinate is 200, the character indicated by the connected edge The size can be determined to be 70 in the width and 100 in the width.

Hangul, Kanji, and Alphabets For lowercase i, j, and other characters, two or more strokes constitute one character. Even in this case, the strokes constituting one character will be adjacent to each other. Using this property, an edge group composed of adjacent edges to the extent satisfying a predetermined criterion is regarded as constituting one character, and an area occupied by the edge group is regarded as a character area.

For example, if the x-axis minimum coordinate of the first edge is 40, the x-axis maximum coordinate is 70, the y-axis minimum coordinate is 48 and the y-axis maximum coordinate is 108, If the x-axis minimum coordinate of the pixel occupied by the second edge constituting the character is 74, the x-axis maximum coordinate is 84, the y-axis minimum coordinate is 40 and the y-axis maximum coordinate is 115, the first and second edges It is determined that the size of the constituent characters is 44 in the horizontal direction and 75 in the vertical direction.

On the other hand, when the number of characters is plural, the size of each character can be determined differently. In this case, the average value of each character size can be used.

The edge of the extracted input image can also be used to determine the thickness of a stroke constituting each character included in the input image. Since the edge can be seen as the outline of each stroke, the thickness of the stroke can be obtained by analyzing the coordinates of each pixel constituting the edge. A difference may occur in each stroke due to an error in the thickness of the stroke. In this case, the average value of the thickness of each stroke can be used.

If the size of the character and the thickness of the stroke are obtained, inquiry is made as to whether or not the font shape information of each of the previously stored fonts matches the character size and the thickness of the stroke. To this end, the present embodiment may further include storing font shape information of each font to be recognized in advance.

The font shape information may include a font type, a character spacing, a major plane, character size information, stroke information (dot thickness information in the case of a dot type font), and the like. The font type indicates whether it is a dot type.

According to some embodiments, the ratio between the size of a character included in an input image and the thickness of a stroke may be further used as a factor for determining an applied font. For example, if the ratio between the character width of the input image and the width of the stroke is 15: 1 and the ratio between the height of the character and the thickness of the stroke is 18: 1, .

In other words, in the present exemplary embodiment, a feature value such as an edge is extracted from an input image (S100), and a font applied to a character displayed on the input image is determined using the feature value and font shape information of each font Wherein the step of determining the applied font comprises the steps of: determining a size of each character displayed on the input image using the extracted feature value and a thickness of a stroke constituting the character; And determining the font having the font shape information corresponding to the size of the character and the thickness of the stroke as the applied font. In this case, the step of determining the size of the character and the thickness of the strokes constituting the character displayed on the input image using the feature value may include determining the width of the edge of the edge, And determining the size of each character and the thickness of the stroke using the coordinates of the pixels constituting each of the edges that are not subjected to the noise processing.

In this embodiment, a histogram of the color values of the input image can be applied to more accurately extract the edges representing the outline of the characters. The character color will have a high frequency in the histogram, so long as each stroke of the character is composed of the same or similar color, and one stroke is composed of the same or similar color. Using this point, an edge can be extracted with respect to a character constituent candidate pixel having a color value occupying a ratio equal to or larger than a predetermined value in the histogram and adjacent pixels among the pixels constituting the input image. In this case, an edge is extracted from the outline of the character, and the edge is treated as the outline of the character, thereby preventing the size of the character from being erroneously determined.

Hereinafter, a font determination method according to an embodiment of the present invention in which a dot type character is included in an input image will be described.

If a dot type character is included in the input image, the outline of each dot forms an edge. Therefore, if edge extraction logic is applied to an image including a dot type character, a plurality of circular edges will be extracted. For example, when edge extraction logic is applied to an image including a first character composed of 20 dots and a second character composed of 18 dots, 38 circular edges will be extracted. Therefore, in the present embodiment, when the number of the circular edges is equal to or greater than the predetermined number as a result of the edge extraction, it is determined that the applied font is one of the dot type fonts.

