JP2004227355A - Scanner system, sheet for setting threshold, and binary image forming method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To effectively form a binary image for OCR processing. <P>SOLUTION: A scanner system (101) of this invention comprises an image acquiring device (10) that acquires an image picture representing a document from a document and a binary image forming device (20) that forms a binary image corresponding to the image picture acquired by the image acquiring device. The binary image forming device comprises a threshold setting part (22) that detects a luminance distribution of a standard color from a standard imaged picture to which the standard color indicating a specified luminance to acquire a binary image is given, and sets as a threshold a luminance value with the highest intensity from the detected luminance distribution, and a binarization processing part (25) that carries out binarization processing, using the threshold to the imaged picture acquired by the image acquiring device and forms a binary image corresponding to the imaged picture. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a scanner system that forms a binary image for determining characters, symbols, and the like represented in an image image obtained from a document.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, in performing an OCR (Optical Character Reader) process of determining an entry character from an image image representing a document such as a form, an image portion in which the entry character is described and a frame line described in a dropout color In order to identify an image portion that is not necessary for forming a binary image as described above, a binary image is formed based on the luminance of the image image. In general, the dropout color is set so as to show a higher luminance distribution than the ideal luminance distribution of the entered character. A technique relating to the formation of such a binary image has been proposed in, for example, Japanese Patent Application Laid-Open No. H11-216, which will be described later.
[0003]
In the technology of the literature, a method of binarizing an image image obtained by a non-contact scanner has been proposed.When setting a binarization threshold, a character to be discriminated is a form similar to a form handled during operation. An unfilled form with no is used. Then, a luminance distribution is detected for each predetermined region in the image image of the blank form, and the minimum luminance value of each luminance distribution is set as a binarization threshold value of the region.
[0004]
[Patent Document 1]
JP-A-2002-34739
[Problems to be solved by the invention]
However, in the above-described technology, since a form corresponding to operation is used as an unfilled form, it is necessary to set a new threshold value each time the layout or dropout color of the form handled during operation is different. Therefore, there is a problem that cost and labor are required.
[0006]
Further, when the density of the entered character is partially light, the luminance value of the portion is higher than that of the other character portions. However, since the binarization threshold of the above technique is set to exclude a pattern showing a luminance higher than the ideal luminance distribution of the entered characters, such as a pattern of a dropout color, the binarization processing is not performed. The high-luminance portion described above is easily removed together with the dropout color pattern, and as a result, the character pattern of the binary image may be rubbed or missing.
[0007]
The present invention has been made in view of the above-described problem, and can improve the efficiency of the binarization process.Also, it is possible to set a threshold value for more appropriately displaying a character pattern in a binary image. It is an object of the present invention to provide a scanner system capable of performing the above.
[0008]
[Means for Solving the Problems]
<Configuration 1>
A scanner system according to the present invention includes an image acquisition device that acquires an image representing a document from a document, and a binary image forming device that forms a binary image corresponding to the image acquired by the image acquisition device. A scanner system, wherein the binary image forming apparatus detects a luminance distribution of the reference color from a reference image image to which a reference color indicating a predetermined luminance for obtaining a binary image is given, and detects the detected luminance. A threshold value setting unit that sets, as a threshold, a luminance value at which the intensity is maximum from the distribution, and performs a binarization process on the image image acquired by the image acquisition device using the threshold value; A binarization processing section for forming a value image.
[0009]
The reference image image may be an image in which the reference color is assigned to each of a plurality of regions that have been divided in advance. In that case, the threshold value setting unit sets the threshold value for each of the regions of the reference image image, and the binarization processing unit performs the binarization process based on the corresponding threshold value for each region. It is desirable to apply.
[0010]
The scanner system according to the present invention further includes a luminance distribution higher than the luminance distribution of the reference color, and a drop-out color that is not required for forming the binary image, from the reference image image given together with the reference color. A dropout determination unit that detects the luminance distribution of the dropout color and compares the detected luminance distribution with the threshold to determine whether or not the dropout color can be excluded can be provided.
