JP2001118032A - Device and method for character recognition, and computer-readable recording medium with recorded program making computer implement character recognizing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To avoide a decrease in recognition rate due to the input of a low- resolution image and to eliminate the need for conventional confirming operation of image input by interpolating image information of low resolution to a specific level. SOLUTION: The character recognition device which inputs image information and recognizes characters of the image information is equipped with a character interpolation part 11 which converts the inputted image information of low resolution to specific resolution through image interpolation and a character recognition part 12 which recognizes the image information converted by the character interpolation part 11.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character recognizing device and a character recognizing method, and a character recognizing method used for image cutting of an image reading device such as an optical character reading device for reading images such as characters and figures. The present invention relates to a computer-readable recording medium on which a program for causing a computer to execute the method is recorded.

[0002]

2. Description of the Related Art Conventionally, image input of a character recognition device has been performed by a scanner device or a facsimile device. In recent years, with the spread of digital cameras, image input from digital cameras has also increased. When using a digital camera as an image input means of a character recognition device,
Since the resolution is not sufficient as compared with a scanner device or a facsimile device, it is difficult to capture the entire document to be recognized in order to capture a character size suitable for the character recognition device.

[0003] In addition, when the entire image cannot be photographed, the user does not know how far the character size is preferable for the character recognition device. Techniques such as displaying information to be shown are also known.

[0004]

However, photographing a user while checking whether or not a preferable character size for the character recognition device can be obtained each time is troublesome in the confirmation operation and the like. If there is an input with a low resolution, the recognition rate is reduced when performing image recognition.

The present invention has been made in view of the above, and by interpolating low-resolution image information to a predetermined level, it is possible to avoid a decrease in recognition rate due to input of low-resolution images. It is another object of the present invention to eliminate the need for a conventional image input confirmation operation.

[0006]

According to a first aspect of the present invention, there is provided a character recognition apparatus for inputting image information and performing character recognition of the image information. The image processing apparatus includes image interpolating means for converting the input low-resolution image information to a predetermined resolution by image interpolation, and recognizing means for recognizing the image information converted by the image interpolating means.

According to the first aspect, the low-resolution image information is converted into a predetermined resolution by the image interpolation means by the image interpolation, and the low-resolution image is recognized by recognizing the converted image information. Even when input, an image having a resolution sufficient for image recognition is input to the recognition means.

According to a second aspect of the present invention, in the character recognition apparatus for inputting image information and performing character recognition of the image information, the input low-resolution color image information is subjected to image interpolation. The image processing apparatus includes an image interpolating unit that converts the image into a predetermined resolution, and a recognizing unit that recognizes the color image information converted by the image interpolating unit.

According to the second aspect, the low-resolution color image information is converted into a predetermined resolution by image interpolation by the image interpolation means, and the converted color image information is recognized, whereby the low-resolution color image information is recognized. Even when a color image is input, an image having a resolution sufficient for image recognition is input to the recognition unit.

According to a third aspect of the present invention, in the character recognition apparatus for inputting image information and performing character recognition of the image information, the input low-resolution color image information is converted into a gray image. Gray conversion means for converting, gray image information converted by the gray conversion means, image interpolation means for converting to a predetermined resolution by image interpolation,
Binarizing the image information converted by the image interpolation means;
And a recognizing means for recognizing the image information binarized by the binarizing means.

According to the third aspect, the low-resolution color image information is converted into a gray image by the gray conversion means,
The image information converted into the gray image is converted into a predetermined resolution by image interpolation by an image interpolation unit, and the color image information having the converted resolution is binarized by a binarization unit, and the white / black pixels are converted. By recognizing an image that has been subjected to good image processing (for example, smoothing processing, edge enhancement processing, etc.) that has been binarized, the image recognition rate is improved even when a low-resolution color image is input. An image of a sufficient level to be input is input to the recognition means.

According to a fourth aspect of the present invention, in the character recognition apparatus for inputting image information and performing character recognition on the image information, the input low-resolution gray image information is converted into an image interpolated image. Interpolating means for converting the gray image information converted by the image interpolating means into binary, and recognizing the image information binarized by the binarizing means. And recognition means.

According to the present invention, the low-resolution gray image information is converted into a predetermined resolution by image interpolation by the image interpolation means, and the gray image information whose resolution has been converted is converted into binary data by the binarization means. A low-resolution gray image is input by recognizing an image that has been binarized and subjected to good image processing (eg, smoothing processing, edge enhancement processing, etc.) in which the white / black pixels have been binarized. In this case, an image of a level sufficient to improve the image recognition rate is input to the recognition unit.

According to a fifth aspect of the present invention, in the character recognition apparatus for inputting image information and performing character recognition of the image information, the input low-resolution color image information is subjected to image interpolation. Interpolating means for converting the color image information converted by the image interpolating means into a predetermined resolution, binarizing means for binarizing the color image information converted by the image interpolating means, and recognizing the image information binarized by the binarizing means. And recognition means.