The shape of the dot is mostly circular, but the shape of the dot may be a non-circular shape, for example, a square, a triangle, or the like. In this case, this embodiment can be generalized to determine the applied font as a dot type font when there are more than a predetermined number of edges of the same shape.

Since the dots constituting one character are adjacent to each other to a degree matching with a predetermined reference, a group is formed of dots adjacent to each other, and the size of a character indicated by the dot group is determined by using a region occupied by the formed dot group . ≪ / RTI >

For example, if the x-axis minimum coordinate of the pixel occupied by the outline edge of each dot included in one dot group is 30, the x-axis maximum coordinate is 100, the y-axis minimum coordinate is 100, and the y- , And the size of the character indicated by the dot group may be determined to be 70 in the horizontal direction and 100 in the vertical direction.

The edge of the extracted input image can also be used to determine the thickness of the dots constituting each character included in the input image. Since the edge can be seen as the outline of each dot, the thickness of the dot can be obtained by analyzing the coordinates of each pixel constituting the edge. A difference may occur in each dot due to an error in the thickness of the dot or the like. In this case, the average value of the thicknesses of the dots can be used.

According to some embodiments, the ratio between the size of the characters included in the input image and the thickness of the dot may be further used as a factor for determining the font to be applied. For example, when the ratio between the character width and the dot thickness of the input image is 22: 1 and the ratio between the character height and the stroke width is 28: 1, it is determined whether there is such a ratio in the font shape information .

In summary, the step of determining the applied font comprises the steps of grouping adjacent circular edges, determining the size of each character using the coordinates of the pixels constituting each circular edge belonging to a group, Determining a thickness of the dot using the diameter of the circular edge; and determining a dot type font having font size information and font shape information corresponding to the size of the character and the size of the dot as the application font.

If the font applied to the characters included in the input image is determined through the above-described operation, the input image is divided into the respective character area images using the font shape information of the applied font (S104). As already mentioned, since the font shape information includes the character spacing and the length information and the size information of each character, even if the character spacing and the length information are further reflected in the area information of each character already obtained through edge extraction A character separation guideline 30 such as shown in Fig. 3 may be set in the input image. Thereafter, the input image may be divided into individual character region images based on the character separation guideline.

Next, a step of recognizing any character for each character area image is performed (S106). At this time, since the already applied font has been determined, the character area image is compared with each character template of the applied font to calculate the similarity to each character template of the character area image, and based on the similarity, The character corresponding to the area image is determined. Hereinafter, this will be described in more detail with reference to FIG.

First, image preprocessing is performed on the character area image to improve recognition accuracy. A series of image processing such as increasing sharpness is performed, and binarization may be performed (S160). By binarization, the character area image has a pixel value of 0 or 1.

Next, the size of the character area image is matched with the size of the image of each character template of the applied font (S162).

Next, a pixel value matching ratio between the character area image and the image of each character template of the applied font is calculated to calculate the similarity (S164). Since the size of the character area image is identical to the size of the character template image, pixel values between pixels at the same position can be matched 1: 1. The character template image is also assumed to be composed of pixels having a pixel value of 0, 1.

Next, the degree of similarity is adjusted according to whether or not the character area image satisfies the feature requirement of each character template of the applied font (S166). Hereinafter, the operation of comparing the character area image with each character template of the applied font to calculate the similarity to each character template of the character area image will be described in more detail.

In FIG. 5, a virtual character area image composed of 5X8 (wide X height) pixels and a '0' template of a specific font are shown. According to the embodiments of the present invention, as the information on the recognition target font, in addition to the font shape information already described, the template image and feature requirement information of each character can be further stored. Figure 5 shows a template image 36 and feature requirement information 37 for the number ' 0 '. The feature requirement information 37 indicates a condition that a pixel value of a specific area in the character area image must be satisfied in order to be recognized as a corresponding character.

First, the similarity of the pixel values of the character area image 35 and the template image 36 to the number '0' is calculated. At this time, it is judged whether or not the pixel values of the pixels at the same position coincide with each other, and the ratio of the pixels whose pixel values match can be the similarity. Next, an evaluation is made as to whether the character area image 35 satisfies the feature requirement of the template for the number '0'.