[0011]
The reference image image may be an image in which the reference color and the dropout color are assigned to each of a plurality of pre-divided regions. In that case, the threshold value setting unit sets the threshold value for each of the regions of the reference image image, and the dropout determination unit determines, for each region, the determination regarding the dropout color based on the corresponding threshold value. It is desirable to carry out.
[0012]
As the image acquisition device of the scanner system according to the present invention, a non-contact scanner device that optically acquires the image image without contacting the document can be applied.
[0013]
<Configuration 2>
The threshold setting sheet according to the present invention is provided with a reference color indicating a predetermined luminance for obtaining a binary image corresponding to an image representing a document.
[0014]
<Configuration 3>
A binary image forming method according to the present invention is a method of obtaining an image image representing a document from a document and forming a binary image corresponding to the obtained image image. The luminance distribution of the reference color is detected from the reference image image to which the reference color indicating the luminance is given, and a luminance value having the maximum intensity from the detected luminance distribution is set as a threshold, and the threshold is set for the image image of the document. This is a method of performing a binarization process using a threshold to form a binary image corresponding to the image image.
[0015]
BEST MODE FOR CARRYING OUT THE INVENTION
An embodiment of the present invention will be described using a specific example.
<Specific example 1>
FIG. 1 is a block diagram showing a configuration of a specific example 1 of the scanner system according to the present invention.
As illustrated in FIG. 1, a scanner system 101 according to a specific example 1 includes a non-contact scanner device 10 serving as an image acquisition device that acquires an image image of a document from a document such as a form, and an image acquired by the non-contact scanner device. And a binary image forming apparatus 20 for forming a binary image corresponding to the converted image image. As the binary image forming apparatus 20, a general-purpose personal computer can be used.
The scanner system 101 further includes a display device 30 such as a CRT display that displays a screen for an operator who operates the binary image forming apparatus 20, and an input operation to the binary image forming apparatus 20. An input device 40 such as a keyboard is provided.
[0016]
FIG. 2 shows a configuration example of the non-contact scanner device 10 in the scanner system 101. The non-contact scanner device 10 is provided with a mat unit 11 for arranging a document from which an image image is to be obtained, and an image obtaining unit 12 for optically reading a document without contacting the document on the mat unit. .
[0017]
In a document handled by the non-contact scanner device 10, a frame line, a ruled line, and the like that are unnecessary in the OCR process are described in a predetermined dropout color indicating a luminance distribution higher than a luminance distribution of a character to be written. . After reading the document by the image acquisition unit 12, the non-contact scanner device 10 performs a filtering process on the image image by a filter unit (not shown) of the non-contact scanner device, and removes most of the dropout colors.
[0018]
As shown in FIG. 1, the binary image forming apparatus 20 includes an image memory 21 for storing an image acquired by the non-contact scanner 10 and an image for a threshold setting sheet (described later) stored in the image memory. A binarization threshold setting unit 22 that sets a binarization threshold based on the binarization threshold; a layout memory 23 that previously holds layout information used when setting the binarization threshold; and a binarization threshold setting unit 22 A binarization threshold value memory 24 that stores the binarization threshold value that has been obtained, and a binarization process using a binarization threshold value of the binarization threshold value memory 24 for a document for which an entry character is to be determined. The image processing apparatus includes a binarization processing unit 25 to be formed, and a control unit 26 that controls the operation of each component of the binary image forming apparatus 20.
[0019]
The binarization threshold setting unit 22 sets a binarization threshold in adjustment processing prior to formation of a binary image. In the setting, a threshold setting sheet corresponding to the reference image shown in FIG. 3A is used. The threshold setting sheet is provided with a reference color that is a color that indicates a critical luminance value that does not cause character rubbing in the binary image. As the luminance of the reference color, a value obtained from an experiment or experience can be used as long as the value satisfies the above-described condition. For example, as a reference color, a color having a luminance distribution of about 100 to 200 with a peak near 150 of 256 gradations under fluorescent lamp illumination can be adopted. Further, as the color of the reference color, any achromatic color or chromatic color can be used as long as the color corresponds to the color specified for the entered character.
