JPH04326666A - Picture encoder - Google Patents

Abstract

PURPOSE:To improve the compression efficiency and to save the memory capacity. CONSTITUTION:Picture information in which a character and a line drawing and a multilevel picture such as a photograph are in existence in mixture is inputted respectively to an extraction circuit 21 and a discrimination circuit 22 to use correlation of text pictures between adjacent blocks and a gradation data extracted by a preceding block is fed back to the discrimination circuit 22. The discrimination circuit 22 takes preceding of the gradation data on others to output a discrimination signal 26, which is used to select a text data extracted from a current block by the extract circuit 21 and the discrimination circuit 22.

Description

[Detailed description of the invention]

0001

[Field of Industrial Application] The present invention relates to an image encoding device.
In particular, the present invention relates to an image encoding device that encodes image information that includes a mixture of text and line images and natural images such as photographs.

0002

2. Description of the Related Art Generally, high resolution is required to process text images such as characters and line drawings, while high gradation is required to process images such as natural pictures. Therefore,
In order to process images that are a mixture of text images and image images, a huge amount of data is required to store both the number of pixels sufficient to ensure high resolution of the text and the number of gradations to maintain the high gradation of the image. There was a problem in that a storage device with a large memory capacity was required.

[0003] One possible method to reduce memory capacity is to compress data and store it in memory. However, if an image containing a mixture of text images and image images is compressed as is, the output image after processing will be Mosquito noise occurs around text parts, so the system is equipped with an image storage unit that separately stores images that require high resolution, such as text, and images that require gradation, such as natural images. An image encoding device has been proposed by the present applicant.

0004

However, in the conventional example described above, for example, the process of extracting text data (data that requires high resolution) from within a certain block is performed independently for each block. Therefore, depending on the block, even within the area where text data exists,
In some cases, the threshold for extraction as text data may not be reached, resulting in blocks from which text data is not extracted. Furthermore, since the gradation data indicating the color of pixels extracted as text data is transferred independently for each block, there is a drawback that the compression efficiency of the gradation data becomes worse.

The present invention has been made to solve the above-mentioned problems, and an object thereof is to provide an image encoding device that can improve compression efficiency and save memory.

[0006]

[Means to solve the problem] and

[Operation] In order to achieve the above object, the image encoding apparatus of the present invention has the following configuration. That is, an image encoding device that encodes image information in which a multivalued image and line drawings are mixed, comprising an extraction means for extracting line drawing information from the image information in which a multivalued image and line drawings are mixed; The apparatus includes a determining means for determining the correlation between the extracted line drawing information, and a storage means for storing the line drawing information extracted by the extracting means and the determination result by the determining means.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 2 is a block diagram showing the configuration of the image encoding device in this embodiment. In the figure, 1 is an input terminal,
2 is a data identification circuit; 3, 4, and 5 are compression circuits, respectively;
6 is a resolution memory, 7 is a gradation memory, 8 is an image memory, 9, 10, and 11 are decompression circuits, 12 is a selector, and 14 is an output terminal. Here, input terminal 1
Image data divided into blocks of a predetermined size (for example, 8 x 8 pixel blocks) is input from a host computer, etc., and the data identification circuit 2 analyzes the data content and identifies the text image (including colored characters) in the block. ) is extracted. The extracted text image data is stored as bitmap data (resolution information for each pixel) and gradation data (color data that indicates the color of the text), and pixels that are not extracted within the block are stored as image data [full color data for each pixel]. (8-bit data for each color of R, G, B)] are stored in separate memories.

[0009] Here, a method for extracting a text image in the data identification circuit 2 will be described. First, a histogram is generated based on the density values (colors) of pixels in one block, and the largest density value of the histogram is selected from among the density values that are equal to or higher than a predetermined threshold value. That is, all pixels having the highest density value in a block are extracted as representative density values (representative colors) of the text image of that block. Therefore, pixels having the same density value are extracted, and images with generally constant density, such as text images, are extracted. Further, if there is no density having a value equal to or greater than the threshold value, it is assumed that no text image exists and no text image data is extracted.