In FIG. 5, three virtual feature requirements for the number '0' are illustrated. The first feature requirement requires that the pixels of the first region 31 be all zeros. In the case of the character area image 35 to be recognized, this condition is not satisfied. The second feature requirement requires that at least one of the pixels of the second region 32 be one. In the case of the character area image 35, this condition is satisfied. The third feature requirement requires that at least one of the pixels of the third region 33 be one. In the case of the character area image 35, this condition is satisfied. That is, in the case shown in FIG. 5, since the character area image 35 does not satisfy only the first feature requirement, in the similarity calculated through comparison between images, the value given to the first feature requirement is subtracted , It can be deducted by the ratio given to the first feature requirement.

That is, each character template may have a plurality of feature requirements, and each feature requirement may have a different degree of importance, so that if the feature requirements are not satisfied, Can be assigned differently.

According to one embodiment, the calculated degree of similarity may be increased upon satisfaction of the feature requirement. That is, when the similarity is adjusted according to whether or not the character area image satisfies the feature requirement of each character template of the applied font, the degree of similarity may be subtracted or the degree of similarity may be increased when the feature requirement is satisfied.

According to another embodiment, with the above feature requirement, pixel value matching for a specific area can be used. For example, a character template of the number '0' may include a region 31 requiring a pixel value of '0' (for example, white), a pixel value of '1' (32, 33), which require color (meaning color). At this time, the evaluation result on the feature requirement can be calculated by how much the character area image 35 satisfies the designated pixel value in the area 31, 32, 33. At this time, the result of the evaluation is not determined only as satisfaction or dissatisfaction, but the adjustment value can be determined according to the degree of satisfaction. The adjustment value may have a positive value or a negative value. The adjustment value is reflected in the pre-calculated degree of similarity. According to the present embodiment, there is an effect that the character recognition can be performed flexibly even when the character printing or the embossing state is irregular.

One character area image 35 may calculate the similarity for all characters included in the applied font. As a result, it can be recognized that the character having the highest similarity is the character displayed in the character area image (S168).

Again, referring back to Fig. The character recognition operation for the character area image described above is repeated for all the character area images included in the input image. As a result, all characters included in the input image can be recognized.

The character recognition method according to the present embodiment checks an applied font according to a feature value of an input image, separates an input image into a character area image, compares each character area image with a character template of the applied font, By recognizing characters through similarity ranking with the template, even if the characters are displayed in the input image is poor, the characters can be correctly recognized. In addition, even if a dot type character is recognized by a plurality of dots correctly by edge extraction, it is recognized that the font is a dot type font, and the characters are accurately displayed by comparing dot type font shape data, images for each character, There is a recognizable effect.

The recognized character or string may be displayed, stored, or transmitted to an external device via a network interface (S108). The apparatus to which the character recognition method according to the present embodiment is applied can provide a user interface as shown in FIG. The user interface may include an area 40 for displaying an input image so that the input image can be recognized, an area 50 for displaying the text to identify the recognized result, and a menu area 60 for processing the recognized result . Through the user interface, the user can collectively process the shooting, recognition, and recognition result management of the material-related information displayed on the surface of the material or embossed.

The character recognition apparatus according to another embodiment of the present invention will be described with reference to FIG. The character recognition device 70 according to the present embodiment may be a hand-held terminal having a handle for easy gripping. The character recognition apparatus 70 is provided with an image sensing unit 71 for sensing an image and a built-in display 72 for displaying a user interface such as shown in Fig. 6, a control unit (not shown), a storage unit And an interface (not shown). The control unit is provided with a processor for executing a computer program for performing the method described with reference to FIGS. 2 to 5 and a storage unit for storing an execution code of the computer program.

