JPH0591347A - Picture encoding device - Google Patents

Abstract

PURPOSE:To encode picture data by a proper method by detecting the picture classification information of an input picture and position information belonging to it. CONSTITUTION:When a blocking means 1 divides an input digital picture signal into the blocks of MXN picture elements (M and N are natural numbers), a detection range setting means 2 sets plural blocks out of divided blocks as the detection range. A detection range scanning means 3 moves the set detection range throughout the area of picture information with each divided block as the movement unit, and a picture classification discriminating means 4 discriminates the picture classification in the detection range at each time of movement. Each time the discriminating operation is terminated, a pertinent block adding means 5 adds the result to a storage means 6. A picture area discriminating means 7 refers to the addition result to discriminate the picture classification and the position, and an encoding means 8 encodes picture data based on discrimination information.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image coding apparatus for appropriately coding image data output from an image processing apparatus by applying a compression method according to its type.

[0002]

2. Description of the Related Art Conventionally, as an apparatus for compressing image data and transmitting the compressed image data, a facsimile apparatus which performs binarization according to a threshold value set for density information of the image data is known.

Generally, the run-length coding method is suitable for compressing binary pixels of white / black such as characters, but even in the case of a binary image, a binary halftone dot image is used. An image compression unit that compresses by referring to adjacent halftone dots by utilizing the periodicity of point image data is suitable. However, in the conventional facsimile apparatus, the same compression method is applied to both the character image and the halftone dot image. Therefore, the overall compression efficiency is low, and various attempts have been made to solve these drawbacks.

For example, in Japanese Utility Model Publication No. 1-2001, "First partial image data which is a binary character image mixed over one page is converted into a second partial image data which is a binary halftone image. On the other hand, the image data for one page to which the control information including the image type information indicating the type of the first and second partial image data and the position information indicating the position occupied in the one page is added. Receiving from the main processing unit, the means for determining the control information for compression,
Based on this determination result, a first image compression means for compressing the first partial image data by a compression method suitable for a character image, and the second partial image data are adjacent to each other by utilizing the periodicity of the binary halftone image data. An image for one page compressed by selecting second image compression means for compressing with reference to halftone dots and compressed first and second partial image data by control information and adding control information. An image information compression device characterized by comprising selective transmission means for transmitting as data "is disclosed. That is, in this apparatus, the overall compression efficiency is improved by using the optimum compression method for each of the character image and the halftone image.

Further, Japanese Patent Laid-Open No. 3-64168 discloses that
The original is read optically and the read information is 2
In an image coding apparatus that performs a three-dimensional discrete cosine transform to perform compression processing, the two-dimensional discrete cosine is referred to by referring to a plurality of bit allocation tables corresponding to a plurality of document contents such as characters, images, and a mixture of characters and images. A technique of performing coding on a converted output is disclosed. The compression method is the same as the conventional one, but the quantization unit aims to improve the overall compression efficiency by performing quantization with the optimal bit allocation table for each of the character image and the image image. Is.

[0006]

However, when the read image can include both an image image such as a color image, a natural image, and a halftone dot image and a character image, the above described fairness 1- No. 25001 and Japanese Unexamined Patent Publication No.
In any of the techniques disclosed in Japanese Patent Laid-Open No. 64-16868, in combination with the input of image data, whether the input image data is a character image or an image image, or a mixture thereof, or a mixture thereof Then, unless the image type information or the attached position information indicating from which area the area of the character image becomes the area, the quantization cannot be performed by the optimum compression method or the optimum bit allocation table. There was a drawback.

The present invention has been made by paying attention to such a problem, and its object is to provide an optimum compression method or quantized bit for an image type of an entire image or a partial image of an input image. To provide an image encoding device capable of detecting an image type only from an input image and performing appropriate encoding on image data based on the image type information for the purpose of compression of the image. is there.

[0008]

In order to achieve the above object, the image coding apparatus of the present invention is a block which divides an input digital image signal into blocks of M × N (M and N are natural numbers) pixels. The blocking means 1, the detection range setting means 2 for setting the detection range with a plurality of blocks divided by the blocking means 1, and the detection range set by the detection range setting means 2. With the block divided by as a movement unit, the detection range scanning unit 3 that moves the detection range for the entire area of the image information, and the image type within the detection range is determined every time the detection range scanning unit 3 moves the detection range. The image type judging means 4 and the storing means 6 having areas corresponding to all the blocks divided by the blocking means 1.
By referring to the corresponding block addition means 5 that adds the result to the storage means 6 each time the image type determination means 4 finishes the determination operation, and the addition result by the corresponding block addition means 5. It is characterized by including an image area discriminating means 7 for discriminating an image type and its position, and an encoding means 8 for encoding image data on the basis of discrimination information by the image area discriminating means 7.

[0009]

That is, in the image coding apparatus of the present invention, the blocking means 1 converts the input digital image signal into M × N.
When a block is divided into blocks of (M and N are natural numbers) pixels, the detection range setting means 2 sets the detection range with a plurality of the divided blocks. Then, the detection range scanning means 3 moves the detection range with respect to the entire area of the image information by using the block divided by the blocking means 1 as the detection range set by the detection range setting means 2 as a movement unit, and the image type The determination unit 4 determines the image type within the detection range for each movement by the detection range scanning unit 3. Each time the image type judging means 4 finishes the judging operation, the corresponding block adding means 5 stores the result in the storage means 6 having an area corresponding to each of all the blocks divided by the blocking means 1. Will be added to. In this way, the image area discriminating means 7 discriminates the image type and its position by referring to the addition result by the block adding means 5, and the encoding means 8 is based on the discrimination information by the image area discriminating means 7. To encode the image data.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Before explaining the embodiments of the present invention, the principle will be described with reference to FIG. 1 in order to avoid misunderstanding of the present invention.

FIG. 1 is a conceptual diagram showing the principle of the present invention.
As shown in the figure, the image coding apparatus of the present invention comprises a blocking means 1, a detection range setting means 2, a detection range scanning means 3, an image type judging means 4, a corresponding block adding means 5, a storage means 6, It is composed of an image area discrimination means 7 and an encoding means 8. The blocking means 1 divides the input digital image signal into blocks of M × N (M and N are natural numbers) pixels.

The detection range setting means 2 sets the detection range 22 with a plurality of blocks 23 in the input image 21 as shown in the example of FIG. 2 in units of blocks divided by the blocking means 1. The detection range scanning means 3 moves the detection range set by the detection setting means 2 in units of one block as shown in FIG.

That is, the detection range 22 moved to the right by one block from the state shown in the input image 31 is shown in the input image 32, and the detection range 22 is further shown in the input image 3.
The input image 33 is moved downward by one block from the state shown by 2.

The image type determination means 4 determines whether or not the image within the detection range is the image type to be determined every time the detection range is moved by the detection range scanning means 3. The storage unit 6 has an area in which numerical values can be added corresponding to the positions of all the blocks divided by the blocking unit 1.

Each time the image type judging means 4 finishes the judging operation, the corresponding block adding means 5 determines the area of the storage unit 6 corresponding to all the blocks in the corresponding detection range in accordance with the result. Count up the value. The image area discrimination means 7 discriminates the image type and its position to be judged from the result of addition to the storage unit 6 at the end of all scanning. The encoding means 8 is the image area discrimination means 7
Appropriate encoding is performed on the image data based on the determination result by the.

In the image coding apparatus having such a configuration, the input digital image signal is first divided into blocks of M × N pixels by the blocking means 1 and a plurality of blocks set by the detection range setting means 2. Is set as the detection range. Then, the detection range scanning means 3 sequentially moves the detection range in units of one block for the entire area of the input image information.

Each time the detection range is moved, the image type determination means 4 determines the type of the image within the detection range. If this is the image type to be determined, 1 is applied, otherwise 0 is applied. The block addition means 5 stores the addition result in the storage areas corresponding to all the blocks in the corresponding detection range of the storage unit 6.

In this way, after the above operation is performed for all areas of the input image information, the addition result of all areas of the storage unit 6 is referred to, and the image type and position of the image to be determined by the image area determination unit 7 are determined. And the encoding means 8 encodes the image data. An embodiment of the image coding apparatus of the present invention will be described below with reference to FIGS. 4 to 9. FIG. 4 is a block diagram showing the arrangement of the image coding apparatus according to the embodiment.

As shown in the figure, the image coding apparatus according to the present embodiment includes a blocking unit 41 and a one-frame image storage unit 4.
2, a detection range setting unit 43, a binary image discrimination unit 44, an addition unit 45, an addition information storage unit 46, a discrimination unit 47, a discrimination information storage unit 48, a CRT 49, and an encoding unit 50.

In the image coding apparatus having such a structure, when digital image data is input, the blocking unit 41 converts the digital image data into M × N pixels (M,
N is a natural number, for example 8 × 8 pixels),
The information of the image divided into the blocks is stored in the one-frame image storage unit 42.

Then, the detection range setting section 43 has a detection range of an appropriate size (in this embodiment, in the block image information stored in the one-frame image storage section 42).
For example, 2 × 2 blocks are set). After the above setting, the binary image determination unit 44 determines that the image in each block within the detection range given by the detection range setting unit 43 is 2
It is determined whether the image is a value image.

Then, the adding section 45 gives a numerical value 0 if the image within the detection range is a binary image, and a numerical value 1 otherwise.
Are sequentially added, and the addition information is sequentially stored in the storage area of the addition information storage unit 46 corresponding to each block in the detection range. The storage area of the addition information storage unit 46 corresponds to the image information of the blocked image stored in the one-frame image storage unit 42.

Next, the detection range setting section 43 moves the detection range in units of one block, and the binary image discrimination section 44 makes the above-mentioned discrimination each time, and the discrimination result is added information. When the addition unit 45 sequentially sends the addition information storage unit 46, the addition information storage unit 4
6 stores the addition information.

When such detection is performed for the entire area of the input image information, the discriminating unit 47 discriminates the image type and its position in each block, and the discriminating information storage unit 48 uses the discrimination result of the discriminating unit 47. The discrimination information regarding a certain image type and its position is stored.

In this way, the encoding unit 50 optimally encodes image data based on the image type and the position information stored in the discrimination information storage unit 48. Further, the information stored in the discrimination information storage section 48 can be displayed by the CRT 49. next,
FIG. 5 shows details of the configuration of the binary image discrimination unit 12.

As shown in the figure, the binary image discrimination section 4
Reference numeral 4 denotes a block scanning unit 440, a density determination unit 441, an addition unit 442, a switching unit 443, a counter 444, and a detection unit 4.
45, a comparison unit 446, a writing unit 447, and a storage unit 448.

In the binary image discriminating unit 44 having such a configuration, the adding unit 442 synchronizes with the block scanning unit 440 every time the block scanning unit 440 scans each block within the set detection range. Then, the density value of each block is added.

At the same time, the density discriminating unit 441 discriminates the density in the block and sends the information to the switching unit 443. The switching unit 443 receives the information regarding this concentration,
The count destination is switched to the counter 444 (counter 0 to counter Z) corresponding to the density. Then, the detection unit 44
In No. 5, a histogram is formed by adding the density values of all the blocks within the detection range, and the number of density peaks is detected from the histogram.

In this way, the comparing unit 446 determines whether the input image is a binary image based on the number of peaks, and the writing unit 4
In accordance with the result of the determination, the storage unit 448 writes 0 if the image is a binary image and 1 otherwise. Below, FIG.
Through reference to FIG. 9, a specific operation performed by the image coding apparatus having the above-described configuration will be described. FIG. 6 shows a part of a state in which the input image captured is divided into blocks by the blocking unit 41 and stored in the one-frame image storage unit 42.

According to the present embodiment, on the block of the input image, the rectangular range of the block from n to n + 8 in the x direction and the block from m to m + 6 in the y direction is defined as a binary image, and the others are imaged. I have an image. Then, the binary image area of the image is detected.

Then, the detection range is fixed to 2 × 2 blocks (plural blocks) as shown in FIG. 7A, and the detection range setting unit 43 uses the detection range scanning method as shown in FIG. 7B. The detection range is set so that the entire area of the image information is scanned in block units, and the binary image discrimination unit 44 determines for each setting whether or not the image is a binary image. 0 is added, and otherwise 1 is added to all blocks corresponding to the detection range of the addition information storage unit 46 by using the addition unit 45. In this addition, scanning of the detection range is performed for all image areas. Will end later.

When the image data as shown in FIG. 6 is given, the addition information storage section 46 is given the addition result as shown in FIG. Detect areas that are not uniform. That is,
In this embodiment, "4" is uniform in FIG.
The other areas are determined as areas where the addition result is not uniform. Thus, the discrimination result as shown in FIG. 9 is obtained. Then, the discrimination result described above is converted into the format of the data example 10 as shown in FIG. 10 by the discrimination unit 47 and written in the discrimination information storage unit 48.

Optimal encoding is performed on the image data based on the image type and start position information written in the discrimination information storage section 48. Further, the information stored in the discrimination information storage section 48 can be displayed by the CRT 49.

In the data string shown in FIG. 10, 1
00 is a heading start code, 101 is data header data, 102 is the total data length, 103 is the number of image types (two types in this embodiment) included in the main image information, 104 is a data unit start code, and 105A and 105B are Image type (binary image and image image in this embodiment), 106 is block information indicating how many partial images of the corresponding image type are, 107 is coordinate information indicating each area, 1
Reference numeral 08 indicates a data unit end code, and 109 indicates a text end code.

As described above, in the present embodiment, particularly in the binary image discriminating section 44, the occurrence frequency distribution of the quantization level of the image is obtained, and it is discriminated whether or not the image is a binary image based on the peak generation pattern. is doing.

In addition to this, there is also a method of determining a binary image when the high frequency component is large due to the frequency distribution of the DCT coefficient, and the base color or luminance level is stored and the level at which the peak of the frequency distribution occurs. By comparison with
It is also possible to adopt a method of determining whether the image data is binary image data.

Further, in the case where the above detection is performed to determine whether to perform DCT encoding as a compression method or to perform binary compression such as modified Huffman, it is particularly determined whether or not. When there is an uncomfortable area, it is optimal to perform DCT encoding, and to the portion that is surely binarized, the compression processing by the above binary compression method is performed.

Although one embodiment of the present invention has been described above, the present invention is not limited to this, and it is needless to say that other changes and modifications can be made. For example, it is also possible to allow the user to confirm the result of the automatic determination by display and correct it.

[0039]

According to the present invention, the size of the input image is reduced for the purpose of detecting the screen or the like by compressing with the optimum compression method or quantization bit for the image type of the whole image or the partial image. By detecting the image type information of the input image and the position information attached to it, which enables optimization such as the case, it is possible to encode the image data by an optimum method. An image encoding device can be provided.

[Brief description of drawings]

FIG. 1 is a conceptual diagram showing the principle of an image encoding device of the present invention.

FIG. 2 is a diagram showing a detection range set by a plurality of blocks in units of blocks divided by the image encoding device of the present invention.

FIG. 3 is one of detection ranges in the image encoding device of the present invention.
It is a figure for demonstrating movement per block.

FIG. 4 is a block diagram showing a configuration of an image encoding device according to an embodiment.

FIG. 5 is a diagram showing details of a configuration of a binary image discrimination unit 12 in the embodiment.

FIG. 6 is a diagram showing a part of a state in which an input image captured by the embodiment is blocked by a blocking unit 41 and stored in a 1-frame image storage unit 42.

7A and 7B are diagrams for explaining a detection range of the embodiment, FIG. 7A is a 2 × 2 block (plural blocks) showing the detection range, and FIG. 7B is a view showing a detection range scanning method by the detection range setting unit. Is.

FIG. 8 is a diagram showing an addition result stored in an addition information storage unit 46 in the embodiment.

FIG. 9 is a diagram showing a discrimination result by a discrimination unit 47 of the image encoding device according to the embodiment.

FIG. 10 is a diagram showing an example of data converted into a format to be written in the discrimination information storage unit 48 of the embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Blocking means, 2 ... Detection range setting means, 3 ... Detection range scanning means, 4 ... Image type determination means, 5 ... Applicable block addition means, 6 ... Storage means, 7 ... Image area determination means, 8
... encoding means.

Claims (1)

[Claims] 1. An input digital image signal is M × N.
(M and N are natural numbers) Blocking means for dividing into blocks of pixels, detection range setting means for setting the detection range with a plurality of blocks divided by the blocking means, and setting by the detection range setting means The detection range
Within the detection range, each time the detection range is moved by the detection range scanning means, the detection range is moved with respect to the entire area of the image information by using the block divided by the blocking means as a movement unit. Image type determining means for determining the image type of, a storage means having an area corresponding to each of all blocks divided by the blocking means, and each time the image type determining means completes the determination operation,
A corresponding block addition means for adding the result to the storage means, an image area determination means for determining the image type and its position by referring to the addition result by the corresponding block addition means, and the image area An image encoding device comprising: an encoding unit that encodes image data based on the discrimination information by the discrimination unit.