JP3742358B2 - Optical character recognition apparatus and program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an optical character recognition apparatus that performs character recognition processing using a full-color image form reading definition body.
[0002]
The present invention also relates to a program for a form definition tool for creating a full color image form reading definition body.
[0004]
[Prior art]
In an OCR (Optical Character Recognition) system that performs character recognition processing using a form reading definition body, conventionally, recognition processing is performed by determining a dropout color for each form. In other words, conventionally, in the design of an OCR form, a form is created using a single color dropout color such as red, green or blue, and the form other than the dropout color is read from the image read by the scanner. The character recognition was performed on the image image of the color.
[0005]
In recent years, in this type of form OCR system, as a scanner, a character recognition personal computer, etc. have been speeded up and storage devices have become cheaper, one line is blue and one line is orange. Alternatively, character recognition has been performed for forms created by changing colors in units of multiple lines.
[0006]
By performing character recognition using a full-color image of the form to be read, it becomes possible to set a dropout color for each field, thereby significantly improving the degree of freedom in designing the form.
[0007]
However, the character recognition of the full-color image of the form to be read specifies the dropout color required at the time of field recognition when defining the field of the form reading definition body, and the plane corresponding to the dropout color is read from the full-color image at the time of reading. Was done by choosing. Here, for example, red dropout, green dropout, and blue dropout correspond to the red plane, green plane, and blue plane of the full-color image, respectively.
[0008]
In this type of full-color image form reading mechanism, when realizing the display function of the screen in each reading field, the storage function of the output image, etc. Since the data is very large, the processing speed may be slow, and there may be practical problems due to memory capacity limitations. In addition, when a binarized image using a plane at the time of reading is used for a screen image, etc., the field frame, wording, etc. will drop out, causing a problem that the entire image cannot be grasped correctly. It was.
[0009]
Therefore, there is a need for means to optimize the display image and output image of the full-color image to the extent that they do not affect the processing speed, storage capacity, etc., but in the past it was simple and satisfied with operability and economy. There was no effective optimization means.
[0010]
[Problems to be solved by the invention]
As described above, there has been a problem in the prior art that there is no effective means for optimizing the display image and output image of the read full-color image to the extent that they do not affect the processing speed, storage capacity, etc., respectively. .
[0011]
The present invention has been made in view of the above circumstances, and it is possible to easily and efficiently optimize the display image, output image, etc. of the screen when reading a full-color image with a simple configuration. It is an object of the present invention to provide an optical character recognition apparatus and program capable of realizing a display function of a screen and a storage function of an output image when reading a full color image excellent in performance.
[0012]
[Means for Solving the Problems]
The present invention provides not only a plane corresponding to a field recognition dropout color but also a plane for displaying a field image on a screen, a plane for generating an output image, and the like when defining a form reading for full color image reading. It is characterized in that it has a mechanism that can arbitrarily specify a color, and an optimal screen and output image can be acquired using a plane selected according to the specification at the time of reading.
[0013]
That is, the present invention provides a scanner that acquires a full color image of a read form, and a plane corresponding to a dropout color of the read field for each read field with respect to the full color image of the read form acquired by the scanner. A form definition tool for creating a form reading definition body in which a display image generation plane and an output image generation plane are set by a graphic user interface; and the form reading definition body created by the form definition tool. Using the definition storage device to be stored and the form reading definition stored in the definition storage device, the character reading according to the setting of the dropout color from the full color image is performed in units of the reading field, A pre-production for generating the display image from the full-color image. A control device that generates a binary image screen according to a screen and generates a binary output image according to an output image generation plane from the full-color image, and a display that displays the binary image screen generated by the control device. An optical character recognition device comprising: a device; and an output image storage device for storing a binary output image generated by the control device.
[0014]
Further, the present invention provides a form definition tool for creating a full-color image form reading definition body in units of reading fields, wherein the form reading definition body includes a plane corresponding to a dropout color at the time of character reading and a screen display. A graphic user interface for arbitrarily setting at least one of the plane and the output image plane.
[0015]
Further, the present invention makes it possible to set a screen display plane in units of reading fields in a form reading definition body applied to reading a full-color image form, and using the set screen display plane, It is characterized by generating and displaying a binary image screen from a full-color image.
[0016]
Further, the present invention makes it possible to set an output image plane in units of reading fields in a form reading definition body applied to reading a full-color image form, and using the set output image plane, Binary image data is generated from a full color image and stored.
[0017]
By providing a screen display image generation function and an output image generation function using the above-described form reading definition body of the present invention, the screen display image and the output image at the time of reading a full-color image form can be simply configured. Therefore, it is possible to easily and efficiently optimize a desired binary image, thereby realizing a screen display function and an output image storage function when reading a full color image having excellent economy and operability. Further, by applying a form definition tool having a graphic user interface (GUI) for arbitrarily setting a plane for screen display and an output image plane, the format definition element of the form reading definition body is a tree structure in the GUI. The format definition elements can be defined in the form reading definition body for each reading field by a simple operation such as dragging and dropping with a pointing device such as a mouse. The screen of the value image and the output image can be acquired.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
[0019]
FIG. 1 is a block diagram showing the configuration of a form OCR system according to the present invention, which includes a scanner 10, a control apparatus 11, a form definition tool 121 and an OCR control program 122 provided in the control apparatus 11, a memory 13, and a form definition. It comprises a body storage device 14, an output image storage device (HDD) 15, a display device 16, and the like.
[0020]
The scanner 10 is a full-color scanner, and sends a full-color image of a read form that has been read by scanning (reading scanning) the read form to the control device 11.
[0021]
The control device 11 controls the entire form OCR system and has a control function for executing the processes of the form definition tool 121 and the OCR control program 122.
[0022]
The form definition tool 121 is a software module having a GUI function for creating a full-color image form reading definition body (FC). In the embodiment of the present invention, according to the processing procedure shown in FIG. The creation processing function of the form reading definition body (FC) including the plane setting in the plane selection table 141 shown in FIG.
[0023]
The OCR control program 122 has a processing function for controlling the entire OCR system including character recognition processing using a form reading definition body (FC). In the embodiment of the present invention, the processing procedure shown in FIG. Referring to the plane selection table 141 shown in FIG. 4, the screen generation and display function of the binary image and the generation and storage of the binary output image using each plane set in the plane selection table 141 are performed. Realize the function.
[0024]
The memory 13 is a storage device having a RAM configuration used for various processes of the control device 11. In the embodiment of the present invention, a full-color image of a form to be read read by the scanner 10 is developed.
[0025]
The form definition storage device 14 stores the form reading definition (FC) created by the form definition tool 121 under the control of the control device 11. In the embodiment of the present invention, a form reading definition body (FC) including the plane selection table 141 shown in FIG. 4 is stored.
[0026]
The output image storage device 15 is a large-capacity storage device that stores an output image of a read result under the control of the control device 11, and is configured by, for example, a hard disk device (HDD). In the embodiment of the present invention, a binary output image generated from a full color image stored in the RAM 13 using the plane selection table 141 is stored in units of reading fields.
[0027]
The display device 16 is a color monitor that displays the GUI screen of the form definition tool 121, the screen of the reading result, and the like under the control of the control device 11. In the embodiment of the present invention, the plane selection table 141 is displayed. A binary image screen generated from a full color image stored in the RAM 13 is displayed in units of reading fields.
[0028]
FIG. 2 is a flowchart showing a procedure for creating a form reading definition body (FC) according to an embodiment of the present invention, which is executed when the control device 11 activates the form definition tool 121.
[0029]
FIG. 3 is a flowchart showing a screen and output image generation and output processing procedure during character recognition processing according to an embodiment of the present invention, and uses display image planes defined in the plane selection table 141. The binary image screen is generated and displayed on the display device 16. Also, a binary output image is generated using the output image plane defined in the plane selection table 141 and stored in the output image storage device 15.
