JP2000295468A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently perform image processing by suppressing degradation of an image. SOLUTION: Image of a document is fetched by an image input part 101, character, photograph, figure, chart and dot areas are extracted from fetched image data by an area identifying part 102, an image class discriminating part 103 discriminates which kind of document the inputted image data are by finding a feature amount such as complexity as a document structure from the presence/ absence of an area to be a key, such as dots or photographs and the degrees of overlapping rectangular areas from the result identified from the characters, photographs, figures, chart or dot for each area. A data transforming part 104 performs transformation processing of an image data format according to the discriminated document kind, and an image output part 105 outputs the output image of the transformed image data format.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for performing a conversion process such as a data format and a density value when filing or copying an image such as a document.

[0002]

2. Description of the Related Art Techniques for capturing a document image as image data and outputting the image as a hard copy or saving the image file as an image file include a technique used for image capture in a copying machine and a personal computer, a filing system and the like. There are technologies used for databases.

[0003] In the case of outputting as a hard copy, image processing is sometimes performed to make it easy for a user to view as a document image. For example, when a character is included in the document image, the edge portion of the character is emphasized by performing the high-pass filter processing, the output becomes clear, and the character becomes easy to read. In addition, in the case of a photograph, a low-pass filter is applied to reproduce a smooth gradation, thereby eliminating a feeling of roughness and providing a beautiful output. If the same photo is a halftone dot photo,
Alternatively, the processing can be switched depending on whether the photograph is a silver halide photograph. As described above, changing the processing method according to the type of the document image is very effective both when outputting as a hard copy and when saving as an image file.

[0004] A document image is usually composed of regions such as characters, figures, photographs, etc., singly or overlapping. Therefore, when performing image processing, it is necessary to detect what area and how are configured in the document image.

Conventionally, in order to identify elements constituting a document image such as a character, a halftone dot, and a photograph, a layout analysis is used as described in Japanese Patent Application No. 8-34702.
There is a method for extracting the structure of a document image. in this way,
After performing the binarization process, the connectivity of pixels is checked, connected pixels are extracted as regions, and identification is performed using feature amounts such as positions and sizes. In the case of this method, since binarization is performed as preprocessing, it has been difficult to accurately extract a region including an intermediate density value such as a photograph.

Further, as described in Japanese Patent Application No. 10-053317, which is an extension of this method, a multivalued image is converted into a plurality of binary images, and a layout is applied to each of the binary images. In some cases, an area including an intermediate density value such as a photograph is accurately extracted by performing analysis.
This method has an advantage that the region can be correctly extracted even when the regions such as characters on the halftone dot overlap.

[0007] An actual document manuscript may have a complicated layout structure. For example, there are magazine covers and advertisements that frequently use graphics, in which characters, figures, and photographs are complicatedly overlapped. If such a document document is subjected to rectangular area identification as described in Japanese Patent Application No. 10-053317, the area may be unnaturally divided. If image processing is performed using the result of the area identification, it may be difficult for the user to see the image.

On the other hand, there is a document manuscript having a very simple layout structure. For example, this case corresponds to a case where the image is composed only of a photograph, or only a character, or only a character and a halftone dot base. In such a case, although the same image processing may be performed on the entire document document, the processing is performed using the identification result for each area, so that it is expected that processing time will be costly.

As described above, the same processing is performed irrespective of the type of the original, which may cause inconvenience.

[0010]

As described above, in a conventional image processing apparatus, a multivalued image is converted into a plurality of binary images, and layout analysis is performed on each of the binary images by using a layout analysis. Since the structure of a document image is extracted as an area and image processing is performed according to the attributes of each area, unnatural areas occur, causing image degradation, unnecessary processing time, and poor efficiency. was there.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an image processing apparatus capable of performing image processing efficiently with less deterioration of an image.

[0012]

SUMMARY OF THE INVENTION An image processing apparatus according to the present invention extracts an area using input means for inputting a document image, and features of pixels from image data input by the input means. And a determination unit for determining the type of the image data from the distribution of the area for each attribute identified by the area identification unit.

An image processing apparatus according to the present invention is characterized in that an input means for inputting a document image, an area is extracted from the image data input by the input means using characteristics of pixels, and an attribute of the extracted area is identified. Area identifying means, determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means, resolution of the image data according to the type determined by the determining means, It comprises a conversion means for converting the compression ratio, the number of colors, and the like, and an output means for outputting the image data converted by the conversion means.

An image processing apparatus according to the present invention is characterized in that an input means for inputting a document image, an area is extracted from the image data input by the input means using characteristics of pixels, and an attribute of the extracted area is identified. Area identifying means for performing the determination, determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means, and performing density conversion or filter processing according to the type determined by the determining means. And correcting means for correcting the image data by applying the image data, and outputting means for outputting the image data corrected by the correcting means.

