JPH05308529A - Image encoder - Google Patents

Abstract

PURPOSE:To improve the reproduciveness of characters or line pictures, especially, characters or line pictures artificially prepared by computers or the like at the device using a compression process. CONSTITUTION:Image information mixing multilevel images such as natural images, characters or line pictures inputted from an input terminal 1 is divided into the blocks of 8X8 picture elements by a block circuit 2 and stored in a buffer 3, and the histogram of the data of all the 64 picture elements is calculated by a historgram calculation circuit 5. Based on the pixel data which historgram is maximum, binary image data are prepared by an extracting color judge circuit 6, reversibly encoded by a bit map data encoding circuit 7 and stored in a bit map memory 8. On the other hand, the density data calculated by the histogram calculation circuit 5 are reversibly encoded by a color data encoding circuit 10 and stored in a color data memory 11.

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, and more particularly to an image coding apparatus for coding an image having a gradation (color), a photograph, and a mixed image of characters and line drawings having a resolution.

[0002]

2. Description of the Related Art The storage capacity required to store a halftone image such as a photograph in a memory is (number of pixels) × (number of gradation bits) ×
(Number of colors), and a particularly large storage capacity is required to store a high-quality color image. For this reason, various information amount compression methods have been proposed, and the information capacity is compressed and then stored in a memory to reduce the memory capacity.

FIG. 10 is a block diagram of a general compressed image storage system. An image signal in which images, characters, and the like are mixed is input from the input terminal 41. The input image signal is read in the block formation circuit 42 in units of, for example, 8 × 8 blocks, and the discrete cosine transform (DC
T) The circuit 43 performs orthogonal transformation. Next, the orthogonally transformed image signal is input to the quantization circuit 44, and the Q table 4
Quantized by 5. The quantized image signal is variable-length coded by the variable-length coding circuit 46 and then stored in the image memory 47 as a compressed image.

Further, the compressed image stored in the image memory 47 is variable-length decoded by the variable-length decoding circuit 48 in synchronization with an output device synchronizing signal (not shown), and inverse-quantized by the inverse quantization circuit 49. After being converted into a discrete cosine inverse transform circuit 50, the discrete cosine inverse transform circuit 50 performs the discrete cosine inverse transform, and the rasterization circuit 51 rasterizes it, and then outputs the expanded image signal from an output terminal 52 to an output device (not shown) such as a page printer. To be done.

The image compression process such as the above-mentioned block formation and encoding is performed by, for example, JPEG (Joint Photograph).
ic Experts Group) has proposed ADCT (Adaptive Discret) as an international standard for color still image coding.
e Cosine Transform) baseline system.

[0006]

However, in the above-mentioned conventional example, in the case of a character or a line drawing having a resolution, there is a drawback that the resolution or the like is lowered and the quality of the character or the line drawing is deteriorated.

In other words, it is necessary to preserve the quality of characters and line drawings.

The present invention has been made in view of the above-mentioned drawbacks, and in a device using a compression process, the reproducibility of characters and line drawings, particularly characters and line drawings artificially created by, for example, a computer is improved. An object is to provide an image coding device.

[0009]

In order to achieve the above object, the image coding apparatus of the present invention has the following configuration. That is, an image encoding device that encodes an image in which a multi-valued image and a line drawing are mixed, a calculation unit that calculates a histogram of an image in which a multi-valued image and a line drawing are mixed, and a histogram calculated by the calculation unit. Based on the line drawing portion of the image, the extracting means extracts the line drawing portion, and the line drawing portion extracted by the extracting means and the encoding means for performing different encoding on the portion other than the line drawing portion.

Further, preferably, each of the above-mentioned means is performed in block units.

More preferably, the extracting means includes a replacing means for replacing the extracted line drawing portion with other image data.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described in detail below with reference to the drawings. <First Embodiment> FIGS. 1 and 2 are block diagrams showing the arrangement of an image coding apparatus according to the first embodiment. First, the process from input of image information to storage in each memory will be described with reference to FIG.