Next, the bitmap data identified by the data identification circuit 2 described above is input to a compression circuit 3, the gradation data to a compression circuit 4, and the image data to a compression circuit 5. In order to maintain high resolution, bitmap data is compressed using reversible encoding, and a resolution memory 6 that stores one bit per pixel.
sent to. In addition, the gradation data represents the gradation (color) of the extracted text, and when using full color (256 gradation) pixels for the input image, red (R), which is the three primary colors,
, green (G), and blue (B), and each block requires 8 bits of data for each plane, totaling 24 bits of data. Storing it in memory as it is requires a considerable amount of memory, so generally the compression circuit 4
It is compressed using , and then stored. The compression method is based on the fact that the gradation (color) data of a text image has a correlation between adjacent blocks, and the same value often appears continuously in the gradation (color) data of the text extracted for each block. Therefore, a method is used in which data is compressed by outputting gradation data only when the gradation data of the previous block and the gradation data of the current processing block are different. At this time, a total of 9 bits of data including 8-bit gradation data and a 1-bit header as shown in FIG. 3(a) is input to the input of the compression circuit 4, and the value is
When it is "1", it is the input data as it is, and when it is "0", it is the header information (1 bit) as shown in Figure 3 (b).
Output only the data and compress the data. The reason why 8 bits are used instead of 24 bits (full color) is because the deterioration of character colors is not so noticeable even with 256 colors.

The image data is compressed by the compression circuit 5. Here, in order to increase the compression ratio, for example, ADCT (A
adaptive Discrete Cosine T
The image data is compressed using irreversible encoding such as (transform) and stored in the image memory 8. The data stored in each of the memories 6 to 8 mentioned above is sent to an expansion circuit 9.
, 10, and 11, respectively, and output as output data. Then, bitmap data output from the resolution memory 6 and expanded by the expansion circuit 9 is supplied to the control terminal of the selector 12 from the signal line 13, and when the value of the bitmap data is "1", the terminal a Data of,
That is, the gradation data of the text is selected, and when it is "0", the data at terminal b, that is, the image data, is selected and supplied to the output terminal 14 through terminal c.

As described above, when storing an image in which a text image of characters or line drawings and an image image such as a natural image are mixed, the image is compressed by an image encoding device equipped with a plurality of compression circuits as described above, and then stored. is configured to do so.

FIG. 1 is a diagram showing the configuration of a data identification circuit in this embodiment. In the figure, 20 is an input terminal, 21 is an extraction circuit, 22 is a determination circuit, 23, 24, and 25 are selectors, 31 is a header information addition circuit, and 28, 29, and 30 are output terminals. The entire circuit shown in FIG. 1 is shown in FIG.
This corresponds to the data identification circuit 2 shown in FIG.

The data input from the input terminal 20 is 1
Data is input in block units and is supplied to the extraction circuit 21 and the determination circuit 22, respectively. The text image extraction method in the extraction circuit 21 is performed using the histogram of the density values of the pixels described above. The bitmap data, image data, and gradation data of the text image extracted by this extraction circuit 21 are sent to each a of selectors 23, 24, and 25.
are supplied to each terminal. On the other hand, in the determination circuit 22,
A text image is extracted from the input block-by-block data in the same manner as in the extraction circuit 21, and the extracted bitmap data, image data, and gradation data are supplied to terminals b of selectors 23, 24, and 25. Ru. Also,
The gradation data extracted from the block input in the previous step (previous block) via the signal line 27 is fed back and input to the determination circuit 22 described above. This determination circuit 22 uses the feedback gradation data to
When extracting a text image from an input block,
Pixels having gradation data having the same density value as the returned gradation data are preferentially extracted.

Specifically, pixels that take gradation data are extracted from the histogram shown in FIG. 4 by increasing the value of the histogram of the gradation data of the previous block fed back by a predetermined value. Make it easy. here,
The amount of increase in the histogram can be set arbitrarily. However, if there is no pixel that takes the gradation data of the previous block at this time, the histogram is not increased.

Therefore, in FIG. 4, when the tone data of the previous block is A, the mode value of the histogram only in the current block is B, but only the value corresponding to the shaded area is present in the density A part. By increasing the value of the histogram, the tone data extracted in the current block becomes A.

According to the above-described method, if the gray scale data extracted by the determination circuit 22 is equal to the gray scale data of the previous block, the 1-bit determination data is transmitted to the value "" through the signal line 26 shown in FIG. If they are different, the determination data is output as a value "1". The determination data is then input to the selectors 23, 24, 25 and the header information addition circuit 31. Here, the selectors 23, 24, and 25 select if the value of the judgment data is "0",
Control terminal c is connected to b side, and the value of judgment data is “
1'', the control terminals c of the selectors 23, 24, 25 are connected to the a side. Therefore, according to the above judgment data, when the gradation data of the previous block is also extracted in the current block, the data output from the judgment circuit 22 is selected, and when the gradation data of the previous block is not extracted, the extraction circuit selects the data output from the judgment circuit 22. The data output from 21 is selected. Then, the gradation data selected by the selector 25 is input to the header information adding circuit 31, and the value of the judgment data input at the same time is used as header information, and the gradation data (8 judgment data (1 bit) is added to the bit), and a total of 9 bits of data is output.