An image processing apparatus according to the present invention is characterized in that an input means for inputting a document image, an area is extracted from the image data input by the input means using characteristics of pixels, and an attribute of the extracted area is identified. Area determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means; and an image format of the image data according to the type determined by the determining means. And storage means for storing the image data of the image format changed by the changing means.

An image processing apparatus according to the present invention is characterized in that an input means for inputting a document image, an area is extracted from the image data input by the input means using characteristics of pixels, and an attribute of the extracted area is identified. Area identifying means, a determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means, and, according to the type determined by the determining means, image filing, printing, Processing means for selecting an image processing application such as transmission, layout understanding, character recognition, ruled line extraction, etc., setting parameters, activating the application, and performing image processing on the image data, and an image processed by this processing means And output means for outputting data.

[0017]

An embodiment of the present invention will be described below with reference to the drawings.

The first embodiment will be described.

FIG. 1 shows a schematic configuration of an image processing apparatus according to a first embodiment of the present invention. In FIG. 1, 1
01 is an image input unit, 102 is an area identification unit, 103 is an image type determination unit, 104 is a data conversion unit, and 105 is an image output unit.

An image input unit 101 is a device for inputting image data. An image input device such as an image scanner, which is a device for reading a document and converting it into image data, captures an image drawn from a document such as a document. Device. Note that the image input unit in the present invention may be configured by a reading device such as an image scanner, or may be configured by a device that captures an image already stored in a filing device or the like as image data.

The area identification section 102 identifies pixels having characters from the image data input by the image input section 101 as areas. First, the image input unit 101
The input image data is separated into a plurality of binary image data according to the state such as the density difference and the saturation of peripheral pixels, and characters or graphics are physically or logically connected from the image. Each region is divided and extracted, and the feature amount such as the position, size, shape, structure, density distribution and the like of the region is measured, and the attribute as a document element is determined. The types of document elements include text, photographs, figures, tables, and halftone dots.

As a specific method of separating the image data into a plurality of binary image data, there is a known method. For example, the method may be realized by a method disclosed in Japanese Patent Application No. 10-053317. . In this case, as the output of the image separating unit 102, seven binary separated image data of a character image, a halftone image, a background image, a halftone image, a color image, a gray image, and a black image are generated.

As a specific method for extracting and identifying document elements, there is a known method, and for example, the method may be realized by a method disclosed in Japanese Patent Application No. 8-34702. The region identification unit 102 integrates not only single image data but also feature amounts of a plurality of image data according to rules, and determines a region attribute of the input image data. For example, when an area is extracted at the same position from both the character image and the halftone image, the type of the area and the size of the area are determined. As a specific example, when a photographic area exists on a character image and a halftone pixel exists on the halftone image at the same position, it is determined to be a continuous tone photographic area. As a specific method of determining an area attribute from a plurality of image data, there is a known method,
For example, it may be realized by a method disclosed in Japanese Patent Application No. 10-053317.

As will be described in detail later, the image type determination unit 103 becomes a key such as a halftone dot or a photograph based on the result of the region identification unit 102 identifying each region as a character, photograph, figure, table, halftone dot, or the like. By determining a feature amount such as the complexity of the document structure based on the presence or absence of a region and the degree of overlap of the rectangular regions, it is determined what type of document the input image data is.

The type of image data referred to here is, for example, one in which the background is uniform and only characters, figures, tables, and the like are present.

(2) Halftone dots on the base and only characters, figures, tables, etc. are present.

One in which only three halftone photographs exist.

One in which only four consecutive photographs exist.

At least one of five characters, figures, tables, and halftone photographs are mixed, and a rectangular area can be extracted, or at least one of characters, figures, tables, and halftone dots A series of continuous photos, and a rectangular area can be extracted.

At least one of the six characters, figures, tables, and halftone photographs are mixed, and a rectangular area cannot be extracted, or at least one of characters, figures, tables, and halftone dots One is a mixture of one and a series of photographs, and it is impossible to extract a rectangular area.

And so on.

The data conversion unit 104 changes the format of the input image data according to the type of the document determined by the image type determination unit 103, which will be described in detail later. Changing the format of image data includes, for example, resolution, compression ratio,
The number of colors may be changed. Alternatively, attribute area information may be converted into an image. As a specific method of converting area information of an attribute into an image, there is a known method, for example, Japanese Patent Application No. 10-05331.
7 may be realized.

The image output unit 105 includes a data conversion unit 104
The image data converted by the above is output.
The output form may be a hard copy or may be stored in storage as an image file.

Next, the processing of the image processing apparatus of the present invention in such a configuration will be described.

FIG. 2 is a flowchart showing an example of a document processing procedure in the image processing apparatus shown in FIG. The processing flow of the image processing apparatus according to the present invention will be described with reference to this flowchart.