From the input terminal 1, image information in which a multi-valued image such as a natural image output from a host computer and characters and line drawings are mixed is input. Here, the input image information is divided into blocks of 8 × 8 pixels by the blocking circuit 2 and stored in the buffer 3. Buffer 3
The histogram calculation circuit 5 calculates the histogram of all 64 pixel data of the image information stored in, and the pixel data having the maximum histogram frequency is transmitted to the extraction color determination circuit 6. Due to the processing in the histogram calculation circuit 5, one block of image information is delayed by the scan time in the one block delay circuit 4 and transmitted to the extracted color determination circuit 6. Then, in the extraction color judgment circuit 6, the inputted image information is compared with the pixel data having the maximum frequency of the histogram inputted pixel by pixel, and if they coincide with each other, the corresponding pixel is set to “1” and the bitmap data is set. It is transmitted to the encoding circuit 7. However, if they do not match, the corresponding pixel is set to “0” and transmitted to the bitmap data encoding circuit 7. The binary image data transmitted to the bitmap data encoding circuit 7 is losslessly encoded by the bitmap data encoding circuit 7 and stored in the bitmap memory 8.

The density data calculated by the histogram calculation circuit 5 is transmitted to the extracted color determination circuit 6 and at the same time to the color data encoding circuit 10 so that the color data encoding circuit 10 performs reversible encoding. And stored in the color data memory 11.

On the other hand, in the discrete cosine transform (DCT) circuit 13, the image information read from the buffer 3 is subjected to the discrete cosine transform, and the discrete cosine transformed image information is used in the quantizing circuit 14 in the Q table. Are quantized. Then, the quantized image information is lossy encoded by the image data encoding circuit 16 and stored in the image memory 17 as a compressed image.

Next, the process until the image information stored in each memory is output to the output device will be described with reference to FIG.

Characters stored in the bitmap memory 8
Binary data such as a line drawing is decoded by the bitmap data decoding circuit 9 and transmitted to the multiplexing circuit 21.
Further, the color information stored in the color data memory 11 is decoded by the color data decoding circuit 12, and the multiplexing circuit 2
1 is transmitted. Then, the image information stored in the image memory 17 is decoded by the image data decoding circuit 18, and the decoded image information is decoded by the inverse quantization circuit 1.
Dequantized by 9. Here, the inversely quantized image information is the discrete cosine inverse transform (IDCT) circuit 20.
Be transmitted to. The image information subjected to the discrete cosine inverse transform is transmitted to the multiplexing circuit 21.

Next, in the multiplexing circuit 21, the pixel portion (binary data “1”) such as a character / line drawing transmitted from the bit map data decoding circuit 9 is transmitted from the color data decoding circuit 12. The color information is converted into the color information value, and for example, it becomes 24 bits of image information per pixel. Then, the image information is compared with the image information transmitted from the IDCT circuit 20, and in the pixel portion where the image data from the bitmap data decoding circuit 9 is "1", the color data decoding circuit 1
In the pixel portion in which the image data from 2 is selected and the image data from the bitmap data decoding circuit 9 is “0”,
The image data from the IDCT circuit 20 is selected, and the selected image data are multiplexed and transmitted to the rasterization circuit 22. The image information transmitted to the rasterization circuit 22 is rasterized and then output from an output terminal 23 to an output device (not shown) such as a page printer.

As described above, although not shown in the embodiment, the input image data is full-color data such as R, G, B having a gradation of 8 bits, and R, G, B are divided into blocks. Performed and stored in each buffer. A histogram is calculated for each buffer, that is, for each R, G, B color, and the extracted color is determined. The extracted pixel area and the extracted color (color data) obtained for each of R, G, and B are
The extracted pixel region is multiplexed by a multiplexing circuit (not shown), losslessly encoded as a binary image by the bitmap data encoding circuit 7, and then stored in the bitmap memory 8, and one block of color data is stored. One full color of 24 bits per loss is losslessly encoded by the color data encoding circuit 10 and then stored in the color data memory 11.

The input image data is not limited to R, G, B, but may be L ^* a ^* b ^* or YUV, for example. When the input image data is R, G, B, from RGB to L ^* a ^*
It may be configured to convert into b ^* or YUV.

Further, in the embodiment, the processing is performed for each R, G, B plane, but the present invention is not limited to this.
It is also possible to collectively process RGB for 24 bits.