[0030]
FIG. 4 shows an example of the configuration of the plane selection table 141 provided in the form reading definition body (FC) according to the embodiment of the present invention. Here, the form reading definition body (FC) is shown in FIG. A plane corresponding to the dropout color, a display image plane, and an output image plane are defined (set) for each of the defined reading fields (fields 1, 2, 3,...), And a full color image. Used as a selection plane above.
[0031]
In FIG. 4, the plane setting area corresponding to the dropout color is indicated as “read”, the plane setting area for the display image is indicated as “display”, and the plane setting area for the output image is indicated as “output”. The red plane is indicated by “R”, the green plane is indicated by “G”, and the blue plane is indicated by “B”. Here, in order to simplify the description, an example using only three planes of “R”, “G”, and “B” is shown, but not limited to this, “R”, “G”, and “B” are combined. It is also possible to use more color planes.
[0032]
Here, the operation in one embodiment of the present invention will be described with reference to the respective drawings.
[0033]
First, a processing procedure when creating and acquiring a full color image form reading definition body (FC) will be described with reference to the flowchart shown in FIG. The full color image form reading definition body creation process is executed when the control device 11 activates the form definition tool 121.
[0034]
In the process of creating a form reading definition body (FC) by the form definition tool 121, first, a full-color read form to be newly read is scanned by the scanner 10, and the full-color image acquired from the scanner 10 is scanned by the control device 11. Is developed on the memory 13 and the creation of a form reading definition body (FC) in the form to be read is started (steps S1 and S2 in FIG. 2).
[0035]
In the creation process of the form reading definition body (FC), first, the full color image developed on the memory 13 is displayed on the display device 16 and the reading field is set on the GUI screen of the form definition tool 121. To do. Subsequently, for each reading field on the GUI screen, the plane corresponding to the dropout color at the time of character reading, the plane for the display image, and the plane for the output image are set, so that the internal memory of the control device 11 is covered. A full-color image form reading definition body (FC) of the reading form is created (steps S3 and S4 in FIG. 2).
[0036]
In the GUI when creating this form reading definition body (FC), the format definition elements of the form reading definition body (FC) are displayed in a tree structure, and the user can display the format definition elements displayed in this tree structure by a mouse or the like. By simply dragging and dropping using a pointing device, copy and input values as needed, each reading field of the entire form and each plane for each reading field can be easily and quickly And can be set. The form reading definition body (FC) created on the internal memory of the control device 11 is stored in the form definition body storage device 14 (step S5 in FIG. 2). At this time, as shown in FIG. 4, the plane corresponding to the dropout color at the time of character reading specified on the GUI, the plane for the display image, and the plane for the output image are planes of the form reading definition body (FC). It is set in the selection table 141.
[0037]
Next, screen display after reading characters of a form to be read and output image saving processing using the plane selection table 141 of the form reading definition body (FC) stored in the form definition body storage device 14 will be described with reference to FIG. This will be described with reference to the flowchart shown in FIG.
[0038]
In the reading process of the form to be read, the control device 11 first loads the form reading definition body (FC) stored in the form definition body storage device 14 into the internal memory in the control device 11 in accordance with the OCR control program 122. (FIG. 3, step S11).
[0039]
Each time a full-color image of a form to be read is acquired from the scanner 10, a character reading process is executed for each full-color image of the entire form with reference to the plane selection table 141 of the form reading definition body (FC). (FIG. 3, step S12). At this time, the control device 11 reads a character by selecting a plane corresponding to the dropout color from the full color image in units of the reading field of the read form according to the setting of the plane selection table 141 of the form reading definition body (FC). I do. For example, in the example of FIG. 4, in the field 1 at the time of reading, the red plane “R” is set as the dropout color, the field 2 is the blue plane “B”, the field 3 is the green plane “G”, and the field 4 is Since the red plane “R” is set as the dropout color, the character is read by dropping out the set color. Based on the read output, the control device 11 performs character recognition.