First, an image of a document is fetched by the image input unit 101 (step ST201). That is, the image input unit 101 converts an image read from a document using an image input device such as a scanner or image file data input from a filing system or the like into image data.

The area discriminating unit 102 reads one line or several lines from the image input device, and separates the data into a plurality of binary image data according to the state of each pixel such as the density difference and saturation of peripheral pixels. And extract each area where data, graphics, etc. are physically or logically connected, and measure the amount of features such as position, size, shape, structure, density distribution, etc. of the area, and type of area Then, the character, the photograph, the figure, the table, and the dot area are extracted as a result (step ST202). Repeat until the processing is completed for all pixels of the input image (step ST2
03).

After all pixel attributes have been determined, the image type determination unit 103 identifies characters, photographs, figures, tables, halftone dots, and the like for each area, and determines the key areas such as halftone dots and photographs. By determining feature quantities such as the presence / absence and the degree of overlap of rectangular areas, such as the complexity of the document structure, it is determined what type of document the input image data is (step ST204). The details of the specific processing in the image type determination unit 103 will be described later with reference to FIG.

After determining the type of the input image data as a document, the data conversion unit 104 performs a conversion process of an image data format according to the document type (step ST205). The details of the specific processing in the data conversion unit 104 will be described later with reference to FIG.

Finally, the image output unit 105 outputs an output image whose image data format has been converted (step ST).
206).

The rough processing operation of the image processing apparatus has been described above. Next, details of processing of each element will be described.

FIG. 3 shows an image type determining unit 1 of the image processing apparatus.
FIG. 3 is a flowchart illustrating details of an example of an image type determination process in Step ST2 shown in FIG.
14 is a flowchart of a process performed in 04. The image type determination process is a process of determining what type of input image data is a document by using a region identification result for each pixel.

First, an example of a document type to be classified in this processing will be described.

A document with a uniform background original is one in which a photograph area and a halftone area do not exist. Therefore, either the character area alone or a blank state where nothing is applied corresponds to this.

The halftone base document refers to a document having a halftone area and a character area within the halftone area.

The halftone photo original is one in which a halftone dot region exists, a photograph region exists in the halftone dot region, and no character region exists.

A continuous tone photographic original is one in which a halftone dot area and a character area do not exist and only a photographic area exists.

The rectangular identification document does not satisfy any of the conditions of a uniform background document, a halftone background document, a halftone photograph document, and a continuous tone document, and does not have a complicated layout structure. Points that can be divided into areas such as points, photographs, charts, etc.

A document which has a complicated layout structure and cannot be divided into regions such as characters, halftone dots, photographs, charts, etc. is distinguished from other documents as an unidentifiable document.

In FIG. 3, pixel attribute data 151 is obtained by determining the attribute of each pixel of the input image data using the type and importance of the region of the plurality of image data identified by the region identification unit 102. It is represented as an area for each document element.

First, data is read from the pixel attribute data 151 to check whether a halftone dot area exists (step ST).
301). If a halftone dot region exists, it is checked whether or not a photograph region exists (step ST302). If a photograph area exists, it is checked whether or not a character area further exists (step ST303). If there is a character area, it becomes the rectangle identification candidate document 152. Rectangle identification candidate document 152
Performs a process of verifying the complexity of the layout structure, and classifies the document into a rectangular identification document or an unidentifiable document. A detailed description of this processing will be described later with reference to FIG.

If there is no character area in step ST303, a halftone photographic original 153 is obtained.

If no photograph area exists in step ST302, it is checked whether a character area exists in the halftone dot area (step ST304). If a character area exists in the halftone dot area, the halftone dot original document 154 is obtained. If no character area exists within the halftone dot area, it is further checked whether a character area exists outside the halftone dot area (step ST305). If there is no character area outside the halftone dot area,
Becomes If a character area exists outside the halftone dot area, the document becomes the rectangle identification candidate document 152.

If no halftone dot region exists in step ST301, it is checked whether a photograph region exists (step ST301).
306). If there is no photo area, uniform base document 1
56. If a photograph area exists, it is checked whether or not a character area further exists (step ST307). If there is no character area, a continuous tone photographic original 155 is obtained. If a character area exists, the document becomes a rectangular identification candidate document 152.

By the above processing, the input image data is
It is determined to be one of a rectangle identification candidate document 152, a halftone photo document 153, a halftone background document 154, a continuous tone photo document 155, and a uniform background document 156.

FIG. 4 is a flowchart showing the details of an example of the image type determination processing in the image type determination section 103 of the image processing apparatus of the present invention, which is performed on the rectangular identification candidate document 152 shown in FIG. It is a flowchart of a process.