A character recognition apparatus according to another embodiment of the present invention will be described with reference to FIG. The character recognition apparatus 100 according to the present embodiment may be a terminal device such as a notebook PC, a desktop PC, a tablet PC, or a smart phone. The character recognition apparatus 100 according to the present embodiment includes a storage unit 104 composed of an image acquisition unit 102, a nonvolatile memory or an auxiliary storage device such as a HDD, a storage unit 104 connected to an external device via a network, A display 110, and a character recognition unit 106. The network interface 108 may include a network interface 108, a display 110,

The image acquiring unit 102 may include, for example, a lens, an image sensor, and a drive controller. The image acquiring unit 102 converts the optical signal input through the lens into digital data, generates an input image, and provides the input image to the character recognizing unit 106.

The character recognition unit 106 is provided with a processor that executes a computer program that performs the method described with reference to FIGS. 2 to 5 and a storage unit that stores an execution code of the computer program. Recognizes the character in the input image, and generates digital output data indicating the recognized character. The character recognition unit 106 stores the digital output data in the storage unit 104 so that the digital output data is stored or provides the digital output data to the display 110 so that the digital output data is displayed, (108).

A character recognition apparatus according to another embodiment of the present invention will be described with reference to FIG. The character recognition apparatus 106 shown in Fig. 9 may be employed as one component in the character recognition apparatus shown in Fig. Accordingly, the character recognition unit 106 has been described with reference to FIG. 9, the character recognition apparatus 106 according to the present embodiment includes an image feature extraction unit 160, a font storage unit 161, a font determination unit 162, a character region dividing unit 163, An image correction unit 164 and a character determination unit 165. [

The image feature extraction unit 160 extracts a feature value by receiving an input image. The feature value may be, for example, an edge, a histogram for a color, or the like.

The font storage unit 161 stores font shape information and a template of each character for a font that supports recognition. The character template includes information on characteristics of character images and characters. The font shape information may include a font type, a character spacing, a major plane, character size information, stroke information (dot thickness information in the case of a dot type font), and the like. The font type indicates whether it is a dot type.

The font determination unit 162 determines an applied font for the character displayed on the input image using the feature value and the font shape information of each font provided from the font storage unit 161. [

The character area dividing unit 163 divides the input image into the respective character area images using the font shape information of the applied font.

The image correcting unit 164 performs predetermined image quality correction on the character area image, performs binarization of the character area image on which the image processing is completed, and outputs the binarized character area image to the character decision unit (Not shown).

The character determination unit 165 compares the character region image with each character template of the applied font to calculate a similarity degree to each character template in the character region image, and calculates a character corresponding to the character region image .

Each component in FIG. 9 may refer to software or hardware such as a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC). However, the components are not limited to software or hardware, and may be configured to be in an addressable storage medium and configured to execute one or more processors. The functions provided in the components may be implemented by a more detailed component or may be implemented by a single component that performs a specific function by combining a plurality of components.

10 is a diagram illustrating a hardware configuration of a character recognition apparatus according to another embodiment of the present invention. The character recognition apparatus 106 of FIG. 9 can be implemented in a hardware structure as shown in FIG.

10, the data processing apparatus 200 according to the present embodiment includes a system bus 202, a processor 212, a random access memory (RAM) 210, a storage 204, An interface unit 206 for connection with the input / output device, and a display 208. [ The computer program code embodying the character recognition method according to the present invention may be stored in the storage 204, loaded into the RAM 210 and executed by the processor 212. [ As a result of execution of the computer program code, output data indicating a character recognition result is stored in the RAM 210 or the storage 204, transmitted to an external device via the network interface 206, or displayed via the display 208 .

The concepts of the invention described above with reference to Figures 1 to 10 can be implemented in computer readable code on a computer readable medium. The computer readable recording medium may be, for example, a removable recording medium (CD, DVD, Blu-ray disk, USB storage device, removable hard disk) . The computer program recorded on the computer-readable recording medium may be transmitted to another computing device via a network such as the Internet and installed in the other computing device, thereby being used in the other computing device.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, You will understand. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive.