[0040]
Next, for the read field subjected to the character recognition processing, the plane selection table 141 of the form reading definition body (FC) shown in FIG. 4 is referred to, and the display image generation plane set in the plane selection table 141 is used. Then, a binary image screen is generated from the full-color image and displayed on the display device 16 (step S13 in FIG. 3). For example, in the example of FIG. 4, the display plane 1 is set with the blue plane “B” as the display color, the field 2 is the red plane “R”, the field 3 is the red plane “R”, and the field 4 is blue. Since the plane “B” is set as the display color, the plane of that color is selected and the full color image is displayed on the display device 16 as a binary display image. As a result, since the display of the binary image and the recognition result are displayed simultaneously, the display item and the character recognition result can be viewed in correspondence with each other. It can also be used to correct recognized characters.
[0041]
Further, using the output image generation plane set in the plane selection table 141, a binary output image is generated from the full color image and stored in the output image storage device 15 (step S14 in FIG. 3). For example, in the example of FIG. 4, in the output field 1, the blue plane “B” is set as the output color, the field 2 is the green plane “G”, the field 3 is the red plane “R”, and the field 4 is green. Since the plane “G” is set as the output color, the full color image can be binarized by selecting the plane of that color and stored in the output image storage device 15.
[0042]
In this way, a plane of arbitrary color is set in the plane selection table 141 of the form reading definition body (FC), and a binary screen and an output image can be generated from the full color image using the set plane. Therefore, it is possible to obtain a desired optimal binary image screen display and an output image easily and easily.
[0043]
In the above-described embodiment, the plane corresponding to the dropout color, the display image plane, and the output image plane are set in the plane selection table 141 of the form reading definition body (FC). By specifying the type of image processing (for example, a sharp binarized image) and the like so that the image processing can be executed at the time of reading, a further optimal screen image and output image can be obtained.
[0044]
【The invention's effect】
As described above in detail, according to the present invention, it is possible to easily and efficiently optimize the display image, output image, and the like of the screen when reading a full-color image, each with a simple configuration. In addition, it is possible to realize a full-color image-compatible form OCR having an excellent operability and a screen display function when reading a full-color image and a storage function of an output image.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a form OCR system according to an embodiment of the present invention.
FIG. 2 is a flowchart showing a procedure for creating a form reading definition body in the embodiment.
FIG. 3 is a flowchart showing a screen and output image generation and output processing procedure during character recognition processing in the embodiment.
FIG. 4 is a diagram showing a configuration example of a plane selection table provided in a form reading definition body in an embodiment of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... Scanner 11 ... Control apparatus 12 ... Form definition tool 13 ... Memory 14 ... Form definition body memory | storage device 15 ... Output image memory | storage part apparatus (HDD)
16 ... Display device 141 ... Plane selection table FC ... Form definition body

Claims (2)

A scanner that acquires a full-color image of the form to be read;
For the full color image of the read form acquired by the scanner, for each reading field, a plane corresponding to the dropout color of the reading field, a display image generation plane, and an output image generation plane A definition storage device for storing a form reading definition body provided with a plane selection table defined respectively;
Using the plane selection table stored in the definition storage device, character reading is performed from the full color image according to the setting of the dropout color in units of the reading field, and the display image is generated from the full color image. A control device for generating a binary image screen according to a plane for generating a binary output image according to a plane for generating an output image from the full color image;
A display device for displaying a screen of a binary image generated by the control device;
An optical character recognition device comprising: an output image storage device for storing a binary output image generated by the control device. In an optical character reader that recognizes a color image of a form using a form reading definition body,
The optical character reader includes, in the form reading definition body, each reading field on the read form, a plane corresponding to a dropout color at the time of reading a character corresponding to the reading field, and a display image generation A function to set a plane selection table that defines each plane and output image generation plane,
The optical character reader performs character recognition according to the setting of the dropout color from the color image of the form for each reading field, using the plane selection table set in the form reading definition body. Generating a binary image according to the display image generation plane from the color image of the form, and generating a binary image according to the output image generation plane from the color image of the form;
A program to realize

Images