If it is determined that the document is a rectangle identification candidate document 152, a process for verifying the complexity of the layout structure is performed, and the document is classified as a rectangle identification document or an unidentifiable document. As an example, the degree of overlap of the rectangular areas may be measured. Although the overlap of rectangles may be directly verified, if the number of rectangular areas increases,
It is expected that the number of comparisons will increase dramatically. Therefore, the rectangular size may be voted in a two-dimensional voting space in which the degree of overlap is prepared in advance, and may be estimated from the state of the voting space (step ST401).

Since this voting space is for measuring the degree of overlap of the rectangular areas, the resolution may be lower than the resolution of the original image. The voting is performed by adding a feature such as the rectangular size of the rectangular area to a position where the coordinate position of the rectangular area is converted into the coordinate position of the voting space. As an example of the rectangle size, any of the width, height, area value, and the like of the rectangle can be used. Also, the voting value may be weighted according to the attributes of the rectangular area, that is, attributes such as characters, photographs, charts, and halftone dots. For example, if an attribute that greatly affects the layout structure, such as a photograph or a chart, is given a higher weight and a character or the like is given a lower weight, the voting value of an area where the picture or chart overlaps becomes larger. Considering the complexity of the layout structure, overlapping of photographs and charts is classified as complicated. Therefore, in this case, it can be easily regarded as complicated. Also,
In order to perform the processing at a higher speed, it is possible to limit the rectangular area to be voted. For example, a small character area may not have a significant effect when determining a layout structure. In such a case, by limiting the size of the rectangle to be voted, the number of votes can be reduced, and as a result, it is expected that the speed will be increased. However, when the character area existing in the image data is composed of only a small character area, it cannot be simply excluded. In such a case, the restriction can be adjusted according to the state of the pixel attribute data output from the area identification unit 102.

After all rectangular-size voting in the voting space is completed, the feature amount of the voting space is measured (step S).
T402). As the feature amount, for example, a statistic amount such as a maximum value or a variance may be used, or a voting space may be regarded as a multi-valued image, and after performing a binarization process, an area may be extracted. . In the former case, the determination can be made by comparing the calculated result with a preset threshold. In the latter case, the determination can be made by measuring the graphic features of the extracted region. is there. In step ST402, the former example is used. In this case, when the maximum value max or the variance value var is smaller than the threshold values a or b, the layout structure is considered to be uncomplicated and the rectangular identification document 157 is determined, and when it is larger, the layout structure is considered to be complex. It is determined that the document 158 cannot be identified (step ST40).
3).

The above is the detailed description of the image type determination unit 103. As a result, the input image data is
53, halftone base document 154, continuous tone photographic document 155,
The original is determined to be one of the uniform background original 156, the rectangular identification original 157, and the unidentifiable original 158.

FIG. 5 shows a data conversion unit 10 of the image processing apparatus.
FIG. 4 is a flowchart showing details of an example of data conversion processing in FIG. 4 and is a step ST205 shown in FIG.
3 is a flowchart of the process performed in step S1. The data conversion process is a process of changing the format of input image data using the image type determination result.

The image type information 161 is stored in the image type
03 indicates the type of the input image data, and includes a halftone photo original 153, a halftone background original 154, and a continuous tone photo original 1
55, a uniform base document 156, a rectangular identification document 157, or an unidentifiable document 158.

First, it is checked whether or not the image type information 161 is an unidentifiable document (step ST501). If the original is not an unidentifiable original, it is checked whether the original is a rectangular identification original (step ST502). If the document is a rectangular identification document, data conversion processing is performed on the input image data for each rectangular area using the information on the pixel attributes obtained by the area identification unit 102 (step ST503).

As data conversion processing, for example, resolution,
There are a compression ratio and the number of colors. A rule of the data conversion process may be set in advance for each pixel attribute, and the data conversion process may be performed according to the rule. For example, in the case of a halftone dot photograph region or a continuous photograph region, remarkable image deterioration may not be observed even if conversion for lowering the resolution is performed or the compression ratio is increased, so that the amount of information can be reduced. If only black characters exist in the character area, the amount of information can be reduced by converting the multi-valued data into binary. By setting the rules of the data conversion process for each pixel attribute in advance, it is possible to convert the data into efficient image data.

As a result of the data conversion processing, converted image data 1
Output as 62.

If it is determined in step ST502 that the document is not a rectangular identification document, uniform data conversion processing is performed on the entire image (step ST504). As a uniform data conversion process, for example, when the image type information 161 is a halftone photo original or a continuous tone photograph, remarkable image deterioration is observed even if conversion for lowering the resolution is performed or the compression ratio is increased. In some cases, the amount of information can be reduced. If the image type information 161 is a uniform background document and only black characters exist,
By converting multi-valued data into binary, the amount of information can be reduced in the same manner. By setting the data conversion rule for each type of image in the same manner as in step ST503, it is possible to convert the image data into efficient image data. In this case, since the same processing is performed on the entire image without performing the data conversion processing for each rectangular area as in step ST503, there is an advantage that the processing can be performed at high speed.