According to the fifth aspect, the low-resolution color image information is converted into a predetermined resolution by image interpolation by the image interpolation means, and the color image information whose resolution has been converted is converted into binary data by the binarization means. By recognizing an image that has been binarized and subjected to good image processing (eg, smoothing processing, edge enhancement processing, etc.) in which the white / black pixels have been binarized, a low-resolution image is input. In this case, an image of a level sufficient to improve the image recognition rate is input to the recognition unit.

In a character recognition method according to a sixth aspect of the present invention, in the character recognition method for inputting image information and performing character recognition of the image information, the input low-resolution image information is determined by image interpolation. And a recognition step of recognizing the image information converted in the image interpolation step.

According to the sixth aspect, the low-resolution image information is converted into a predetermined resolution by the image interpolation in the image interpolation step, and the image information having the converted resolution is recognized. Even when input, an image having a resolution sufficient for image recognition is input to the recognition process.

In a character recognition method according to a seventh aspect of the present invention, in the character recognition method for inputting image information and performing character recognition of the image information, the input low-resolution color image information is subjected to image interpolation. The method includes an image interpolation step of converting the image into a predetermined resolution, and a recognition step of recognizing the color image information converted in the image interpolation step.

According to the present invention, the low-resolution color image information is converted into a predetermined resolution by image interpolation in the image interpolation step, and the converted color image information is recognized, whereby the low-resolution color image information is recognized. Even when a color image is input, an image having a resolution sufficient for image recognition is input to the recognition process.

In the character recognition method according to the present invention, in the character recognition method for inputting image information and performing character recognition of the image information, the input low-resolution color image information is converted into a gray image. A gray conversion step of converting, and an image interpolation step of converting the gray image information converted in the gray conversion step to a predetermined resolution by image interpolation,
2 for binarizing the image information converted in the image interpolation step
And a recognition step of recognizing the image information binarized in the binarization step.

According to the present invention, the low-resolution color image information is converted into a gray image by the gray conversion step.
The image information converted to the gray image is converted to a predetermined resolution by image interpolation in an image interpolation step, and the color image information having the converted resolution is binarized in a binarization step, and the white / black pixels By recognizing an image that has been subjected to good image processing (for example, smoothing processing, edge enhancement processing, etc.) that has been binarized, the image recognition rate is improved even when a low-resolution color image is input. An image of a sufficient level to be input to the recognition process.

In a character recognition method according to a ninth aspect, in the character recognition method for inputting image information and performing character recognition of the image information, the input low-resolution gray image information is subjected to image interpolation. An image interpolating step of converting the gray image information converted in the image interpolating step into a binary value, and recognizing the image information binarized in the binarizing step. And a recognition step.

According to the ninth aspect, the low resolution gray image information is converted to a predetermined resolution by image interpolation in an image interpolation step, and the gray image information whose resolution has been converted is converted to a binary image in a binarization step. A low-resolution gray image is input by recognizing an image that has been binarized and subjected to good image processing (eg, smoothing processing, edge enhancement processing, etc.) in which the white / black pixels have been binarized. In such a case, an image of a sufficient level to improve the image recognition rate is input to the recognition process.

According to a tenth aspect of the present invention, in the character recognition method for inputting image information and performing character recognition of the image information, the input low-resolution color image information is subjected to image interpolation. An image interpolating step of converting the image data into a predetermined resolution, a binarizing step of binarizing the color image information converted in the image interpolating step, and recognizing the image information binarized in the binarizing step. And a recognition step.

According to the tenth aspect, the low-resolution color image information is converted into a predetermined resolution by image interpolation in an image interpolation step, and the color image information having the converted resolution is converted into a binary image in a binarization step. Value, and 2 for that white / black pixel
By recognizing an image that has been subjected to good valued image processing (eg, smoothing processing, edge enhancement processing, etc.), the image recognition rate is improved even when a low-resolution image is input. Are input to the recognition process.

[0026] In a computer-readable recording medium according to claim 11, the computer-readable recording medium according to claims 6 to 10 is provided.
A program for causing a computer to execute the character recognition method according to any one of the embodiments is recorded.

According to the eleventh aspect, claims 6 to 10
A program for causing a computer to execute the character recognition method according to any one of the first to third aspects.
10 can be realized by a computer.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a character recognition device and a character recognition method according to the present invention and a computer-readable recording medium storing a program for causing a computer to execute the character recognition method will be described with reference to the accompanying drawings. Refer to and describe in detail. The present invention is not limited by the embodiment.

(Embodiment 1) In this embodiment 1,
An example will be described in which when a low-resolution image is input, resolution conversion is performed by image interpolation, and the image after the resolution conversion is recognized.

FIG. 1 is a block diagram showing the configuration of the character recognition device according to the first embodiment of the present invention. The character recognizing device 10a includes a character interpolating unit 11 for converting an input low-resolution image into a predetermined resolution by image interpolation, a character recognizing unit 12 for character recognizing image data whose resolution has been converted by the character interpolating unit 11, It has.