In this way, each data output from the selectors 23, 24 and the header information addition circuit 31 is supplied to the output terminals 28, 29, 30, and output from the data identification circuit 2 shown in FIG.

As explained above, according to this embodiment,
In the process of extracting gradation (color) data in the current block by using the correlation of text images between adjacent blocks, the gradation data extracted in the previous block and returned is prioritized. , the compression effect is enhanced and the gradation memory can be saved.

[0020]

Other Embodiments Next, other embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 5 is a diagram showing the configuration of a data identification circuit in another embodiment. In the figure, 40 is an input terminal;
1 is a determination circuit, 42 is an extraction circuit, 43 is a header information addition circuit, and 46, 47, and 48 are output terminals, respectively. Note that only the points different from the embodiments described above will be explained.

Input data in units of blocks inputted from the input terminal 40 is inputted to a determination circuit 41 . Further, the extraction result of the gradation data in the extraction circuit 42 of the previous block is fed back via the signal line 44 and input to the determination circuit 41 . This determination circuit 41 uses the same method as in the embodiment described above to increase the value of the histogram so as to give priority to the gradation data of the previous block. Further, based on the increase, it is determined whether gradation data can be extracted, and the determination result is outputted to the signal line 45 as 1-bit data. That is, in the histogram, it is determined whether the gradation data portion of the previous block is a mode value. Furthermore, when the determination data is output, the data input to the determination circuit 41 is output as is and input to the extraction circuit 42 . The data input to the extraction circuit 42 is extracted in the same manner as in the previous embodiment, and bitmap data and image data are supplied to output terminals 46 and 47, respectively. Further, the gradation data outputted by the extraction circuit 22 is inputted to the header information addition circuit 43. The operation of the header information adding circuit is the same as that in the embodiment described above, and a description thereof will be omitted.

As explained above, according to this embodiment, priority is given to the gradation data of the previous block in the operation of the determination circuit, and as a result, it is only determined whether the gradation data of the previous block can be extracted. The actual extraction process depends only on the extraction circuit, which simplifies the circuit configuration.

The present invention may be applied to an image processing system composed of a plurality of devices such as an image scanner, a printer interface, and a printer, or to an apparatus composed of a single device. Furthermore, a system to which multiple printers can be connected may be used. It goes without saying that the present invention can also be applied to a case where the present invention is achieved by supplying a program to a system or device.

The data inputted from the input terminal 1 may be, for example, commands in a page description language such as Postscript. In this case, the data identification circuit 2 performs command analysis to output bitmap data, image data, and gradation data. In addition, the input data is image data for each pixel, and there are characters (
(including color characters), the bitmap data and image data may be distinguished by, for example, a known image area separation technique.

Further, a hard copy device such as a laser beam printer or an inkjet printer, or a soft copy device such as a display can be connected to the output terminal 14. In addition, as a compression algorithm, DC
T (Discrete Cosine Transform
In addition to the above-mentioned ADCT using rm), other compression methods can be used, such as vector quantization.

Furthermore, the resolution memory 6 may store raw bitmap data for each pixel instead of compressed data.

[0028]

[Effects of the Invention] As explained above, according to the present invention,
It is possible to improve compression efficiency and save memory.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the configuration of a data identification circuit in this embodiment.

FIG. 2 is a block diagram showing the configuration of an image encoding device in this embodiment.

FIG. 3 is a diagram showing the structure of data output from a header information addition circuit.

FIG. 4 is a diagram showing an increase in the value of a histogram in a determination circuit.

FIG. 5 is a diagram showing the configuration of a data identification circuit in another embodiment.

[Explanation of symbols]

21 Extraction circuit 22 Judgment circuit 31 Header information addition circuit

Claims (1)

[Claims] 1. An image encoding device for encoding image information in which a multi-valued image and line drawings are mixed, comprising an extraction means for extracting line drawing information from image information in which a multi-valued image and line drawings are mixed; An image characterized by comprising a determining means for determining the correlation of the line drawing information extracted by the extracting means, and a storage means for storing the line drawing information extracted by the extracting means and the determination result by the determining means. Encoding device.