As a result of the data conversion processing, step ST50
3, and output as converted image data 162.

If the document is an unidentifiable document in step ST501, the input image data is output as it is as the corrected image data without performing the data conversion process using the information of the pixel attribute obtained by the area identifying section 102 ( Step ST505). The data conversion process (step ST
503 and step ST504) are resolution, compression ratio,
Not only processing for input image data such as the number of colors but also processing for converting attribute area information into image data as described above may be used.

As described above, the type of the input image data is determined from the result of the region identification, and the data conversion process is performed according to the determined type of the image data. It can be converted.

Next, a second embodiment will be described.

The second embodiment is an image processing apparatus that corrects image data by performing appropriate density conversion or filter processing in accordance with the result of determining the type of an image.

FIG. 6 shows a schematic configuration of an image processing apparatus according to the second embodiment. 6, 101 is an image input unit, 102 is an area identification unit, 103 is an image type determination unit, 106 is an image correction unit, 105 is an image output unit,
In this embodiment, the data converter 104 in the first embodiment shown in FIG. Since the image input unit 101, the region identification unit 102, the image type determination unit 103, and the image output unit 105 have already been described, the image correction unit 106 will be described here.

FIG. 7 is a flowchart showing an example of a document processing procedure in the image processing apparatus shown in FIG. FIG.
In the flowchart of the configuration shown in FIG. 1, the data conversion processing (step ST205) is changed to an image correction processing (step ST207).

After determining the type of the image data input by the image type determination unit 103 as a document, the image correction unit 10
In step 6, the image data is corrected by performing appropriate density conversion or filter processing according to the type of the document (step ST207).

FIG. 8 is a flowchart showing details of an example of image correction processing in image correction section 106, and is a flowchart of processing performed in step ST207 shown in FIG.

The image type information 161 is stored in the image type
03 indicates the type of the input image data, and includes a halftone photo original 153, a halftone background original 154, and a continuous tone photo original 1
55, a uniform base document 156, a rectangular identification document 157, or an unidentifiable document 158.

First, it is checked whether or not the image type information 161 is an unidentifiable document (step ST601). If the document is not an unidentifiable document, it is checked whether the document is a rectangular identification document (step ST602). If the document is a rectangular identification document, image correction processing is performed on input image data for each rectangular region using the pixel attribute information obtained by the region identification unit 102 (step ST603).

The image correction processing includes, for example, density conversion processing and filter processing. A rule of the image correction processing may be set in advance for each pixel attribute, and the image correction processing may be performed according to the rule. For example, a high-pass filter process may be performed so that the characters are easy to see in a character region, and a low-pass filter process may be performed in a halftone dot photograph region to obtain a smooth image without moiré. As a result, the image data is corrected to high-quality image data, and can be output as corrected image data 163. By setting the image correction processing rule for each pixel attribute in advance in this way, it is possible to correct image data of high quality.

If the document is not a rectangular identification document in step ST602, uniform image correction processing is performed on the entire image (step ST604). As a uniform image correction process, if the document is a uniform background document, the entire image is subjected to high-pass filter processing so that the characters are easy to see. A low-pass filter may be applied to the whole. As described above, by setting the rule of the image correction process for each type of image as in step ST603, it is possible to correct the image data to high-quality image data. In this case, step ST
Since the same processing is performed on the entire image without performing image correction processing for each rectangular area as in 603, there is an advantage that high-speed processing can be performed. As a result of the image correction processing, step ST6
In the same manner as in step S03, the image data is output as corrected image data 163.

If the document is an unidentifiable document in step ST601, the input image data is directly output as corrected image data without performing the image correction process using the pixel attribute information obtained by the area identification section 102 ( Step ST605).

As described above, by determining the type of the input image data from the result of the region identification and performing the image correction processing according to the determined type of the image data, it is possible to obtain high-quality image data. Become.

Next, a third embodiment will be described.

The third embodiment is an image processing apparatus for selecting an image format to be stored in accordance with the result of discriminating the type of the image, converting the format into the format, and storing the format in filing the image.

FIG. 9 shows a schematic configuration of an image processing apparatus according to the third embodiment. 9, reference numeral 101 denotes an image input unit, 102 denotes an area identification unit, 103 denotes an image type determination unit, 107 denotes an image format change unit, and 105 denotes an image storage unit, and data conversion in the first embodiment shown in FIG. Part 10
4 as an image format change unit 107 and an image output unit 105 as an image storage unit 108. Image input unit 1
01, the region identification unit 102, and the image type determination unit 103 are the same as those already described, and a description thereof will be omitted. Image storage unit 1
08 is the output destination after changing the image format, it can be considered that the same operation as the image output unit 105 is performed. Therefore, here, the image format changing unit 107 will be described.