Claims (15)

Extracting a feature value from an input image;
Determining an applied font for a character displayed on the input image using the feature value and font shape information of each font;
Dividing the input image into respective character area images using font shape information of the applied font;
Comparing the character area image with each character template of the applied font to calculate a similarity degree to each character template of the character area image; And
And determining the character corresponding to the character area image using the similarity.
Character recognition method. The method according to claim 1,
Wherein determining the applied font comprises:
Determining a size of each character displayed on the input image using the extracted feature value and a thickness of a stroke constituting the character; And
Determining a font having the font shape information corresponding to the size of the character and the thickness of the stroke as the applied font,
Character recognition method. 3. The method of claim 2,
Wherein the step of extracting feature values from the input image comprises:
Extracting an edge from the input image,
Wherein the step of determining the size of each character displayed in the input image using the feature value and the thickness of the stroke constituting the character,
Noise processing the edge having a length less than a predetermined minimum length among the extracted edges; And
And determining the size of each character and the thickness of the stroke using the coordinates of the pixels constituting each edge that are not noise-
Character recognition method. The method of claim 3,
The step of extracting the edge comprises:
Generating a color histogram of the input image; And
And extracting the edge from the pixels constituting the input image, the candidate pixel having a color value occupying a ratio equal to or greater than a predetermined value in the histogram and adjacent pixels thereof,
Character recognition method. The method according to claim 1,
Wherein the step of extracting feature values from the input image comprises:
Extracting an edge from the input image,
Wherein determining the applied font comprises:
Determining a dot type font if there are more than a predetermined number of circular edges among the extracted edges;
Character recognition method. 6. The method of claim 5,
Wherein determining the applied font comprises:
Grouping adjacent circular edges;
Determining the size of each character using coordinates of pixels constituting each circular edge belonging to a group;
Determining the thickness of the dot using the diameter of each circular edge; And
Determining a dot type font having the font shape information corresponding to the size of the character and the thickness of the dot as the application font,
Character recognition method. The method according to claim 1,
The step of calculating the degree of similarity may include:
And matching the size of the character area image with the size of the image of each character template of the applied font.
Character recognition method. The method according to claim 1,
The step of calculating the degree of similarity may include:
Computing a pixel value matching ratio between the character area image and the image of each character template of the applied font to calculate the similarity; And
And adjusting the similarity by whether the character area image satisfies a feature requirement of each character template of the applied font.
Character recognition method. 9. The method of claim 8,
The character template of the applied font has the first feature requirement and the second feature requirement,
Wherein the step of adjusting the degree of similarity comprises:
Wherein the character region image does not satisfy the first feature requirement, subtracts the first value from the calculated similarity, and when the character region image does not satisfy the second feature requirement, Subtracting the figures,
Character recognition method. 9. The method of claim 8,
The feature requirement is that the pixel value of a specific region must satisfy,
Character recognition method. The method according to claim 1,
Wherein the distinguishing step comprises:
Setting a character separation guideline on the input image using the character spacing and the length of the font characteristic information of the applied font; And
And separating the input image into respective character area images based on the character separation guideline.
Character recognition method. Dividing an input image into respective character area images using font feature information;
Comparing the character region image with each character template to calculate a similarity degree to each character template of the character region image; And
Determining a character corresponding to the character area image using the similarity,
The step of calculating the degree of similarity may include:
And reflecting whether the character area image satisfies a feature requirement of each character template.
Character recognition method. A computer program for performing the method of any one of claims 1 to 12, wherein the computer program is recorded. An image feature extraction unit for extracting a feature value from an input image;
A font determining unit for determining an applied font for a character displayed on the input image using the feature value and font shape information of each font;
A character region dividing unit for dividing the input image into respective character region images using font shape information of the applied font; And
And a character determination unit for comparing the character region image with each character template of the applied font to calculate a similarity degree for each character template in the character region image and determining a character corresponding to the character region image using the similarity degree doing,
Character recognition device. 15. The method of claim 14,
An image correction unit that performs predetermined image quality correction for the character area image, performs binarization of the character area image that has been subjected to image processing, and provides the character area image that is binarized to the character determination unit Further included,
Character recognition device.

Images