Next, the calculation processing in the above-mentioned histogram calculation circuit 5 will be described below with reference to the flowcharts shown in FIGS.

In step S1 of FIG. 3, the registers A and B are
₀ , B ₁ , ..., B ₆₃ , C ₀ , C ₁ , ..., C ₆₃ , D in total 1
All 30 are initialized to "0". Here, the register A is a register that temporarily stores the captured pixel data, the register B _n (n = 0 to 63) is a register that stores pixel data of different densities, and the register C
_m (m = 0 to 63) is a register for counting the histogram of the pixel data stored in the register B _n .
The register D is a register that stores the maximum value of the histogram. In the next steps S2 and S3, i =
0 and k = 0, and 8 × 8 in steps S4 and S5.
The i + 1th pixel of the block is fetched and the register A
To store. Then, in steps S6, S7, and S9, it is checked whether or not the pixel data up to the i-th pixel data has the same density as the content of the register A. That is, in step S6, the value of i and the value of k are compared, and if smaller (i> k), the process proceeds to step S7, and it is determined whether the contents of register A and the contents of register B _k are equal. Here, if they are not equal, the process proceeds to step S9, the value of k is incremented by 1, the process returns to step S6, and the above process is repeated. However, if they are equal, the process proceeds to step S10, and the contents of the histogram register C _k is incremented by one.

On the other hand, in step S6, the value of i and k
If the values are equal, the process proceeds to step S8, and the i-th
The first pixel data is newly stored. That is, the contents of the register A are stored in the register B _i , and the next step S1
At 0, the contents of the histogram register C _k corresponding to the pixel data are incremented by 1. And step S1
In 1, the value of i, which is the index of the pixel data, is incremented by 1, and in the next step S12, it is checked whether or not the process has been completed for one block (64 pixel data). If not, the process proceeds to step S3. Is returned, and the histogram creation process is repeated. After that, when a histogram for 64 pixels is created (i = 64), the process proceeds to step S13 in FIG. 4 to find the maximum value of the histogram.

First, in step S13, the index
k is initialized to "0", and in steps S14 and S15,
Dista C₀ And B₀ Store the contents of D in A and A respectively
It Then, in step S16, the value of K is increased by 1
Then, in the following step S17, k is compared with “64”,
Check whether all have been processed. As a result, still
If it is in the middle (k <64), the process proceeds to step S18.
Register D and register C_k Compare the contents of the
Ta C_k Is less than or equal to (D ≧ C_k )
The process returns to step S16. But register C_k Is better
If larger (D <C _k ) The process proceeds to step S19,
Register C_k Is stored in register D and the next step
In step S20, the register which is the density of the corresponding pixel data
B_k Is stored in register A, and the process goes to step S16.
Return processing.

Next, when the above process is repeated and the process is completed for the pixel data in one block, step S
The process proceeds from step 17 to step S21, and the content of the register A, that is, the pixel data having the maximum density of the histogram is transmitted to the extraction color determination circuit 6 and the color data encoding circuit 10.

As described above, according to the first embodiment, it is possible to prevent the deterioration of the image quality of characters, line drawings, etc. and efficiently compress the image data. <Second Embodiment> Next, a second embodiment according to the present invention will be described below with reference to the drawings.

FIG. 5 is a block diagram showing the first half configuration of the image coding apparatus according to the second embodiment. The configuration of the latter half is the same as that of FIG. 2, and the blocks having the same functions as those in the first embodiment are designated by the same reference numerals. In this embodiment, the function of replacing the extracted portion of the non-extracted image is added.

6 to 8 show one block, an extracted image and a non-extracted image, FIG. 6 is an original image before extraction, FIG. 7 is an extracted image, and FIG. Is a non-extracted image. Then, the extracted portion (white portion) of FIG. 8 is replaced with another image as shown in FIG.