FIG. 10 is a flowchart showing an example of a document processing procedure in the image processing apparatus shown in FIG. In the flowchart of the configuration in FIG. 1 shown in FIG. 2,
The data conversion process (step ST205) is changed to an image format change process (step ST208).

After the type of the image data input as a document is determined by the image type determining unit 103, the image format changing unit 107 stores the image data in the image storage unit 1 according to the type of the document.
Then, a process of selecting an image format suitable for storing the data in the format 08 is performed (step ST208).

FIG. 11 is a flowchart showing details of an example of the image format changing process in the image format changing unit 107, and is a flowchart of the process performed in step ST208 shown in FIG.

The image type information 161 is stored in the image type
03 indicates the type of the input image data, and includes a halftone photo original 153, a halftone background original 154, and a continuous tone photo original 1
55, a uniform base document 156, a rectangular identification document 157, or an unidentifiable document 158.

First, it is checked whether the image type information 161 is an unidentifiable document (step ST701). If the original is not an unidentifiable original, it is checked whether the original is a rectangular identification original (step ST702). If the document is a rectangular identification document, an image format change process is performed on the input image data for each rectangular region using the pixel attribute information obtained by the region identification unit 102 (step ST703).

The image format includes, for example, the JPEG format and the TIFF (G4) format. A rule for changing the image format may be set in advance for each pixel attribute, and the image format may be changed according to the rule. For example, in the case of a character area, the data amount can be reduced by changing to the TIFF (G4) format, and the image data is changed to efficient image data. Further, if the area is a halftone photographic area or a continuous tone photographic area, the data amount can be compressed while suppressing the deterioration of the image quality by using the JPEG format. In this case, the image data is changed to efficient image data. As a result, the image data can be stored as the format-changed image data 164. In this case, image data having a plurality of formats is generated from one image data. Therefore, when storing, they may be stored separately. At this time, the structure of the original image data and the name of the format-changed image data generated therefrom are held as a reference file, and when restoring, the original image data is converted from the format-changed image data 164 by referring to the reference file. Reconfiguration is also possible. By setting the rule of the image format change process for each pixel attribute in advance in this way, it is possible to change to an efficient image data format with a reduced data amount.

If it is determined in step ST702 that the document is not a rectangular identification document, a uniform image format change process is performed on the entire image (step ST704). As uniform image format change processing, if there is no photo if the original is a uniform background document, TIFF
By changing to the (G4) format, the data amount can be reduced.
By adopting the PEG format, it is possible to compress the data amount while suppressing the deterioration of the image quality. Thus, step S
By setting the rule of the image format change process for each type of image in the same manner as in T703, it is possible to change to an efficient image data format with a reduced data amount. In this case, since the same processing is performed on the entire image without performing the image format change processing for each rectangular area as in step ST703, the image format change processing can be performed at high speed, and a plurality of image formats are generated. Since there is no need for a reference file, there is an advantage that the original image data can be restored at high speed from the stored format-changed image data. As a result of the image format change processing, it is possible to store the format change image data 164 as in step ST703.

If the original is not identifiable in step ST701, the input image data is output as is as the format-changed image data without performing the image format change process using the pixel attribute information obtained by the area identification unit 102. (Step ST705).

As described above, the type of the input image data is determined from the result of the region identification, and the image format is changed in accordance with the determined type of the image data. This makes it possible to reduce the amount of information when storing.

Next, a fourth embodiment will be described.

The fourth embodiment is an image processing apparatus that selects an application to be started, determines start parameters, and starts the application according to the result of determining the type of image.

FIG. 12 shows a schematic configuration of an image processing apparatus according to the fourth embodiment. In FIG. 12, 101
Is an image input unit, 102 is an area identification unit, 103 is an image type determination unit, 109 is an application activation unit, and 110 is an application processing result output unit. The data conversion unit 104 in the first embodiment shown in FIG. The starting unit 109 is used, and the image output unit 105 is used as an application processing result output unit 110. Image input unit 101, area identification unit 102, image type determination unit 1
03 is as described above, and the description is omitted. The application processing result output unit 110 starts the application and is the output destination of the result of the processing. Therefore, it can be considered that the operation is substantially the same as that of the image output unit 105. Therefore, here, the application starting unit 109 will be described.

FIG. 13 is a flowchart showing an example of a document processing procedure in the image processing apparatus shown in FIG.
In the flowchart of the configuration in FIG. 1 shown in FIG. 2, the data conversion process (step ST205) is changed to the application activation process (step ST209), and the image output process (step ST206) is changed to the application process result output process (step ST210). Form.