[0020]
As shown in FIG. 3A, an achromatic reference color is assigned to five preset positions on the threshold setting sheet of this specific example. In the illustrated example, the reference color is applied to each of two places above and below the form, and one place near the center of the form. This arrangement is an example set when the non-contact scanner device 10 reads the document, taking into account that the influence of illumination is different at each location of the form. The arrangement is not limited to the illustrated example, and the arrangement can be appropriately set. .
[0021]
FIG. 3B shows layout information stored in the layout memory 23 in advance. The layout information defines which part of the threshold setting sheet is to be detected for the luminance distribution of each part where the reference color is applied.
FIG. 3C shows an image image in which the reference image image shown in FIG. 3A is associated with the layout information shown in FIG. 3B.
[0022]
In this specific example, the binarization threshold setting unit 22 sets a binarization threshold for each range divided by the boundary shown in FIG. Specifically, as shown in FIG. 4, the luminance distribution of the reference color is detected for each region such as “region 1: (xs1, ys1)-(xe1, ye1)” in the layout information of the layout memory 23. Then, the luminance value having the maximum intensity in the detected luminance distribution is stored in the binarization threshold memory 24 for each region as a binarization threshold such as “Threshold 1” shown in FIG.
[0023]
During operation, the binarization processing unit 25 performs binarization processing on the image of the completed form for each of the above-described regions. In the binarization processing, for example, in the histogram shown in FIG. 4, a pattern indicating a luminance value lower than 150, which is the binarization threshold, is determined to be a pattern of a character to be subjected to the OCR processing. It is determined that the pattern indicating a high luminance value is a pattern like a frame to be excluded in the binarization processing.
[0024]
The operation of the scanner system 101 of the first specific example will be described with reference to the flowchart of the adjustment process shown in FIG. 5 and the flowchart at the time of operation shown in FIG.
In the adjustment processing according to the flowchart of FIG. 5, when the non-contact scanner device 10 reads a document from the threshold setting sheet, the binary image forming device 20 stores the reference image, which is the image, in the image memory 21. (Step S1). The white peak of the reference image stored in the image memory 21 is normalized by the well-known shutter control by the non-contact scanner device 10.
[0025]
The binarization threshold setting unit 22 generates a reference color histogram as shown in FIG. 4 for each of the above-described regions obtained from the layout information of the layout memory 23 for the reference image image. Then, the luminance value corresponding to the peak in the histogram is stored in the binarization threshold memory 24 for each region as a binarization threshold (step S2). With this, the pre-operation adjustment processing ends.
[0026]
In the operation of the scanner system 101 shown in FIG. 6, when the non-contact scanner device 10 reads a completed form, that is, a form to be subjected to OCR processing, the binary image forming apparatus 20 stores the image in the image memory 21. (Step S3). In this image image as well, the white peak is normalized as in the case of the above-described reference image image.
[0027]
The binarization processing unit 25 uses the binarization threshold value corresponding to each of the image images stored in the image memory 21 for each range divided as shown in FIG. Become Furthermore, a binary image is formed by integrating the binarized images for each range, and this is stored in the image memory 21 (step S4). Thereafter, the binary image stored in the image memory 21 is subjected to character determination by an OCR processing unit (not shown).
[0028]
According to the scanner system 101 of the specific example 1, since the shared binarization threshold is set for various forms using the threshold setting sheet, it is not necessary to set the threshold every time the form is changed. , And a binary image for the OCR process can be efficiently formed. Further, since the binarization threshold is set based on the reference color of the character, the binarization threshold hardly exists in the luminance distribution of the entered character, and as a result, it is possible to prevent the character from being rubbed or missing. it can.
[0029]
<Specific example 2>
FIG. 7 is a block diagram showing a configuration of a specific example 2 of the scanner system according to the present invention.