The process from input of image information to storage in each memory will be described with reference to FIG. In FIG. 5, image information output from a host computer or the like is input from an input terminal 1, and the input image information is divided into blocks by a blocking circuit 2 and stored in a buffer 3. A histogram is calculated by the histogram calculation circuit 5 from the image information stored in the buffer 3, and the pixel data having the maximum histogram is extracted.
To transmit. The one-block delay circuit 4 delays the histogram calculation processing time, and the image information is read from the buffer 3 to the extracted color determination circuit 24. In the extraction color determination circuit 24, the image information input from the buffer 3 is compared pixel by pixel with the pixel data for which the histogram input from the histogram calculation circuit 5 is maximum, and if they match, to the bitmap data encoding circuit 7. Replace "1" with replacement circuit 2
6 is transmitted, and when they do not match, the bit map data encoding circuit 7 is replaced with “0” and the replacement circuit 26
"1" is transmitted to. The binary image data transmitted to the bit map data encoding circuit 7 is stored in the bit map memory 8 through the same process as in the first embodiment.
The pixel data transmitted to the color data encoding circuit 10 is also the first
The color data is stored in the color data memory 11 through the same process as in the above embodiment.

On the other hand, in the extraction color judgment circuit 24, only the time until the extraction color is judged is 2 block delay circuit 2
After being delayed by 5, the image information in the buffer 3 is read into the replacement circuit 26. In the replacement circuit 26, each pixel of the image information from the buffer 3 is compared with the binary data transmitted from the extracted color determination circuit 24, and the pixel whose binary data corresponds to “0” is a replacement data calculation circuit (not shown). Is replaced with a value (average value of non-extracted pixels) calculated in advance, and a pixel corresponding to binary data “1” is not replaced.

Next, the image data replaced as described above is stored in the image memory 17 as a compressed image through the same process as in the first embodiment described above. The process of decoding the image information from each memory and outputting it to the output device is the same as in the first embodiment, and will not be repeated here.

As described above, according to the second embodiment, higher image data compression can be achieved.

In the above-mentioned embodiment, the average value of the non-extracted portion is used as the replacement data, but the present invention is not limited to this. For example, a value that minimizes the mean square error may be used.
Alternatively, it may be replaced with a completely different value using a table.

The present invention may be applied to a system including a plurality of devices such as a reader, an interface, and a printer, or may be applied to an apparatus including a single device such as a copying machine. .. Further, it goes without saying that the present invention can also be applied to the case where it is achieved by supplying a program stored in a floppy disk, an IC card or the like to a system or an apparatus.

[0036]

As described above, according to the present invention,
In a device using a compression process, it is possible to improve the reproducibility of characters and line drawings, especially characters and line drawings artificially created by a computer or the like.

[Brief description of drawings]

FIG. 1 is a block diagram showing an encoding process in a first embodiment.

FIG. 2 is a block diagram showing a decoding process in the first and second embodiments.

[Figure 3]

FIG. 4 is a flowchart showing a histogram process in this embodiment.

FIG. 5 is a block diagram showing an encoding process in the second embodiment.

[Figure 6]

[Figure 7]

[Figure 8]

FIG. 9 is a diagram for explaining an extraction unit and a non-extraction unit.

FIG. 10 is a block diagram showing a configuration of a general image encoding device.

[Explanation of Codes] 1 Input Terminal 2 Blocking Circuit 3 Buffer 4 1 Block Delay Circuit 5 Histogram Calculation Circuit 6 Extraction Color Judgment Circuit 7 Bitmap Data Encoding Circuit 8 Bitmap Memory 9 Bitmap Data Decoding Circuit 10 Color Data Code Coloring circuit 11 Color data memory 12 Color data decoding circuit 13 DCT circuit 14 Quantization circuit 15 Q table 16 Image data coding circuit 17 Image memory 18 Image data decoding circuit 19 Inverse quantization circuit 20 IDCT circuit 21 Multiplexing circuit 22 Rasterization circuit 23 Output terminal

Claims (3)

[Claims] 1. An image coding apparatus for coding an image in which a multi-valued image and a line drawing are mixed, wherein a calculation means for calculating a histogram of an image in which a multi-valued image and a line drawing are mixed, and calculation by the calculation means. An extraction unit for extracting a line drawing portion of the image based on the histogram, and an encoding unit for performing different encoding on the line drawing portion extracted by the extraction unit and a portion other than the line drawing portion. Image encoding device. 2. The image coding apparatus according to claim 1, wherein each of the means performs a block unit. 3. The image coding apparatus according to claim 2, wherein the extraction unit includes a replacement unit that replaces the extracted line drawing portion with other image data.