After the type of the image data input as a document is determined by the image type determination unit 103, an application suitable for performing image processing according to the type of the document is selected by the application activation unit 109, and the activation parameter Is determined, and processing for activating the application is performed (step ST209).

FIG. 14 shows the application starting unit 109.
FIG. 14 is a flowchart showing details of an application activation process as an example in FIG.
It is a flowchart of the process performed in T209.

The image type information 161 is stored in the image type
03 indicates the type of the input image data, and includes a halftone photo original 153, a halftone background original 154, and a continuous tone photo original 1
55, a uniform base document 156, a rectangular identification document 157, or an unidentifiable document 158.

First, it is checked whether the image type information 161 is an unidentifiable document (step ST801). If the original is not an unidentifiable original, it is checked whether the original is a rectangular identification original (step ST802). If the document is a rectangular identification document, an application is started for each rectangular region on the input image data using the information on the pixel attribute obtained by the region identification unit 102, and each image processing is performed (step ST803).

The application may be one that performs image processing such as image filing, printing, facsimile transmission, layout understanding, character recognition, and ruled line extraction. A rule of an application to be activated may be set in advance for each pixel attribute, and the application activation processing may be performed according to the rule. For example, if it is a character area, the layout may be understood, a document structure such as a column may be recognized, and the character may be encoded by character recognition. If the area is a halftone picture area or a continuous tone photographic area, the image filing may be activated to save the image. By attaching the printer, the output destination printer may be automatically changed to a color printer. Depending on the application, the processing result is application processing result data 165, which can take various forms such as a printout, a file, and a character recognition result. By setting the rules of the application start processing for each pixel attribute in advance in this way, various processing can be automatically performed as soon as image data is input.

If it is determined in step ST802 that the document is not a rectangular identification document, the application is activated to perform uniform image processing on the entire image (step ST804). As a uniform image format conversion process, if there is no photo if it is a uniform background document, understand the layout, recognize the document structure such as paragraphs, and perform the character encoding process by character recognition. If it is a halftone or continuous tone photo original, activate image filing to save the image, print directly
X transmission can be performed. The result of the processing by the application is the application processing result data 165 as in step ST803, and can take various forms such as a printout, a file, and a character recognition result.

If the document is an unidentifiable document in step ST801, the application starting process using the information on the pixel attribute obtained by the area identifying unit 102 is not performed.
It is also possible to output the input image data as it is as the application processing result data, or to activate the image filing or the like, store the image data as a file as it is, and then allow the user to select the processing later (step ST805).

As described above, the type of input image data is determined from the result of area identification, and an application for performing image processing according to the determined type of image data is activated. Various processing can be performed in a specific manner, leading to labor saving.

As described above, in the image processing apparatus of the present invention, a document image is fetched by the image input unit 101, and the area identification unit 102 physically or logically connects the document image based on the state of the data of each pixel and neighboring pixels. After extracting what is done as one area, measure the feature amount such as position, size, shape, structure, density distribution etc. of each area on the image, and based on the measurement results based on predetermined rules Then, identification is performed as an area having attributes such as characters, photographs, and halftone dots as document constituent elements. Then, based on the attribute information and the like, the image type determination unit 103 determines the type of the document from the characteristics such as the complexity of the entire document image input, and the data conversion unit 104 determines the image data format according to the type of the document. Can be converted.

The image correcting section 106 performs an appropriate density conversion or filter processing in accordance with the type of the document to correct the image data, and the image format changing section 107 stores the image in the image storage section according to the type of the document. In this case, it is possible to select an image format suitable for storing in the storage unit 108 and to perform processing for changing the format.

Further, in the application starting unit 109, it is possible to select an application suitable for performing image processing according to the type of document, determine a starting parameter, and perform a process of starting the application. .

Therefore, in the document image data input by the present apparatus, areas are extracted by distinguishing different document elements such as characters, photographs, charts, and halftone dots. By determining a feature amount such as complexity as a document structure from the degree of overlap of the rectangular areas, it is possible to determine what kind of document the input image data is, and use the determination result. By doing so, it is possible to convert the image data into efficient image data with little deterioration in image quality, and to perform image correction for obtaining high-quality image data.

Also, when saving as an image file, a format suitable for image data can be automatically selected, and efficient filing can be performed.

Further, as soon as the document image data is input,
Since the image processing application is automatically selected and started, labor is saved.

As described above, according to the embodiment of the present invention, when a document is to be hard-copyed or converted into image data and stored, characters, figures, Different document elements such as tables, photos, halftone dots, etc. are distinguished and extracted and identified, and features such as the presence / absence of key areas such as halftone dots, photos, and the degree of overlap of rectangular areas, such as the complexity of the document structure, are determined. By obtaining, it is possible to determine what kind of document the input image data is. By using this determination result, the most suitable image format can be selected when performing data conversion processing such as resolution, compression ratio, number of colors, filtering processing, and saving as an image file, so that efficient filing can be performed. become.