The operation of the character recognition device 10a will be described with reference to the flowchart shown in FIG. First, it is determined whether or not the resolution of the input image data to be recognized is equal to or higher than a predetermined level (step S11). The determination of the resolution is made based on, for example, resolution information stored in the header portion of the image.

If it is determined in step S11 that the resolution of the input image data is not higher than the predetermined level, that is, the resolution is low, the character interpolating unit 11 converts the input image data into the predetermined resolution (step S12).
To enter. The character interpolation (resolution conversion) unit 11 uses a method of substituting the closest pixel value before interpolation, a method of performing linear interpolation, and a method of performing cubic interpolation to determine a pixel after interpolation. Create an image of

After converting the low-resolution image data to a predetermined resolution in step S12, the character recognition unit 12
Performs character recognition on the image data whose resolution has been converted (step S13). In the character recognition unit 12, for example,
A generally known method based on pattern recognition is used. That is, the feature of the image is extracted, and in the pattern matching process,
The recognition result is obtained by comparing the characteristics of the standard pattern obtained in advance.

By the way, in general, in English text etc., 200 d
Pi (dot per inch; the number of dots (points) included in a one-inch length); in Japanese, a resolution of 300 dpi to 400 dpi is recommended. Therefore, for example, in the case of a low-resolution image of 100 dpi, if the target is English, the image interpolation processing is doubled, and if the target is Japanese, the image interpolation processing is tripled to quadrupled. The interpolation magnification is obtained by a method manually specified by the user, a method of estimating the resolution by any method, a method of calculating the interpolation magnification from the resolution information stored in the header portion of the image, or the like.

Further, a preferable resolution for character recognition can be calculated from the viewpoint of the character size of the character to be recognized. For example, in one character recognition device,
At 400 dpi, it is recommended that the number of pixels is 36 dots or more. In this case, it is also possible to calculate the character size of the low-resolution image by any method and calculate the interpolation magnification so as to be 36 dots or more.

Therefore, as described above, an image input at a low resolution is converted into a predetermined high-resolution image by image interpolation processing, and the converted image is recognized as a character. It can be performed.

(Embodiment 2) In this embodiment 2,
The difference from the first embodiment is that the input image to be recognized is a color image. That is, in the second embodiment, an example will be described in which, when a low-resolution color image is input, resolution conversion is performed by image interpolation, and an image after the resolution conversion is recognized.

FIG. 3 is a block diagram showing the configuration of the character recognition device according to the second embodiment of the present invention. This character recognition device 10a is basically the same as the configuration of FIG. 1 described above, and includes a character interpolation unit 11 for converting an input low-resolution color image to a predetermined resolution by image interpolation, and a character interpolation unit 11 And a character recognizing unit 12 for character recognizing the color image data whose resolution has been converted in the step (a).

The operation of the character recognition device 10a will be described with reference to the flowchart shown in FIG. First, the input image data to be recognized is a color image, that is, R image data.
(Red), G (green), B (blue) or Y
It is determined whether or not the color image signal is composed of a plurality of planes such as (yellow), M (magenta), C (cyan), and K (black) and has a low resolution (step S21).

If it is determined in step S21 that the image is a low-resolution color image, the character interpolation (resolution conversion) unit 11 converts the image into a resolution higher than a predetermined level, as in the first embodiment (step S22). , The converted image is recognized by the character recognition unit 12 (step S
23). Note that this character interpolation and character recognition use the same method as in the first embodiment.

Therefore, as described above, R,
An image obtained by inputting a color image composed of a plurality of planes such as G, B or Y, M, C, and K at a low resolution is converted into a predetermined high-resolution image by an image interpolation process. Since the character is recognized, more accurate character recognition can be performed.

(Embodiment 3) In this embodiment 3,
The difference from Embodiment 1 described above is that the input image to be recognized is a gray image. That is, in the third embodiment, an example will be described in which when a low-resolution gray image is input, resolution conversion is performed by image interpolation, and the image after the resolution conversion is recognized.

FIG. 5 is a block diagram showing the configuration of the character recognition device according to the third embodiment of the present invention. This character recognition device 10a is basically the same as the configuration of FIG. 1 described above, and includes a character interpolation unit 11 for converting an input low-resolution gray image to a predetermined resolution by image interpolation, and a character interpolation unit 11 And a character recognizing unit 12 for character recognizing the gray image data which has been subjected to the resolution conversion.

The operation of the character recognition device 10a will be described with reference to the flowchart shown in FIG. First, the input image data to be recognized is a gray image, for example, R,
It is determined whether or not the color image data composed of a plurality of planes such as G, B, Y, M, C, and K is a gray image obtained by converting a single plane into a gray image and has a low resolution ( Step S31).

