JPH07121655A - Document reading device - Google Patents

Abstract

PURPOSE:To attain the input of images of high picture quality in a photo area without deteriorating the recognition rate of a character area nor deteriorating the picture quality of a graphic area. CONSTITUTION:This document reading device consists of an image input part 11 which scans the characters, graphics or photos, etc., on a recording medium to acquire the image data, an image member 12 which stores the image data, and area identifying part 13 which discriminates the character, graphic and photo areas based on the image data, a result output part 15 which outputs the character codes, the graphic data and the photo data obtained from the image data corresponding to those areas discriminated by the part 13 through a character recognizing part 14, etc., as the document data, an output device 16 which is represented by a printer, etc., a graphic data reading part 18 which reads and supplies the image data based on the area recognizing result acquired by the part 13, a photo data input part 19, and a control part 17 which controls the whole of the device.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document reading device for reading characters, figures, photographs, etc. described in a document.

[0002]

2. Description of the Related Art A conventional document reading apparatus for reading a document in which characters, figures, photographs, etc. are mixed, first scans the characters, figures, or photographs in order to recognize the characters in an input section, and then simply scans them. Input the image data by converting the value. The image data obtained here is temporarily stored in the memory. Then, the identification section identifies each area such as a character, a figure, or a photograph based on the image data stored in the memory. Then, for the portion of the character area in the identification result obtained by the identification unit, the character code is output after performing the character recognition processing, and for the remaining graphic area or photograph area,
It is output as simple binary image data.

[0003]

However, the conventional document reading device described above has the following disadvantages. In the conventional document reading apparatus, since the image is input by the simple binary conversion in the same manner as the character area or the graphic area in the photo area, image deterioration such as moire fringes occurs. Here, the moire fringes are vertical and horizontal stripes that are included when a halftone image is copied. On the contrary, in order to maintain the image quality of the photo area, if not only the photo area but also the character area and the graphic area are input by the pseudo multi-value conversion, the recognition rate of the character area decreases and the image quality of the graphic area decreases. Had also deteriorated.

The present invention has been made in order to solve such inconveniences of the prior art, and the high image quality is achieved even in the photographic area without lowering the recognition rate of the character area and the image quality of the graphic area. It is an object of the present invention to provide a document reading device capable of inputting the image.

[0005]

The document reading device of the present invention is an image input means for scanning characters, figures or photographs recorded on a recording medium, converting them into image signals, and further converting the image signals into digital signals. And an image memory for storing the image data obtained by the processing of the image input means, and an area identification for identifying each area such as a character, a figure or a photograph from the image data stored in the image memory. Means and the area identification result thus obtained by the area identifying means, a line and a character in the line are cut out from the image data in the image memory corresponding to the character area, and the character Character recognition means for recognizing each image data to obtain a character code, and the image data in the image memory corresponding to a graphic area in the area identification result Graphic data read-out means for reading out as graphic data, and photograph data input means for digitally converting the image data in the image memory corresponding to the photograph area out of the area identification result in the image input means to input it as photograph data. And output means for outputting the character code, the graphic data, and the photograph data as document data in an arbitrary format.

Further, in the document reading apparatus of the present invention, image data obtained by digitally converting, in the image input means, an area wider than the photographic area identified by the area identifying means by the photographic data input means. Is input and stored in the image memory, and the photographic area in the image data is identified by the area identifying means, and the image data corresponding to the identified photographic area is used as photographic data. It is what

[0007]

According to the document reading apparatus of the present invention, the means for inputting an image at the time of capturing as image data is changed according to each area such as a character, a figure, or a photograph recorded on the recording medium. As a result, it is possible to input and output image data having a high character recognition rate and high image quality without a reduction in the recognition rate of the character area or deterioration of the image such as moire fringes. Further, according to the document reading apparatus of the present invention, when detecting a photographic area, the image data that has been subjected to area identification and once subjected to input processing is subjected to area identification for a wider area, and image data obtained by this Is output as the image data of the photo area, the correctly recognized photo is not affected by the deviation of the document itself caused by inputting the image twice and the change of the position of the photo area. It becomes possible to input and output the image data of the area.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a document reading device of the present invention, and FIG. 2 is a flow chart showing a first embodiment of the document reading device of the present invention. The image input unit 11 scans a document to be read, converts characters, figures, photographs, etc. recorded on the document into an image signal by photoelectric conversion,
Further, this image signal is converted into binary image data. At this time, the conversion of the image signal into binary image data is simple binary or pseudo multi-level conversion. In addition, the image input unit 11 also transfers image data to the image memory 12. The image memory 12 is a memory for storing the image data transferred from the image input unit 11, and can be accessed by the area identification unit 13, the character recognition unit 14, and the result output unit 15, which will be described later. The area identification unit 13 identifies the image data stored in the image memory 12 for each area such as a character, a figure, or a photograph, and the area identification result is obtained by the image input unit 11, the character recognition unit 14, and the result. Transfer to the output unit 15.

Next, the character recognizing unit 14 operates the image memory 12
Of the image data stored in the area identification unit 13
A character detected from the image data corresponding to the area identification result (character area in this case) by the identification processing in step S1 is recognized as a character code, and the result output unit 15 of this character code is recognized.
Transfer to. As a result, the result output unit 15 corresponds to each of the graphic area and the photograph area based on the character code obtained from the character recognition unit 14 and the area identification result read out from the image memory 12 and identified by the area identifying unit 13. The image data and the document data are converted into document data of an arbitrary format and output to the output device 16. Output device 1 here
Although not shown, 6 indicates a magnetic disk device for storing the document data sent from the result output unit 15, a printer for printing the document data, and the like.

Then, the graphic data reading section 18 reads out from the image memory 12 the image data corresponding to the graphic area of the area identification result transferred from the area identifying section 13, and the photographic data input section 19 Area identification unit 13
The image data obtained by digitally converting the photographic area of the area identification result transferred from is input.
The control unit 17 also controls each operation of the entire document reading apparatus.

Next, the processing operation of the first embodiment of the document reading apparatus configured as described above will be described. FIG. 3 is an example of an image showing the entire one page of the read document, and FIG. 4 is an explanatory diagram showing the area identification result by the area identification unit. First, an image is input (step S1). Specifically, by the image input unit 11, all characters, figures, photographs, etc. of the entire one page recorded on the document to be read this time,
It is converted into an image signal by photoelectric conversion, further, image data is created by simple binarization conversion of this image signal, and this image data is stored in the image memory 12. An example of an image of the entire page of the document at this time is as shown in FIG. When the image input is completed, the area identification unit 13 causes the image data stored in the image memory in step S1 to display the character area 41, the graphic area 42 or the photograph area 43 as shown in FIG.
A region identification process for identifying is performed (step S2). This area identification processing is performed using, for example, a method using a peripheral distribution histogram of image data (see geometrical characteristics). After that, the area identification unit 13 transfers the area identification result to the image input unit 11, the character recognition unit 12, or the result output unit 15, respectively.

Next, character recognition processing is performed by the character recognition unit 14 (step S3). The character recognition unit 14 detects a line from the image data stored in the image memory 12 corresponding to the character area 41 of the area identification result transferred from the area identification unit 13, and further detects a character from the line. To detect. Then, the character code is obtained by referring to the image memory 12 again for the detected image data of each character, and recognizing the image data of each character. All of these character codes are transferred to the result output unit 15. Subsequently, the image data corresponding to the graphic area 42 in the area identification result transferred from the area identification unit 13 is read from the image memory 12 by the graphic data reading unit 18 (step S4). Then, the image input unit 11 obtains the document to be read this time, which has been image-inputted in step S1, by subjecting the photographic region 43 of the region identification result transferred from the region identification unit 13 to pseudo multi-value conversion. The obtained image data is input from the photograph data input unit 19 (step S5), and further, these image data (graphics, photographs) are transferred to the result output unit 15 via the image memory 12.

Finally, in the result output section 15, the character area 4
The character code obtained in step S3 for 1 and the image data obtained in steps S4 and S5 for the graphic area 42 and the photograph area 43 are converted into document data of an arbitrary document format, The process ends by outputting the data to the output device 16 (step S6). As described above, the character region 41 and the graphic region 42 are image-inputted by the simple binary conversion, and the photograph region 43 is image-inputted by the pseudo multi-valued conversion. It is possible to output high-quality image data even in the photographic area while maintaining the high image quality in the above.

Next, a second embodiment of the present invention will be described in detail with reference to the drawings. In explaining the second embodiment,
The description of the portions, such as the configuration and operation shown in the block diagram showing the document reading apparatus of the present invention, which are the same as those in the first embodiment, will be omitted. FIG. 5 is a flowchart showing a second embodiment of the document reading apparatus of the present invention, FIG. 6 is an explanatory view showing an enlarged area of the photograph area, and FIG. 7 is image data of the enlarged photograph area. It is an explanatory view showing. After the reading of the image data corresponding to the graphic area 42 in step S4 of FIG. 2 described above is completed, the image input unit 11 inputs an image by a pseudo multi-value conversion in an area wider than the photograph area (step S5).
1). Specifically, with respect to the reading target document image-inputted in step S1, an enlarged area 62 (see FIG. 6) wider than the photograph area 61, which is the area identification result in the area identification unit 13, is displayed in FIG. Image data 7
It is input as 2. Of course, this image data 72 is also stored in the image memory 12.

Here, an example of how to obtain the enlarged area will be introduced. The enlarged region 62 shown in FIG. 6 is in the X direction (image input unit 1
1 in the main scanning direction: left / right direction, w = W / 30 (W is the X direction of the image of the entire document: page width) is enlarged in the left and right directions by w, and the Y direction (image input unit 1
1 in the sub-scanning direction: up and down), h = H / 30 (H is Y direction of the image of the entire document: page height) and an area enlarged in both up and down directions by h is shown. Then, when the image input of the enlarged area 62 shown in step S51 is completed, the area identifying unit 13 identifies the photograph area 71 from the image data 72 (see FIG. 7) stored in the image memory 12. Is performed (step S52), and the identification processing result here is transferred to the result output unit 15. Upon receiving this, the result output unit 15
The image data corresponding to the photographic area 71, which is the area identification result sent from the area identification unit 13, is read from the image memory 12 (step S53), and the image data read in step S53 is read as in the first embodiment. All the processing is completed by converting the document data into an arbitrary document format and outputting the document data to the output device 16.

Thus, in the second embodiment of the present invention,
When the photographic area is detected, the image data once subjected to pseudo multi-value conversion is input to the area identifying unit 13 again to identify a wider area. As a result, the first image input (step S1: image input by simple binary conversion)
And a second image input (step S51: image input by pseudo multi-value conversion), which causes a deviation of the document itself, and a photographic area due to a change in image data generated at the first and second photoelectric conversions. It is possible to input correctly recognized image data of a photographic area without being affected by a change in the position of the. Lastly, in the first and second embodiments, the pseudo multi-value conversion has been described as an example of the means for inputting the image data, but the present invention is not limited to this, and the monochrome It is effective even for multi-value conversion and re-input by color.

[0017]

As described above, according to the document reading apparatus of the present invention, an image is input when capturing it as image data according to each area such as a character, a figure, or a photograph recorded on a recording medium. By changing the means, it becomes possible to output image data with high character recognition rate and high image quality without deterioration of recognition rate of character areas and deterioration of images such as moire fringes. It was Further, according to the document reading apparatus of the present invention, when detecting a photographic area, the image data that has been subjected to area identification and once subjected to input processing is subjected to area identification for a wider area, and image data obtained by this Since it is output as image data of the photo area, the correctly recognized photo area is not affected by the deviation of the document itself caused by inputting the image twice and the change of the position of the photo area. It became possible to input and output the image data of.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a document reading device of the present invention.

FIG. 2 is a flow chart showing a first embodiment of the document reading device of the present invention.

FIG. 3 is an example of an image showing an entire page of a read document.

FIG. 4 is an explanatory diagram showing an area identification result by an area identification unit.

FIG. 5 is a flowchart showing a second embodiment of the document reading device of the present invention.

FIG. 6 is an explanatory diagram showing an enlarged area of a photograph area.

FIG. 7 is an explanatory diagram showing image data of an enlarged photographic area.

[Explanation of symbols]

 11 Image Input Section 12 Image Memory 13 Area Identification Section 14 Character Recognition Section 15 Result Output Section 16 Output Device 17 Control Section

Claims (2)

[Claims] 1. An image input means for scanning a character, a figure, a photograph or the like recorded on a recording medium to convert the image signal into an image signal, and further digitally converting the image signal, and an image obtained by the processing of the image input means. An image memory for storing data, an area identifying means for identifying each area of characters, figures, photographs, etc. from the image data stored in the image memory, and the area identifying means. A character recognizing unit that obtains a character code by cutting out a line and a character in the line from the image data in the image memory corresponding to a character region in the region identification result, and recognizing the image data of each character just cut out. And graphic data reading means for reading, as graphic data, the image data in the image memory corresponding to a graphic area in the area identification result. Of the area identification result, the image data in the image memory corresponding to the photograph area is digitally converted into photograph data by the image input means, and the photograph data input means for inputting the photograph data, the character code And a output unit for outputting the graphic data and the photo data as document data. 2. The photographic data input means inputs image data obtained by digitally converting an area wider than the photographic area identified by the area identifying means by the image input means, and stores the image data in the image memory. 2. The document according to claim 1, wherein the document data is stored, and the photo region in the image data is identified by the region identification means, and the image data corresponding to the photo region identified here is taken as the photo data. Reader.