Also, an image processing application suitable for processing input image data can be started.

[0114]

As described in detail above, according to the present invention,
It is possible to provide an image processing apparatus capable of performing image processing efficiently with less deterioration of an image.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of an image processing apparatus according to a first embodiment of the present invention.

FIG. 2 is a flowchart illustrating an example of image processing in the image processing apparatus according to the first embodiment.

FIG. 3 is a flowchart illustrating an image type determination process in the image processing apparatus according to the first embodiment.

FIG. 4 is a flowchart illustrating an image type determination process in the image processing apparatus according to the first embodiment.

FIG. 5 is a flowchart illustrating data conversion processing in the image processing apparatus according to the first embodiment.

FIG. 6 is a block diagram illustrating a schematic configuration of an image processing apparatus according to a second embodiment of the present invention.

FIG. 7 is a flowchart illustrating an example of image processing in the image processing apparatus according to the second embodiment.

FIG. 8 is a flowchart illustrating an image correction process in the image processing apparatus according to the second embodiment.

FIG. 9 is a block diagram illustrating a schematic configuration of an image processing apparatus according to a third embodiment of the present invention.

FIG. 10 is a flowchart illustrating an example of image processing in the image processing apparatus according to the third embodiment.

FIG. 11 is a flowchart illustrating an image format change process in the image processing apparatus according to the third embodiment.

FIG. 12 is a block diagram illustrating a schematic configuration of an image processing apparatus according to a fourth embodiment of the present invention.

FIG. 13 is a flowchart illustrating an example of image processing in the image processing apparatus according to the fourth embodiment.

FIG. 14 is a flowchart illustrating an application activation process in the image processing apparatus according to the fourth embodiment.

[Explanation of symbols]

 101 image input unit (input means) 102 area discriminating unit (identifying means) 103 image type determining unit (determining means) 104 data conversion unit (converting means) 105 image output unit (output means) 106 image correction Unit (correction unit) 107 ... Image format change unit (change unit) 108 ... Image storage unit (storage unit) 109 ... Application start unit (start processing unit) 110 ... Application processing result output unit (output unit)

Continuing from the front page (72) Inventor Naorou Kodaira 1 Tokoba, Komukai Toshiba-cho, Kawasaki-shi, Kanagawa Prefecture Inside the Toshiba R & D Center (72) Inventor Hiroaki Kubota 1, Komukai Toshiba-cho, Koyuki-ku, Kawasaki-shi, Kanagawa F-term in Toshiba Research and Development Center Co., Ltd. (reference) PP28 PP37 PQ22 5L096 BA12 BA17 EA03 FA43 FA44 FA45 LA05

Claims (6)

[Claims] 1. An input means for inputting a document image, an area identifying means for extracting an area from image data input by the input means using characteristics of pixels, and identifying an attribute of the extracted area; Determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means. 2. The image processing apparatus according to claim 1, wherein the image data of the uniform background, the image data of the entire image having the halftone background, the image data of the halftone photograph only, the image data of the continuous tone photograph only, and the area identification based on the rectangle are provided. 2. The image processing apparatus according to claim 1, wherein at least one type is determined from possible image data and image data in which a region cannot be identified by a rectangle. 3. An input means for inputting a document image, an area identifying means for extracting an area from the image data input by the input means by using a feature of a pixel, and identifying an attribute of the extracted area; Determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means; resolution, compression ratio, number of colors, etc. of the image data according to the type determined by the determining means An image processing apparatus comprising: a conversion unit configured to convert the image data; and an output unit configured to output image data converted by the conversion unit. 4. An input means for inputting a document image, an area identifying means for extracting an area from the image data input by the input means using a feature of a pixel, and identifying an attribute of the extracted area; Determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means; and performing density conversion or filter processing according to the type determined by the determining means to convert the image data. An image processing apparatus comprising: a correcting unit for correcting; and an output unit for outputting image data corrected by the correcting unit. 5. An input means for inputting a document image, an area extracting means for extracting an area from the image data input by the input means using characteristics of pixels, and identifying an attribute of the extracted area; Determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means; changing means for changing the image format of the image data according to the type determined by the determining means; A storage unit for storing image data in the image format changed by the changing unit. 6. An input means for inputting a document image, an area extracting means for extracting an area from the image data input by the input means using characteristics of pixels, and identifying an attribute of the extracted area; Determining means for determining the type of the image data from the distribution of the area for each attribute identified by the area identifying means; and image filing, printing, transmission, layout understanding, and text according to the type determined by the determining means. Processing means for selecting an application for image processing such as recognition and ruled line extraction, setting parameters, and activating the application to perform image processing of the image data; and output means for outputting image data image-processed by the processing means An image processing apparatus, comprising: