JP2012073749A - Image formation device and image formation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To smoothly generate a high-quality searchable file.SOLUTION: An image formation device includes: an OCR analysis unit 153 for extracting character information from a target image of predetermined image quality; a panel 13 for displaying output images of the character information and the target image; and an image processing unit 152 for generating searchable PDF data of the target image where the character information can be detected, on the basis of the character information and the target image. The panel 13 displays an output image of character information extracted from the target image of changed image quality and an output image of the target image. The image processing unit 152 generates searchable PDF on the basis of the output image of the character information extracted from the target image of changed image quality and the output image of the target image.

Description

  The present invention relates to an image forming apparatus and an image forming program for generating file data such as PDF, and more particularly to an image forming apparatus and an image forming program for generating a searchable file with invisible text data.

In recent years, when a document is converted into file data such as PDF (Portable Document Format) or XPS (XML Paper Specification), a searchable PDF or searchable PDF in which a desired character string can be searched by adding character information included in the document. The use of XPS and the like (hereinafter collectively referred to as searchable files) is increasing.
As shown in FIG. 9, in order to extract character information from a manuscript, a searchable file uses OCR (Optical Character Reader) technology that extracts text data from raster data obtained by binarizing image data (JPEG data or the like). Used. For this reason, when generating a searchable file, appropriate resolution conversion is required to achieve accurate pattern matching of character objects.
That is, the resolution corresponds to the character recognition level, and the character recognition accuracy can be adjusted by increasing or decreasing the resolution (see, for example, Patent Document 1).

By the way, as shown in FIG. 10, the searchable file hides a text layer made up of text data extracted by OCR processing, and the text layer and the image data layer are formed in a hierarchy.
For this reason, the user cannot judge at a glance whether or not the character string embedded in the searchable file is correct. If the user does not actually perform a character string search through a viewer or the like, the accuracy will be improved. There was a problem that could not be determined.
In order to solve such a problem, a character recognition method including a step of displaying a character string extracted by OCR processing and a step of correcting an erroneously extracted character has been proposed (for example, see Patent Document 2). .

JP 2006-172055 A JP 2004-318760 A

However, according to the character recognition method described in Patent Document 2, since the character itself is corrected manually by the user, there is a possibility that a secondary error may occur.
Further, according to the character recognition device described in Patent Document 1, such a mistake cannot occur, but various keywords and scores for each keyword must be set in advance in order to determine OCR recognition accuracy. It was complicated. Furthermore, since there is no means for setting the reading resolution to an arbitrary value and confirming the result in advance, it is inflexible and inconvenient for the user.

  The present invention has been made in view of the circumstances as described above, and in generating a searchable file, the user can arbitrarily set the character recognition level and can check the accuracy of character extraction. An image forming apparatus and an image forming program capable of smoothly generating a high-quality searchable file by repeating the above process are provided.

  In order to achieve the above object, an image forming apparatus of the present invention includes character information extraction means for extracting character information from a target image having a predetermined image quality, display means for displaying the character information and an output image of the target image, File data generating means for generating file data of the target image capable of detecting the character information based on the character information and the target image; and character accuracy changing means for changing the image quality of the target image in accordance with an input operation. The display means displays the character information extracted from the target image with the changed image quality and the output image of the target image, and the file data generation means extracts the character information extracted from the target image with the changed image quality. In addition, the file data is generated based on the target image.

  Further, the image forming program of the present invention includes a computer of an image forming apparatus, character information extracting means for extracting character information from a target image having a predetermined image quality, display means for displaying the character information and an output image of the target image, Functions as file data generating means for generating file data of the target image capable of detecting the character information based on the character information and the target image, and character accuracy changing means for changing the image quality of the target image in accordance with an input operation. And displaying the character information extracted from the target image having the changed image quality and the output image of the target image on the display unit, and extracting the image data from the target image having the changed image quality on the file data generating unit. The file data is generated based on the character information and the target image.

  According to the image forming apparatus and the image forming program of the present invention, a high-quality searchable file can be generated smoothly.

1 is a block diagram illustrating a configuration of an MFP (image forming apparatus) according to an embodiment of the present invention. It is explanatory drawing for demonstrating the image process at the time of image data input. It is explanatory drawing for demonstrating the image process in the process of extracting text data and displaying a preview. It is explanatory drawing for demonstrating the image process at the time of the reset of a character recognition level. FIG. 5 is a diagram showing a preview image of a document and a preview image of a character string. It is the figure which showed the adjustment bar of a character recognition level. It is explanatory drawing for demonstrating the image process at the time of searchable PDF production | generation. It is a flowchart for demonstrating the procedure of the image processing which concerns on one Embodiment of this invention. It is the 1st explanatory view showing the composition of a searchable file. It is the 2nd explanatory view showing the composition of a searchable file.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of an MFP (image forming apparatus) according to an embodiment of the present invention.
As illustrated in FIG. 1, the MFP 10 according to the present exemplary embodiment includes a communication interface 11, a display interface 12, a panel 13, a scanner device 14, and a control unit 15.

  The communication interface 11 is a network interface for communicating necessary data and signals with a personal computer 20 (hereinafter referred to as a PC 20) via a network such as a LAN (Local Area Network). For example, image data (for example, image data such as JPEG or MMR) of a document output from the PC 20 based on execution of a utility or a Web browser in the PC 20 is received.

  The display interface 12 is an input / output interface that transmits and receives various data to and from the panel 13. For example, a process for delivering a searchable PDF creation request output from the panel 13 to the control unit 15, a process for delivering the generated preview image to the panel 13, and the like are performed.

  The panel 13 includes a touch panel unit that combines a touch panel and a color LCD (Liquid Crystal Display). The panel 13 displays various screens and operation buttons for inputting various operation commands by a user's touch operation. Configure the operation panel. For example, the panel 13 displays a preview image of a document or a preview image of a character string extracted from the document, or displays an operation button or other user interface for making a searchable PDF creation request (FIGS. 5 and 6). reference).

The scanner device 14 is a scanner device such as an image sensor mounted as a scanner unit, and acquires image data (image data such as JPEG and MMR) by reading a document on which images and characters are drawn.
The scanner device 14 outputs the acquired document image data to the control unit 15 (a data transmission / reception unit 151 described later).

The control unit 15 includes a microcomputer such as a CPU, a ROM that stores control programs and data, a RAM that functions as a work memory, and an I / O port. The control unit 15 executes the stored control program to control each unit constituting the MFP 10.
For example, the control unit 15 performs image processing such as converting image data received via the communication interface 11 or image data input via the scanner device 14 into raster data.
Note that raster data obtained by image processing is normally used for print processing as an output image (output image). Specifically, a toner image based on the raster data is formed on the photosensitive drum, and this toner image is temporarily transferred to the transfer belt. Next, the toner image on the transfer belt is transferred to a sheet, and the toner image is fixed on the sheet by a transfer roller, whereby a printing process is performed (photosensitive drum, transfer drum, and transfer roller are not shown).

Here, the control unit of the MFP according to the embodiment of the present invention will be described in detail.
As shown in FIG. 1, the control unit 15 includes a data transmission / reception unit 151, an image processing unit 152, an OCR analysis unit 153, and a font rasterizer 154 as functional blocks.
The data transmission / reception unit 151 exchanges necessary data or signals between the control unit 15 and other components.
For example, image data input from the communication interface 11 or the scanner device 14 is output to the image processing unit 152, or raster data input from the image processing unit 152 is output to the communication interface 11 or the display interface 12.

The image processing unit 152 generates an output image (raster data) by decomposing the input image data into pixels based on a predetermined output image quality (resolution) and converting them into binary data (output image generation means).
The image processing unit 152 can also generate an output image after changing the output image quality, such as changing the resolution and image size, or performing rotation and color conversion, as necessary.
The image processing unit 152 outputs the generated output image to the OCR processing unit 153.
Further, the image processing unit 152 converts the output image into PDF data, and generates a PDF file (searchable PDF) by combining the text data extracted by the OCR analysis unit 153 and the PDF data (file data generation unit). ).

The OCR analysis unit 153 analyzes the output image and extracts text data (character information extraction unit). Specifically, the OCR analysis unit 153 takes out pixels constituting a character from the raster data and attaches attribute information such as a corresponding character code, size, and position.
The font rasterizer 154 generates a preview image of a character string by converting text data into raster data. A preview image of the generated character string is sent to the PC 20 or the panel 13 and displayed (see FIG. 5).

The PC 20 is electrically connected to the MFP 10 via a network such as a LAN (Local Area Network).
A PC browser and other PC utilities having a bitmap image preview function are installed in the PC 20, and various information can be exchanged with the MFP 20 by the PC utility.

  Next, image processing according to an embodiment of the present invention will be described in detail. Specifically, i) image processing at the time of image data input, ii) image processing in the step of extracting and previewing text data, iii) image processing at the time of resetting the character recognition level, and iv) searchable PDF Each of the image processing at the time of generation will be described.

i) Image Processing at the Time of Image Data Input FIG. 2 is an explanatory diagram for explaining image processing at the time of image data input in one embodiment of the present invention.
Specifically, as shown in FIG. 2, image processing performed between the input of image data and the output of an output image to the OCR analysis unit 153 will be described.

As shown in FIG. 2, first, image data of an original is input via the PC 20 or the scanner device 14.
For example, the PC 20 outputs a searchable PDF creation request signal and document image data to the communication interface 11 via a Web browser or other utility that can exchange information with the MFP 10. The communication interface 11 outputs the input signal and data to the data transmission / reception unit 151.
On the other hand, when generating a searchable PDF using document image data read by the scanner device 14, the panel 13 outputs a search request signal for the searchable PDF to the data transmission / reception unit 151, and the scanner device 14 outputs the image data of the document. Is output to the data transmission / reception unit 151.

The data transmission / reception unit 151 outputs the input searchable PDF creation request signal and the image data of the document to the image processing unit 152.
The image processing unit 152 binarizes the input document image data based on a predetermined output image quality (resolution), and acquires an output image composed of raster data.
The image processing unit 152 generates a document preview image based on the acquired output image, holds the generated preview image, and outputs the output image to the OCR analysis unit 153.

ii) Image Processing in Step of Extracting Text Data and Displaying Preview FIG. 3 is an explanatory diagram for explaining image processing performed from extraction of text data from an output image to preview display.

Here, first, the OCR analysis unit 153 extracts character information from the output image, converts it into text data such as plain text and rich text, and outputs the text data to the image processing unit 152.
Next, the image processing unit 152 converts the input text data into raster data using the font rasterizer 154 and acquires it as a preview image of a character string. The image processing unit 152 holds the input text data.
Further, the image processing unit 152 outputs the stored document preview image and character string preview image to the data transmission / reception unit 151. At this time, the image processing unit 152 can process the preview image of the document and the preview image of the character string side by side so that they can be compared with each other and output the processed preview image to the data transmission / reception unit 151.
Next, the image processing unit 152 outputs such a preview image and the text data input from the OCR analysis unit 153 to the data transmission / reception unit 151.

The data transmission / reception unit 151 outputs the input preview image and text data to the communication interface 11 or the display interface 12.
The communication interface 11 transmits the input preview image and text data to the PC 20, and the display interface 12 outputs the input preview image and text data to the panel 13.
The PC 20 or the panel 13 to which the preview image is input displays the document preview image and the character string preview image (display means). In addition to the preview image of the character string, other character information such as the number of recognized characters, language, and processing time can be displayed.
Therefore, the user can refer to the two displayed preview images while comparing them, and can easily confirm whether the characters extracted by the OCR analysis unit 153 are correct.

iii) Image processing at the time of resetting the character recognition level FIG. 4 is an explanatory diagram for explaining image processing at the time of resetting the character recognition level.
The resetting of the character recognition level is arbitrarily performed at the user's discretion, for example, when the preview image is confirmed and it is determined that many character errors are unacceptable.
As shown in FIGS. 5 and 6, the user refers to a monitor screen (not shown) connected to the PC 20 and a preview screen displayed on the display area of the panel 13 through a user interface such as operation buttons and an adjustment bar. Reset the character recognition level.

Specifically, the user selects a “recognition level readjustment” button displayed in the display area shown in FIG. When the “recognition level readjustment” button is selected, the character recognition level adjustment bar (operation unit) shown in FIG. 6 is displayed (operation unit display means). According to this adjustment bar, the ratio between the character recognition accuracy (extraction accuracy) and the performance (processing speed) can be adjusted by a button operation or a slide operation (character accuracy changing means). For example, in FIG. 6, the user can select the left of the level selection button and set the accuracy high, but in this case, the performance is set low. If the right of the level selection button is selected, the performance can be set high, but the accuracy is set low.
When the character recognition level is reset and the “OK” button is selected, the reset character recognition level setting information is output.

  The character recognition level setting information is a parameter relating to output image quality. Specifically, according to the resetting of the character recognition level, the image processing unit 152 changes the output image quality such as the resolution when converting the image data into the output image (raster data) (character accuracy changing means). In other words, when the recognition accuracy as the character recognition level is set higher than the original recognition accuracy, a high-resolution output image is generated accordingly, and when the recognition accuracy is set lower than the original recognition accuracy, Accordingly, a low-resolution output image is generated.

The character recognition level setting information may be determined according to the type of document. For example, by using the pre-scan function of the scanner device 14 or the like, the type of the original such as a black and white character original, a color character original, or a mixed image is determined, and the output image quality (resolution) is set according to the determined original type. It can also be changed.
In addition to the resolution, other parameters relating to image quality such as image size and color conversion can also be used as setting information for the character recognition level.

The specific image processing flow is as follows.
First, when the character recognition level is reset through the operation of the mouse attached to the PC 20, the PC 20 transmits the reset character recognition level setting information to the communication interface 11, and the communication interface 11. Outputs this setting information to the data transmission / reception unit 151.
On the other hand, when the character recognition level is reset by a key operation on the panel 13, the panel 13 outputs the reset character recognition level setting information to the display interface 12, and the display interface 12 displays the setting information. Is output to the data transmission / reception unit 151.
Next, the data transmission / reception unit 151 outputs the character recognition level setting information related to the resetting to the image processing unit 152.

The image processing unit 152 converts the image data of the document into an output image based on the character recognition level setting information related to the resetting, and outputs the output image to the OCR analysis unit 153.
For example, in the resetting of the character recognition level, when a resolution higher than the original resolution is set, the original image data is converted into raster data based on the newly set resolution, and this is converted to the OCR analysis unit 153. Output.
Similar to the image processing in ii) described above, the OCR analysis unit 153 extracts text data from the output image, and the font rasterizer 154 converts this into raster data to generate a character string preview image. The processing unit 153 outputs the text data to the PC 20 or the panel 13.

iv) Image processing when generating a searchable PDF FIG. 7 is an explanatory diagram for explaining image processing when generating a searchable PDF in one embodiment of the present invention.
As shown in FIG. 7, the PC 20 or the panel 13 outputs a searchable PDF file creation execution signal in response to a predetermined operation. This operation is executed by selecting an “execute” key displayed on the preview screen shown in FIG.
When a file creation execution signal is output via the PC 20, the communication interface 11 outputs this to the data transmission / reception unit 151. On the other hand, when a file creation execution signal is output via the panel 13, the display interface 12 outputs this to the data transmission / reception unit 151.
The data transmission / reception unit 151 outputs the input file creation execution signal to the image processing unit 152.

  The image processing unit 152 that has received the execution signal for file creation converts the output image of the document held at that time into PDF data, and also stores the text data (that is, the character set immediately before) at that time. The searchable PDF is generated by combining the text data (extracted by the OCR analysis unit 153 based on the recognition level) and the PDF data. When generating a searchable PDF, the text layer forming the text data is hidden, and the text layer and the PDF data are combined in a hierarchical manner.

By the way, the searchable PDF can be generated by the user via the PC 20 or the panel 13 through the process of correcting the text data.
In this case, the user deletes, adds, or changes the corresponding character by a keyboard operation of the PC 20 or a key operation of the panel 13. Then, in response to a predetermined operation, the PC 20 or the panel 13 outputs the corrected text data and a file creation execution signal to the data transmission / reception unit 151.
The data transmission / reception unit 151 having received the execution signal for file creation takes out the corrected text data input together with the execution signal. Then, the original output image held at that time is converted into PDF data, and the retrieved corrected text data and the PDF data are combined to generate a searchable PDF.

Next, an image processing method according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 8 is a flowchart showing an image processing method according to an embodiment of the present invention.
As shown in FIG. 8, first, the MFP 10 receives a searchable PDF file creation request and document image data (S1). Specifically, a searchable PDF file creation request signal and document image data are output in response to a user's predetermined operation on the PC 20, and the data transmission / reception unit 151 receives the request signal via the communication interface 11. Alternatively, a searchable PDF file creation request signal output in response to a user's predetermined operation on the panel 13 is input to the data transmission / reception unit 151 via the display interface 12 and the image data of the original read by the scanner device 14 Is input by the data transmission / reception unit 151.
The data transmission / reception unit 151 outputs the input file creation request signal and the original image data to the image processing unit 152.

The image processing unit 152 that has received the file creation request signal adjusts the image data of the document input together with the request signal to a predetermined output image quality (predetermined resolution) to create an output image (raster data). Further, based on the output image, a preview image of the document is generated (S2).
The image processing unit 152 outputs the generated output image of the document to the OCR analysis unit 153.
The OCR analysis unit 153 extracts text data from the output image of the document (S3) and outputs it to the image processing unit 152.
Further, the image processing unit 152 converts the text data into raster data using the font rasterizer 154, and generates a preview image of the character string (S4).
Next, the image processing unit 152 sends the generated document preview image, character string preview image, and text data to the PC 20 or the panel 13 (S5). The image processing unit 152 holds text data.

