JPH04220086A - Image data control system - Google Patents

Abstract

PURPOSE:To efficiently coding even if the image area of a document has complicated structure which is not separated into a rectangular area. CONSTITUTION:Image data is inputted to a picture element attribute deciding part 2 and decided whether it is a picture element in a gradation area or a picture element in a monochromatic area. At every area devided by an area dividing part 3, an area deciding part 4 decides color/monochrome and character/natural picture, background. A coding processing part 5 imparts an area number to each area, a header is constituted of the area number and an area attribution, and at every kind of area, coding data is stored in a memory 6. At this time the background and a base aren't coded.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data management method in an image recording device such as an electronic file.

0002

[Prior Art] In recent years, with the development of multimedia,
The number of office documents expressed in color images is increasing. Color document images like this require a much larger amount of data than conventional monochrome binary documents.
Even when a large capacity storage medium such as an optical disk is used, compression processing is required to reduce the amount of data when storing color document images.

By the way, in the case of office documents, the text area still occupies most of the text area, and when compressing document data that includes natural images such as photographs, compression techniques suitable for both binary images and natural images are needed. Since it has not yet been developed, conventionally, document data is compressed by distinguishing between binary image areas and natural image areas, and using a compression method suitable for each area.

For example, one-dimensional encoding (Modified Huffman encoding method,
MH) or two-dimensional sequential processing encoding (Modify
d MR encoding method (MMR), etc., and performs reversible compression such as
ADCT (Adaptive
Discrete Cosine Transform
er, applied discrete cosine transform).

[0005]

[Problems to be Solved by the Invention] However, as the document structure becomes more complex, the adaptive encoding process that involves separation as described above not only becomes more complex, but also the compression effect does not improve due to the increase in header information. Therefore, it is desired to structure the document as simply as possible during adaptive encoding.

[0006] From this point of view, an adaptive encoding system including document structure description has been proposed (1988 Institute of Electronics, Information and Communication Engineers Spring National Conference, 1-317, 318). This method separates and extracts individual regions such as text, photographs, and diagrams from a document image, performs optimal encoding processing on each region, and uses a document structure description language to create documents containing attribute information. Accumulating image data.

[0007] Since the description using the above document description language assumes that the constituent elements of the document are rectangular areas, documents (for example, technical documents) created with a word processor or DTP (desk top publishing) cannot be used. However, in the case of general printed materials such as catalogs, it is not possible to clearly separate document components into rectangular areas, so separating and extracting in rectangular areas results in extremely fine areas. There is a problem in that the encoding efficiency deteriorates due to the occurrence of overlapping regions or the occurrence of overlapping regions.

SUMMARY OF THE INVENTION An object of the present invention is to provide an image data management system that can efficiently encode even if the image area of a document has a complex structure that cannot be separated into rectangular areas.

[0009]

[Means for Solving the Problems] In order to achieve the above object, in the invention according to claim 1, regions are separated from a document image in which a plurality of different regions coexist, and an optimal In image processing methods that perform encoding processing,
The present invention is characterized in that an image data code string is constructed by header information in which attribute information for identifying the separated area is encoded and encoded data following the header information.

According to a second aspect of the invention, the plurality of areas include a black character area, a color character area, a color gradation area, a monochrome gradation area, a color background area, and a base area.

[0011] According to a third aspect of the present invention, among the plurality of regions, the gradation region and the character region are encoded, and the background region and the base region are not encoded.

[0012] According to a fourth aspect of the invention, the attribute information includes an area number.

[0013]

[Operation] Image data is input to the pixel attribute determination unit, which determines whether the pixel is in a gradation area or a monochromatic area, and the area division unit divides the area based on the determination result. For each divided area, the area determination unit determines color/monochrome, text/natural image, and background. The encoding processing unit assigns an area number to each area, configures a header with the area number and area attribute, and stores encoded data in a memory for each type of area.

Therefore, according to the present invention, it is possible to efficiently manage encoded data even for a document having a complicated structure in which the image area of the document cannot be separated into rectangular areas, and it is possible to efficiently manage encoded data. Since it is not encoded, the amount of stored data can be reduced.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. To explain the case where the present invention is applied to an electronic filing device as an example, the electronic filing device reads a color document with an optical reading device such as a color scanner, stores it in a storage device such as an optical disk, and stores it in a storage device such as an optical disk. This is a system that searches and outputs according to the information. Then, when storing the image data in the storage device, compression processing (MH, MMR, ADCT, etc.) is performed.

[0016] Furthermore, the color document image to which the present invention is directed is an image containing various constituent elements such as a color natural image, color text, black text, background, color background, and diagrams.

FIG. 1 is a block diagram for implementing the image data management system of the present invention. 1 is an image input unit that inputs a color document image using a multi-gradation image reading device; 2 is a pixel attribute determination unit that determines the attribute of each pixel; 3
4 is an area determining unit that divides the area based on attribute information; 4 is an area determining unit that determines the divided area; 5 is an encoding processing unit that performs adaptive encoding for each area; 6 is encoded image data 7 is a control section that controls each section.

FIG. 2 shows an example of a multimedia document including three color natural images 21, 22, 23, two color text areas 24, 25, a black text area 26, a background 27, and a color background 28 in one image. It shows.

The operation of the present invention will be explained in detail below with reference to the flowchart of the adaptive encoding process shown in FIG. 3. In step 31, one image is read by the multi-tone image reading device 1, The read image data is input to the pixel attribute determination section 2. The pixel attribute determining unit 2 determines the attribute of each pixel using the local area information, and determines whether the pixel is in a gradation area or a monochromatic area (step 32). In step 33, based on the attribute information determined by the pixel attribute determining section 2, the region dividing section 3 generates a cluster of regions. At this time, a small single-color area surrounded by the gradation area is included in the gradation area, and processing is performed so that the area is not extremely subdivided. Next, in step 34, based on the pixel data of the divided areas,
The area determination unit 4 determines whether the image is color/monochrome, text/natural image, background, etc.

[0020] The regions are separated by the above steps, and there are various algorithms for such region separation methods, such as the method described in the above-mentioned literature, and any one of them may be adopted.

When the separation of regions is completed, in step 35, the encoding processing unit 5 performs encoding by selecting the optimal encoding method for each region. For example, monochrome/
For color character areas, encode by MMR,
Monochrome/color natural image areas are encoded by ADCT.

Next, the encoding processing unit 5 performs step 36
Management of encoded data, which is a feature of the present invention, is performed as follows. That is, suppose that as a result of region separation of the color document image of FIG. 2, regions as shown in FIG. 4 are obtained. In FIG. 4, regions of the color natural images 21, 22, and 23 and the background portion 28 of FIG. 2 are extracted as irregular shapes.

[0023] In the present invention, information on each of these irregularly shaped areas is grouped with information having the same area attribute, and encoded data is structured in a format corresponding to this area attribute. That is, as shown in FIG. 4, an area number is given to each area in order from the upper left of the figure, and encoded data is stored in the memory 6 along with the attributes of the area. Then, a header is configured with the area number and area attribute, and encoded data is stored in the memory 6 for each type of area in the format shown in FIG. Note that the area attribute value is the attribute of each area (monochrome single color, multivalue, color single color, multivalue)
This is a predetermined value given to distinguish between

In FIG. 5, what is actually encoded are natural image areas such as photographs and character areas, and the background stores its area number, area attribute value, and color code, and the background itself is not encoded. . In addition, since the density information has no meaning in the area corresponding to the paper base, no information is stored therein.

Among the color document images shown in FIG. 2, for example, color natural images 21 and 22 are stored or transmitted in the following format. That is, area number 4, area attribute value (color natural image), image encoding code within area number 4,
Area number 5, area attribute value (color natural image), area number 5
The images are stored in the order of the image encoding codes within.

In the present invention, in order to identify the position of each region during decoding, the entire image to which the region numbers shown in FIG. 4 are assigned is encoded. At this time, since the same values are continuous in each area, the amount of data to be accumulated is compressed by run-length encoding. Using the positional information of the run-length encoded area as the leading information, the header and encoded code data are stored for each type of area described above.

[0027]

Effects of the Invention As explained above, claims 1, 2,
According to the invention described in item 4, encoded data can be efficiently managed even for a document having such a complicated structure that the image area of the document cannot be separated into rectangular areas. Furthermore, according to the third aspect of the invention, since the background and base portions are not encoded, the amount of accumulated data can be reduced, and as a result, the decoding process can be performed at high speed.

[Brief explanation of the drawing]

FIG. 1 is a block configuration diagram for implementing an image data management method of the present invention.

FIG. 2 is a diagram showing an example of a multimedia document.

FIG. 3 is a flowchart of adaptive encoding processing of the present invention.

FIG. 4 is a diagram in which regions are separated.

FIG. 5 is a data format corresponding to the type of area stored in the memory.

[Explanation of symbols]

1 Image input section 2 Pixel attribute determination section 3 Area division part 4 Area determination section 5 Encoding processing unit 6 Memory 7 Control section

Claims (4)

[Claims] Claim 1: In an image processing method that separates regions from a document image in which a plurality of different regions coexist and performs optimal encoding processing on each of the separated regions, for identifying the separated regions. An image data management system characterized in that an image data code string is constructed by header information in which attribute information is encoded and encoded data following the header information. 2. The image data management method according to claim 1, wherein the plurality of areas include a black text area, a color text area, a color gradation area, a monochrome gradation area, a color background area, and a base area. . 3. The image data management system according to claim 1, wherein among the plurality of areas, the gradation area and character area are encoded, and the background area and base area are not encoded. 4. The image data management system according to claim 1, wherein the attribute information includes an area number.