If it is determined in step S31 that the image is a low-resolution gray image, the character interpolation (resolution conversion) unit 11 converts the image into a resolution higher than a predetermined level, as in the first embodiment (step S32). , The converted image is recognized by the character recognition unit 12 (step S
33). Note that this character interpolation and character recognition use the same method as in the first embodiment.

Therefore, as described above, an image obtained by inputting a gray image at a low resolution is converted into a predetermined high-resolution image by image interpolation processing, and the converted image is recognized as a character. Character recognition can be performed.

(Embodiment 4) In this embodiment 4,
An example will be described in which, when a low-resolution color image is input, resolution conversion is performed by image interpolation, the image after resolution conversion is converted to a gray image, and the gray image is recognized.

FIG. 7 is a block diagram showing a configuration of a character recognition device according to the fourth embodiment of the present invention. The character recognizing device 10b includes a character interpolating unit 11 for converting an input low-resolution color image into a predetermined resolution by image interpolation.
A character conversion unit 13 that performs gray conversion on the color image data whose resolution has been converted by the character interpolation unit 11, and a character recognition unit 12 that recognizes characters of the image data after gray conversion.

The operation of the character recognition device 10b will be described with reference to the flowchart shown in FIG. First, the input image data to be recognized is a color image, that is, R,
It is determined whether or not the color image signal is composed of a plurality of planes such as G, B or Y, M, C, K and has a low resolution (step S41).

If it is determined in step S41 that the image is a low-resolution color image, the character interpolation (resolution conversion) unit 11 converts the image into a resolution higher than a predetermined level, as in the first embodiment (step S42). The converted image is converted into a gray image by the gray conversion unit 13 (step S43). Then, the character conversion section 12 performs character recognition on the gray-converted image data (step S44).

In the gray conversion section 13, R, G, B
Alternatively, a color image signal composed of a plurality of planes such as Y, M, C, and K is converted into a gray image of one plane. This conversion uses, for example, a method of selecting only a G (green) plane from among a plurality of planes, or a method using a value of (R + G + B) / 3.

Therefore, as described above, R,
An image obtained by inputting a color image composed of a plurality of planes such as G, B or Y, M, C, and K at a low resolution is converted into a predetermined high-resolution image by an image interpolation process. Is gray-converted, and character recognition of the image data after the gray conversion is performed, so that more accurate character recognition can be performed.

(Embodiment 5) In this embodiment 5,
An example will be described in which, when a low-resolution color image is input, the color image is converted to a gray image, the image is converted in resolution by image interpolation, and the image after the resolution conversion is recognized.

FIG. 9 is a block diagram showing a configuration of a character recognition device according to the fifth embodiment of the present invention. The character recognition device 10c includes a gray conversion unit 13 for converting an input low-resolution color image into gray, a character interpolation unit 11 for converting image data after gray conversion to a predetermined resolution by image interpolation, and a character interpolation unit. And a character recognizing unit 12 for character recognizing the image data whose resolution has been converted at 11.

The operation of the character recognition device 10c will be described with reference to the flowchart shown in FIG. First, the input recognition target image data is a color image, that is,
It is determined whether or not the color image signal is composed of a plurality of planes such as R, G, B or Y, M, C, K and has a low resolution (step S51).

If it is determined in step S51 that the image is a low-resolution color image, gray conversion is performed by the gray conversion unit 13 (step S52). Then, the data converted to the gray image is converted to a resolution equal to or higher than a predetermined level by the character interpolation (resolution conversion) unit 11 (step S53), and the converted image is subjected to character recognition by the character recognition unit 12 for the image data. Is performed (step S54).

The gray conversion unit 13 converts a color image signal composed of a plurality of planes such as R, G, B or Y, M, C, K into one plane, as in the fourth embodiment. Convert to gray image. This conversion uses, for example, a method of selecting only a G (green) plane from among a plurality of planes, or a method using a value of (R + G + B) / 3.

Therefore, as described above, R,
An image obtained by inputting a color image composed of a plurality of planes such as G, B or Y, M, C, K at a low resolution is gray-converted, and the image after the gray conversion is subjected to interpolation processing to a predetermined high-resolution image. Since the image is converted into an image and character recognition of the image after the conversion is performed, more accurate character recognition can be performed.

(Embodiment 6) In this embodiment 6,
When a low-resolution image is input, resolution conversion is performed by image interpolation, and the image after the resolution conversion is binarized,
An example of recognizing an image after the binarization processing will be described.

FIG. 11 is a block diagram showing a configuration of a character recognition device according to the sixth embodiment of the present invention. The character recognizing device 10d includes a character interpolating unit 11 for converting an input low-resolution image into a predetermined resolution by image interpolation, and a binarizing unit for performing binarization processing on the image data whose resolution has been converted by the character interpolating unit 11. 14 and a character recognition unit 12 for character recognition of the image data after the binarization processing.

The operation of the character recognition device 10d will be described with reference to the flowchart shown in FIG. First, it is determined whether or not the resolution of the input recognition target image data is low (step S61). This step S61
If it is determined that the resolution of the input image data is low, the character interpolation unit 11 converts the resolution to a predetermined resolution (step S62) and inputs the resolution to the binarization unit 14.

The binarizing unit 14 converts the image data converted to a predetermined resolution into white or black pixels (pixels including characters and background pixels) (step S63). This conversion method is to process the entire image by comparing the size with a preset threshold value, or to perform a smoothing process,
Processing such as edge enhancement processing is performed, and adaptive binarization processing is performed on each pixel. Further, the binarizing section 14 directly converts the color image and outputs the R.D. G, B, Y, M, C, K
May be converted into pixels including characters and the like and background pixels.

Therefore, according to the sixth embodiment,
When a low-resolution image is input, resolution conversion is performed by image interpolation, the image after the resolution conversion is binarized into white / black pixels, and the image after the binarization processing is recognized. Accurate character recognition is realized.

(Embodiment 7) In this embodiment 7,
When a low-resolution color image is input, resolution conversion is performed by image interpolation, the converted image is converted to a gray image, the gray image is binarized, and the binarized image is converted to a gray image. An example of recognition will be described.

FIG. 13 is a block diagram showing a configuration of a character recognition device according to the seventh embodiment of the present invention. The character recognizing device 10e includes a character interpolating unit 1 that converts an input low-resolution color image into a predetermined resolution by image interpolation.
1, a gray conversion unit 13 that converts the image data whose resolution has been converted by the character interpolation unit 11 into a gray image, a binarization unit 14 that binarizes the gray image, and an image data that has been binarized. A character recognition unit 12 for character recognition.

The operation of the character recognition device 10e will be described with reference to the flowchart shown in FIG. First, it is determined whether or not the input recognition target image data is a low-resolution color image (step S71). This step S
If it is determined in step 71 that the image data is a low-resolution color image, the image data is converted into a predetermined resolution by the character interpolation unit 11 (step S72) and input to the gray conversion unit 13.

Thereafter, the image is converted into a gray image by the gray conversion unit 13 (step S 73) and input to the binarization unit 14. The binarization unit 14 converts the image data converted to a predetermined resolution into white or black pixels (pixels including characters and background pixels), as in the sixth embodiment (step S74). Then, character recognition is performed from the image data after the binarization processing (step S75).

Therefore, according to the seventh embodiment,
Low resolution color images (R, G, B or Y, M,
(Color image composed of multiple planes such as C and K)
Is input, the resolution is converted by image interpolation, the image after the resolution conversion is converted to a gray image, the gray image is binarized, and the image after the binarization processing is recognized with higher accuracy. Character recognition is realized.

(Eighth Embodiment) In the eighth embodiment,
When a low-resolution color image is input, the color image is converted to a gray image, the resolution is converted by image interpolation, the image after the resolution conversion is binarized, and the image after the binarization processing is recognized. An example will be described.

FIG. 15 is a block diagram showing a configuration of a character recognition device according to the eighth embodiment of the present invention. The character recognition device 10f includes a gray conversion unit 13 for converting an input low-resolution color image into a gray image, a character interpolation unit 11 for converting the input low-resolution color image to a predetermined resolution by image interpolation, and a resolution conversion performed by the character interpolation unit 11. A binarizing unit 14 for binarizing the image data, and a character recognizing unit 12 for character recognizing the binarized image data.

The operation of the character recognition device 10f will be described with reference to the flowchart shown in FIG. First, it is determined whether or not the input recognition target image data is a low-resolution color image (step S81). This step S
If it is determined in 81 that the image data is a low-resolution color image, the image data is converted into a gray image by the gray conversion unit 13 (step S82). Further, the gray image is converted into a predetermined resolution by the character interpolation unit 11 (step S83) and input to the binarization unit 14.

Thereafter, the binarizing section 14 performs the processing in the sixth embodiment.
Similarly to the above, the image data converted to the predetermined resolution is converted into white or black pixels (pixels including characters and the like and background pixels) (step S84). Then, character recognition is performed from the image data after the binarization processing (step S85).

Therefore, according to the eighth embodiment,
Low resolution color images (R, G, B or Y, M,
(Color image composed of multiple planes such as C and K)
Is input, the image is converted to a gray image, the resolution is converted by image interpolation, the image after the resolution conversion is binarized, and a higher-precision character that recognizes the image after the binarization is processed. Recognition is realized.

(Embodiment 9) In this embodiment 9,
An example will be described in which, when a low-resolution gray image is input, the resolution is converted by image interpolation, the gray image after the resolution conversion is binarized, and the image after the binarization processing is recognized.

FIG. 17 is a block diagram showing a configuration of a character recognition device according to the ninth embodiment of the present invention. The character recognizing device 10g includes a character interpolating unit 1 that converts an input low-resolution gray image into a predetermined resolution by image interpolation.
1, a binarizing unit 14 for binarizing the image data whose resolution has been converted by the character interpolating unit 11, and a character recognizing unit 12 for character recognizing the image data after the binarizing process.

The operation of the character recognition device 10g will be described with reference to the flowchart shown in FIG. First, it is determined whether or not the input recognition target image data is a low-resolution gray image (step S91). This step S
If it is determined in step 91 that the image data is a low-resolution gray image, the gray image is converted to a predetermined resolution by the character interpolation unit 11 (step S92) and input to the binarization unit 14. The binarization unit 14 converts the resolution-converted image data into white or black pixels (pixels including characters and background pixels) (step S93). And 2
Character recognition is performed from the image data after the value processing (step S94).

Therefore, according to the ninth embodiment,
When a low-resolution gray image is input, resolution conversion is performed by image interpolation, the image after the resolution conversion is binarized, and higher-precision character recognition that recognizes the image after the binarization processing is realized. I do.

(Embodiment 10) In the tenth embodiment, when a low-resolution color image is input, the resolution is converted by image interpolation, the color image after the resolution conversion is binarized, and An example of recognizing an image after the conversion processing will be described.

FIG. 19 is a block diagram showing a configuration of a character recognition device according to the tenth embodiment of the present invention. The character recognition device 10g differs from the above-described ninth embodiment in that the image data input to the character interpolation unit 11 is low-resolution color image data.

The operation of the character recognition device 10g will be described with reference to the flowchart shown in FIG. First, it is determined whether or not the input recognition target image data is a low-resolution color image (step S101). If it is determined in step S101 that the image data is a low-resolution color image, the color image
To a predetermined resolution (step S102),
The data is input to the binarization unit 14. The binarization unit 14 converts the resolution-converted image data into white or black pixels (pixels including characters and background pixels) (step S103). Then, character recognition is performed from the image data after the binarization processing (step S104).

Therefore, according to the tenth embodiment, when a low-resolution color image is input, resolution conversion is performed by image interpolation, and the resolution-converted image is converted to two.
Character recognition with higher precision is realized than by performing the binarization process and recognizing the image after the binarization process.

The above-described character recognition method can be programmed, recorded on a computer-readable recording medium, and executed on a computer. Also, a part of the character recognition method can be provided on a network and realized through a communication line.

That is, the character recognition method described in the present embodiment is realized by executing a prepared program on a computer (CPU 20) such as a personal computer or a workstation, as shown in FIG. This program uses keyboard 2
5, the memory 21, the hard disk 2
4, FD (floppy disk) 27, CD-ROM2
6. The program is recorded on a computer-readable recording medium such as an MO and a DVD, and is executed by being read from the recording medium by the computer (CPU 20). If necessary, the character recognition processing data can be communicated from the communication device 22 to an external device.

As shown in FIG. 22, this program can be distributed to devices 31 to 33 such as personal computers via the recording medium via a network such as the Internet 30.

[0085]

As described above, according to the first aspect of the present invention,
According to the method, the low-resolution image information is converted into a predetermined resolution by image interpolation by the image interpolation means, and the image information having the converted resolution is recognized. Since an image having a stable resolution sufficient for recognition is input to the recognition means, it is possible to avoid a reduction in the recognition rate due to the input of the low-resolution image, and to use a character size preferable when the user performs character recognition. Since it is no longer necessary to confirm the existence of
It is possible to provide a character recognition device that does not require the operation of confirming the appropriateness / inappropriateness of an image to be input, which is performed when a low-resolution input device is used.

According to the second aspect of the present invention, the low-resolution color image information is converted into a predetermined resolution by image interpolation by the image interpolation means, and the color image information having the converted resolution is recognized. Even when a low-resolution color image is input, an image having a stable resolution sufficient for image recognition is input to the recognition unit, so that a reduction in the recognition rate due to the input of the low-resolution image can be avoided. In addition to this, it is not necessary for the user to perform troublesome confirmation of whether or not the character size is preferable when performing character recognition, so that it is possible to provide a character recognition device that does not require a conventional image input confirmation operation.

According to the third aspect of the present invention, the low-resolution color image information is converted into a gray image by the gray conversion means, and the image information converted into the gray image is subjected to image interpolation by the image interpolation means. The color image information is converted to a predetermined resolution, the color image information converted to the predetermined resolution is binarized by the binarizing means, and the binarized image is recognized by the white / black pixels, whereby the low-resolution color image is obtained. Is input to the recognizing unit even when the image is input, the reduction of the recognition rate due to the input of the low-resolution image can be avoided, and the user can perform character recognition. In such a case, it is not necessary to perform a troublesome check as to whether or not the character size is preferable. Therefore, it is possible to provide a character recognition device that does not require a conventional image input check operation.

According to a fourth aspect of the present invention, low-resolution gray image information is converted into a predetermined resolution by image interpolation by an image interpolation means, and the gray image information whose resolution has been converted is converted into binary image information. By binarizing the image with the binarizing unit and recognizing the binarized image with the white / black pixels, an image of a level sufficient for image recognition is input to the recognizing unit even when a low-resolution gray image is input. Therefore, it is possible to avoid a decrease in the recognition rate due to the input of the low-resolution image, and it is not necessary to perform a troublesome check as to whether or not the user has a preferable character size when inputting an image to be subjected to character recognition. Therefore, it is possible to provide a character recognition device that does not require a conventional image input confirmation operation.

According to a fifth aspect of the present invention, the low-resolution color image information is converted into a predetermined resolution by image interpolation by the image interpolation means, and the converted color image information is converted into binary data. The binarizing unit binarizes the image and recognizes the binarized image with the white / black pixels, so that even if a low-resolution image is input, an image of a level sufficient for image recognition is transmitted to the recognizing unit. Since the input is performed, it is possible to avoid a decrease in the recognition rate due to the input of the low-resolution image, and it is not necessary for the user to perform a troublesome check as to whether or not the character size is preferable when performing the character recognition. It is possible to provide a character recognition device that does not require an image input confirmation operation.

According to the sixth aspect of the present invention, the low-resolution image information is converted to a predetermined resolution by image interpolation in the image interpolation step, and the low-resolution image information is recognized, whereby the low-resolution image information is recognized. Even when a high-resolution image is input, an image having a resolution sufficient for image recognition is input to the recognition process, so that a reduction in the recognition rate due to the input of the low-resolution image can be avoided, and the user can use the image. Since it is not necessary to perform a troublesome check of a preferable character size when inputting an image for character recognition, the appropriateness / inappropriateness of an image to be input has been conventionally performed when a low-resolution input device is used. And a character recognition method that does not require a confirmation operation.

According to the seventh aspect of the present invention, low-resolution color image information is converted into a predetermined resolution by image interpolation in the image interpolation step, and the color image information having the converted resolution is recognized. Even when a low-resolution color image is input, an image having a resolution sufficient for image recognition is input to the recognition process, so that a reduction in the recognition rate due to the input of the low-resolution image can be avoided. In addition, since the user does not need to perform troublesome confirmation of whether or not the character size is preferable when performing character recognition, it is possible to provide a character recognition method that does not require a conventional image input confirmation operation.

According to the eighth aspect of the present invention, the low-resolution color image information is converted into a gray image by the gray conversion step, and the image information converted into the gray image is subjected to the image interpolation by the image interpolation step. The color image information is converted to a predetermined resolution, and the color image information whose resolution has been converted is binarized in a binarization step, and the binarized image is recognized by the white / black pixels, thereby obtaining a low-resolution color image. Is input to the recognition step even when the image is input, a reduction in the recognition rate due to the input of the low-resolution image can be avoided, and the user can perform character recognition. In this case, it is not necessary to perform a troublesome check on whether the character size is preferable or not, so that it is possible to provide a character recognition method that does not require a conventional image input check operation.

According to the ninth aspect of the present invention, low-resolution gray image information is converted into a predetermined resolution by image interpolation in an image interpolation step, and the gray image information whose resolution is converted is converted into binary image information. By recognizing the binarized image in the white / black pixels in the binarization process, an image of a level sufficient for image recognition is input to the recognition process even when a low-resolution gray image is input. Therefore, it is possible to avoid a decrease in the recognition rate due to the input of the low-resolution image, and it is not necessary for the user to perform a troublesome check as to whether or not the character size is preferable when performing the character recognition. A character recognition method that does not require an input confirmation operation can be provided.

According to a tenth aspect of the present invention, low-resolution color image information is converted to a predetermined resolution by image interpolation in an image interpolation step, and the converted color image information is converted to binary data. Binarization in the whitening process
By recognizing an image binarized by black pixels, an image of a sufficient level for image recognition is input to the recognition process even when a low-resolution image is input. It is possible to avoid a reduction in the recognition rate due to the input, and it is not necessary for the user to perform a troublesome check of a preferable character size when performing character recognition. A recognition method can be provided.

According to an eleventh aspect of the present invention, a program for causing a computer to execute the character recognition method according to any one of the sixth to tenth aspects is recorded.
A computer-readable recording medium capable of realizing the character recognition operation according to any one of claims 6 to 10 by a computer is obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a character recognition device according to a first embodiment of the present invention.

FIG. 2 is a flowchart illustrating a character recognition procedure according to the first embodiment of the present invention;

FIG. 3 is a block diagram illustrating a configuration of a character recognition device according to a second embodiment of the present invention;

FIG. 4 is a flowchart illustrating a character recognition procedure according to the second embodiment of the present invention;

FIG. 5 is a block diagram illustrating a configuration of a character recognition device according to a third embodiment of the present invention;

FIG. 6 is a flowchart illustrating a character recognition procedure according to the third embodiment of the present invention.

FIG. 7 is a block diagram illustrating a configuration of a character recognition device according to a fourth embodiment of the present invention.

FIG. 8 is a flowchart illustrating a character recognition procedure according to the fourth embodiment of the present invention.

FIG. 9 is a block diagram showing a configuration of a character recognition device according to a fifth embodiment of the present invention.

FIG. 10 is a flowchart showing a character recognition procedure according to the fifth embodiment of the present invention.

FIG. 11 is a block diagram showing a configuration of a character recognition device according to a sixth embodiment of the present invention.

FIG. 12 is a flowchart showing a character recognition procedure according to the sixth embodiment of the present invention.

FIG. 13 is a block diagram illustrating a configuration of a character recognition device according to a seventh embodiment of the present invention.

FIG. 14 is a flowchart showing a character recognition procedure according to the seventh embodiment of the present invention.

FIG. 15 is a block diagram illustrating a configuration of a character recognition device according to an eighth embodiment of the present invention.

FIG. 16 is a flowchart showing a character recognition procedure according to the eighth embodiment of the present invention.

FIG. 17 is a block diagram illustrating a configuration of a character recognition device according to a ninth embodiment of the present invention.

FIG. 18 is a flowchart showing a character recognition procedure according to the ninth embodiment of the present invention.

FIG. 19 is a block diagram showing a configuration of a character recognition device according to a tenth embodiment of the present invention.

FIG. 20 is a flowchart showing a character recognition procedure according to the tenth embodiment of the present invention.

FIG. 21 is a block diagram illustrating an example of a computer system that implements the character recognition method according to the present invention by software.

FIG. 22 is a block diagram showing an example of a system for realizing a character recognition method according to the present invention on a network.

[Explanation of symbols]

 10a to 10g Character recognition device 11 Character interpolation (resolution conversion) unit 12 Character recognition unit 13 Gray conversion unit 14 Binarization unit

Claims (11)

[Claims] 1. A character recognition device for inputting image information and performing character recognition of the image information, comprising: an image interpolation means for converting input low-resolution image information to a predetermined resolution by image interpolation; A character recognition device comprising: recognition means for recognizing image information converted by the means. 2. A character recognition device for inputting image information and performing character recognition on the image information, comprising: an image interpolation unit for converting the input low-resolution color image information into a predetermined resolution by image interpolation; A character recognition device, comprising: recognition means for recognizing color image information converted by interpolation means. 3. A character recognition apparatus for inputting image information and performing character recognition of the image information, comprising: a gray conversion unit for converting the input low-resolution color image information into a gray image; Image interpolation means for converting the obtained gray image information to a predetermined resolution by image interpolation, and binarizing the image information converted by the image interpolation means.
A character recognition device, comprising: a value conversion unit; and a recognition unit that recognizes image information binarized by the binarization unit. 4. A character recognition apparatus for inputting image information and performing character recognition of the image information, comprising: an image interpolation unit for converting the input low-resolution gray image information into a predetermined resolution by image interpolation; Characters comprising: a binarization unit for binarizing gray image information converted by an image interpolation unit; and a recognition unit for recognizing image information binarized by the binarization unit. Recognition device. 5. A character recognition apparatus for inputting image information and performing character recognition of the image information, comprising: an image interpolation unit for converting the input low-resolution color image information into a predetermined resolution by image interpolation; Characters comprising: a binarizing unit for binarizing color image information converted by an image interpolating unit; and a recognizing unit for recognizing the image information binarized by the binarizing unit. Recognition device. 6. A character recognition method for inputting image information and performing character recognition of the image information, comprising: an image interpolation step of converting input low-resolution image information to a predetermined resolution by image interpolation; A recognition step of recognizing the image information converted in the step. 7. A character recognition method for inputting image information and performing character recognition of the image information, comprising: an image interpolation step of converting the input low-resolution color image information to a predetermined resolution by image interpolation; A recognition step of recognizing the color image information converted in the interpolation step. 8. A character recognition method for inputting image information and performing character recognition of the image information, comprising: a gray conversion step of converting the input low-resolution color image information into a gray image; An image interpolation step of converting the obtained gray image information into a predetermined resolution by image interpolation, and binarizing the image information converted in the image interpolation step.
A character recognition method, comprising: a binarization step; and a recognition step of recognizing image information binarized in the binarization step. 9. A character recognition method for inputting image information and performing character recognition of the image information, comprising: an image interpolation step of converting input low-resolution gray image information to a predetermined resolution by image interpolation; A character recognition comprising: a binarization step of binarizing gray image information converted in an image interpolation step; and a recognition step of recognizing image information binarized in the binarization step. Method. 10. A character recognition method for inputting image information and performing character recognition of the image information, comprising: an image interpolation step of converting input low-resolution color image information to a predetermined resolution by image interpolation; Character recognition comprising: a binarization step of binarizing color image information converted in an image interpolation step; and a recognition step of recognizing image information binarized in the binarization step. Method. 11. A computer-readable recording medium on which a program for causing a computer to execute the character recognition method according to claim 6 is recorded.