Next, the PC 20 or the panel 13 displays the received (input) preview image of the original and the preview image of the character string (S6). At this time, other character information such as the number of recognized characters, language, and processing time may be displayed together with the preview image.
Thus, the user can check the accuracy of character extraction by the OCR analysis unit 153 by looking at the preview image.
When the character recognition level is reset (S7: YES), the image processing unit 152 acquires the reset character recognition level setting information, and the original image data based on the setting information. Is converted into raster data to generate an output image (S8). The generated output image of the document is output to the OCR analysis unit 153, and the processes after step S3 are performed again.

On the other hand, if the character recognition level has not been reset (S7: NO), it is determined whether the text data has been modified (S9). That is, it is determined whether or not the text data has been corrected manually by the user.
When the text data is corrected (S9: YES), the data transmitting / receiving unit 151 receives (inputs) the corrected text data from the PC 20 or the panel 13 and outputs it to the image processing unit 152 (S10).
Then, the image processing unit 152 combines the PDF data of the output image and the corrected text data to generate a searchable PDF (S11).
If the text data has not been corrected (S9: NO), the image processing unit 152 combines the PDF data of the output image and the stored text data to generate a searchable PDF (S11).

As described above, according to the MFP 10 according to the embodiment of the present invention, when generating a searchable file, the user can change the character recognition level (output image quality) to perform OCR processing, and character extraction in this OCR processing. The accuracy can be confirmed.
By enabling such recursive OCR processing, it is possible to generate a searchable file with high accuracy.

In addition, an adjustment bar that can set the ratio between the character information extraction accuracy and the character information extraction speed in the OCR analysis unit 153 is displayed, and the output image quality is changed based on the ratio determined by the setting of the adjustment bar. OCR processing can be performed.
For this reason, it is possible to generate a searchable PDF having a certain accuracy while maintaining a balance with the speed performance.

  In addition, since it is also possible to manually correct characters, for example, the entire document is corrected by resetting the character recognition level (output image quality), and the user makes pinpoint corrections for fine corrections. Flexible response is possible.

The image forming apparatus of the present invention has been described with reference to the preferred embodiments. However, the image forming apparatus according to the present invention is not limited to the above-described embodiments, and various modifications can be made within the scope of the present invention. It goes without saying that implementation is possible.
For example, in the MFP 10 according to the embodiment of the present invention, an example of generating a searchable PDF has been described, but a searchable file may be generated using other document data that is useful in combination with text data. For example, a searchable XPS composed of a combination of XPS data and text data can be generated.
In the above-described embodiment, the searchable PDF is generated by the MFP 10. However, the PC 20 may have all the necessary configurations, and the searchable PDF may be generated by the PC 20 alone.

  The present invention can be suitably used for an image forming apparatus such as a copying machine, a printer, or a multifunction machine.

10 MFP (image forming apparatus)
13 Panel 14 Scanner Device 15 Control Unit 152 Image Processing Unit 153 OCR Analysis Unit 154 Font Rasterizer 20 PC (Host Computer)

Claims (5)

Character information extraction means for extracting character information from a target image of a predetermined image quality;
Display means for displaying the character information and an output image of the target image;
File data generating means for generating file data of the target image capable of detecting the character information based on the character information and the target image;
Character accuracy changing means for changing the image quality of the target image in response to an input operation,
The display means displays the character information extracted from the target image of the changed image quality and the output image of the target image;
The image forming apparatus, wherein the file data generation unit generates the file data based on the character information extracted from the target image having a changed image quality and the target image. Output image generating means for binarizing the target image based on a predetermined image quality or a changed image quality;
The character information extraction means is
The image forming apparatus according to claim 1, wherein character information is extracted from the binarized target image. An operation unit display unit for displaying an operation unit for setting a ratio between the character information extraction accuracy and the character information extraction speed in the character information extraction unit;
The image forming apparatus according to claim 1, wherein the character accuracy changing unit changes the image quality of the target image based on the ratio determined by the setting of the operation unit.   The image forming apparatus according to claim 1, wherein the character accuracy changing unit changes an image quality of the target image according to a type of the target image. The computer of the image forming device
Character information extraction means for extracting character information from a target image of a predetermined image quality;
Display means for displaying the character information and an output image of the target image;
Based on the character information and the target image, function as file data generation means for generating file data of the target image capable of detecting the character information, and character accuracy change means for changing the image quality of the target image in accordance with an input operation. With
The display means displays the character information extracted from the target image with the changed image quality and the output image of the target image,
An image forming program for causing the file data generation means to generate the file data based on the character information extracted from the target image having a changed image quality and the target image.

Images