The scanner system 102 of the specific example 2 has the same configuration as that of the scanner system 101 of the specific example 1 described above, and further includes whether or not dropout colors can be eliminated by the binarization threshold set by the binarization threshold setting unit 22. A drop-out availability determination unit 27 that determines whether the drop-out is possible.
[0030]
As shown in FIG. 8A, the threshold setting sheet of the specific example 2 is provided with a limit color of a dropout color adjacent to a reference color of a character arranged at each position. The limit color of the dropout color is a color in which the luminance value corresponding to the peak of the luminance intensity indicates a value higher than the binarization threshold and a color indicating a luminance distribution that can be removed by the binarization threshold. As the luminance distribution of the limit color, a value obtained from an experiment or experience can be used as long as the value satisfies the above-described condition. For example, as the limit color, a color having a luminance distribution of about 160 to 255 having a peak near 210 of 256 gradations under fluorescent lamp illumination can be adopted. Further, the limit color is not limited to a chromatic color, and an achromatic color having the same luminance as the chromatic color can be used.
[0031]
As shown in FIG. 8B, the layout information of the layout memory 23 of the specific example 2 detects the luminance of the dropout color shown in FIG. Area to be defined.
FIG. 8C shows an image image in which the reference image image shown in FIG. 8A is associated with the layout information shown in FIG.
[0032]
In the adjustment processing prior to the operation, the dropout availability determination unit 27 detects the luminance distribution of the dropout color for each region from the reference image image, and determines whether the dropout color can be removed by the binarization threshold. Is determined. For example, as shown in FIG. 9, when a dropout color having a luminance distribution having a peak at the value of (A) is detected, the dropout color is determined by a binarization threshold obtained from the reference color of the character. It is determined that it can be properly excluded. Further, for example, when a dropout color having a peak at the value of (B) is detected, it is determined that the dropout color cannot be properly excluded because a binarization threshold exists in the luminance distribution. Is determined.
[0033]
The operation in the adjustment process of the scanner system 102 of the specific example 2 will be described with reference to the flowchart shown in FIG.
The procedure from steps S10 to S11 for setting the binarization threshold after the non-contact scanner device 10 reads the document on the threshold setting sheet is the procedure from steps S1 to S2 in the scanner system 101 of the specific example 1 described above. Therefore, the description is omitted.
[0034]
After the binarization threshold is set, the dropout availability determination unit 27 makes a determination on the exclusion of the above-described dropout color for each region of the image described with reference to FIG. 8C (step S12). ). In this determination, when it is determined that the dropout color cannot be properly removed, for example, a message indicating that there is a problem in the installation environment of the non-contact scanner device 10 and noise due to illumination is occurring is displayed on the display device 30. Can be displayed.
[0035]
If it is determined in the determination that the dropout color can be properly eliminated, the scanner system 102 can be used in the same manner as the operation procedure described in the first embodiment without changing the installation environment of the non-contact scanner device 10. Can be operated.
[0036]
According to the scanner system 102 of the specific example 2, the same effect as that of the scanner system 101 of the specific example 1 can be obtained. Further, since the determination as to whether or not the dropout color can be excluded is performed, the non- It is possible to confirm whether the installation environment of the contact scanner device 10 is good or not. As a result, during operation, it is possible to more appropriately eliminate the dropout color, and it is possible to form a highly accurate binary image.
[0037]
In each of the specific examples described above, an example is shown in which a single reference color is given to the threshold setting sheet, but a color having a plurality of PCS (Print Contrast Scale) values is also used as the reference color. be able to. In that case, for example, by a linear interpolation process, it is possible to calculate a luminance that does not cause indirect character rubbing.
[0038]
Further, in each of the specific examples described above, the example of the scanner system using the non-contact scanner device has been described. However, the present invention can be applied to a scanner system that reads a document using a contact-type scanner device.
In each of the specific examples described above, the threshold setting sheet is divided into a plurality of regions in advance. However, a threshold setting sheet in which the reference color is assigned to one place without being divided in this way can be used.
[0039]
【The invention's effect】
ADVANTAGE OF THE INVENTION According to the scanner system which concerns on this invention, since the threshold value of the common sheet | seat for a binarization is set to various forms using the sheet | seat for a threshold value setting, it is not necessary to set a threshold value for every form of a document. , A binary image for the OCR process can be efficiently formed. Moreover, since the threshold value for binarization is set based on the reference color, the threshold value is unlikely to exist in the luminance distribution of the entered character, and as a result, it is possible to prevent the character from being rubbed or missing.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a specific example 1 of a scanner system according to the present invention.
FIG. 2 is an explanatory diagram illustrating a configuration of a non-contact scanner device.
FIG. 3 is an explanatory diagram illustrating a threshold setting sheet according to a specific example 1.
FIG. 4 is an explanatory diagram for explaining setting of a binarization threshold.
FIG. 5 is a flowchart illustrating a procedure of an adjustment process of the scanner system according to the first embodiment.
FIG. 6 is a flowchart illustrating a procedure when the scanner system according to the first embodiment is operated.
FIG. 7 is a block diagram showing a configuration of a specific example 2 of the scanner system according to the present invention.
FIG. 8 is an explanatory diagram illustrating a threshold setting sheet according to a second specific example.
FIG. 9 is an explanatory diagram for explaining availability determination regarding removal of a dropout color. FIG. 10 is a flowchart illustrating a procedure of an adjustment process of the scanner system according to the second embodiment.
[Explanation of symbols]
101, 102 Scanner system 10 Non-contact scanner device 11 Mat unit 12 Image acquisition unit 20 Binary image forming device 21 Image memory 22 Binarization threshold setting unit 23 Layout memory 24 Binarization threshold memory 25 Binarization processing unit 26 Control Unit 27 Dropout availability determination unit 30 Display device 40 Input device

Claims (7)

An image acquisition device that acquires an image image representing the document from a document, and a scanner system including a binary image forming device that forms a binary image corresponding to the image image acquired by the image acquisition device,
The binary image forming apparatus includes:
A luminance distribution of the reference color is detected from a reference image image to which a reference color indicating a predetermined luminance for obtaining a binary image is added, and a luminance value having the maximum intensity from the detected luminance distribution is set as a threshold value. A threshold setting unit;
A scanner that performs a binarization process on the image image acquired by the image acquisition device using the threshold value and forms a binary image corresponding to the image image. system. In the reference image image, the reference color is given to each of a plurality of regions that have been divided in advance,
The threshold value setting unit sets the threshold value for each of the regions of the reference image image,
The scanner system according to claim 1, wherein the binarization processing unit performs the binarization processing for each of the regions based on the corresponding threshold value. Furthermore, the luminance distribution of the drop-out color is obtained from a reference image image that shows a luminance distribution higher than the luminance distribution of the reference color and is used together with the reference color, and a drop-out color that is not used for forming the binary image is provided. 2. The scanner system according to claim 1, further comprising: a dropout determination unit that detects and compares the detected luminance distribution with the threshold to determine whether or not the dropout color is excluded. In the reference image image, the reference color and the dropout color are given to each of a plurality of regions that are previously divided,
The threshold value setting unit sets the threshold value for each of the regions of the reference image image,
The scanner system according to claim 3, wherein the dropout determination unit performs the determination regarding the dropout color based on the corresponding threshold value for each of the areas. The scanner system according to claim 1, wherein the image acquisition device is a non-contact scanner device that optically acquires the image image without touching the document. A threshold setting sheet, which is provided with a reference color indicating a predetermined luminance for obtaining a binary image corresponding to an image image representing a document. A binary image forming method for obtaining an image image representing the document from a document and forming a binary image corresponding to the obtained image image,
A luminance distribution of the reference color is detected from a reference image image to which a reference color indicating a predetermined luminance for obtaining a binary image is added, and a luminance value having the maximum intensity from the detected luminance distribution is set as a threshold. And performing a binarization process on the written image image using the threshold value to form a binary image corresponding to the image image.

Images