JP2002077626A - Apparatus and method for image processing and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform a suitable encoding in response to kinds of images. SOLUTION: In an apparatus for image processing comprising a dividing means to divide image data into a plurality of blocks, a conversion means to subject the image data to a discrete cosine transform by the blocks and a quantizing means to quantize the transformed image data, if the image data is an image data which is obtained by rendering a page description language, the quantizing means selects quantization parameters for quantization based on an image area flag data attached to the image data, and if the image data is obtained by an image reader, the quantizing means selects the parameters to quantize the image data based on an encoding amount of the blocks which are previously quantized.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing technique, and more particularly to an image coding technique corresponding to an image type.

[0002]

2. Description of the Related Art Conventionally, compression methods for color still images include:
JPEG method using discrete cosine transform, Wavelet
A method using conversion is often used. In this type of encoding system, basically, the entire image is uniformly processed. Therefore, when the encoding system is changed for each image partial region, image region flag information for identifying the region is separately provided. Is used.

As a mode of changing the encoding method by dividing the image area, for example, a character area and a photographic area are divided into areas, and the quantization of the photographic area where the image quality deterioration is relatively inconspicuous is coarsened.
The quantization of character areas where image quality degradation is relatively conspicuous is finer. There is a mode in which quantization of a computer graphics area that is relatively conspicuous is made fine.

When the total code amount predicted from the generated code amount monitor value exceeds the target code amount,
There is a mode in which the quantization step after the determination is made coarse.

[0005]

However, even if the coding method is changed, the purpose is different, so that the coding method is often implemented separately, and the two methods are not consistent.

Accordingly, a main object of the present invention is to provide an image processing apparatus capable of executing appropriate encoding according to the type of an image,
An object of the present invention is to provide an image processing method and a recording medium.

[0007]

According to the present invention, means for dividing image data into a plurality of blocks, means for performing discrete cosine transform of the image data for each of the blocks, and means for performing discrete cosine transform of the image data When the image data is image data obtained by rendering a page description language, an image area flag attached to the image data. A quantization parameter is selected and quantized based on data, and when the image data is image data obtained by an image reading device, a quantization parameter is determined based on the code amount of the block that has been previously quantized. Is selected and quantized.

Further, according to the present invention, a step of dividing image data into a plurality of blocks, a step of discretely cosine transforming the image data for each block, and a step of quantizing the discrete cosine transformed image data A quantization step;
In the quantization step, when the image data is image data obtained by rendering a page description language, a quantization parameter is selected based on image area flag data attached to the image data. When the image data is image data obtained by an image reading device, a quantization parameter is selected and quantized based on the code amount of the block that has been previously quantized. A featured image processing method is provided.

According to the present invention, a computer is
Means for dividing the image data into a plurality of blocks, means for discrete cosine transform of the image data for each block,
Quantizing means for quantizing the discrete cosine-transformed image data, a recording medium recording a program to function as, the quantization means, the image data,
If the image data is obtained by rendering a page description language, a quantization parameter is selected and quantized based on the image area flag data attached to the image data, and the image data is processed by an image reading device. If the image data is obtained, a recording medium is provided, wherein a quantization parameter is selected and quantized based on the code amount of the block previously quantized.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of an image processing apparatus for digital images according to an embodiment of the present invention. In FIG. 1, 100 is an image input unit, 101 is an encoding unit including a block line buffer for compression, 102 is an encoding unit for image area flag data, 103 is a quantization / encoding unit, 104 is a compression memory, and 105 is a compression memory. Large-capacity external storage device, 106 is a compression memory, 107 is an inverse quantization / decoding unit, 126 is a decoding unit for image area flag data, 109 is a decoding unit, 110
Is an output unit, which is an I / F with an image generation unit such as a printer (not shown).

The input unit 100 is connected via a selector 122 to an image scanner unit 120 which is an image reading device such as a scanner, and a page description language rendering unit 121 such as a printer. One of the image data is input by the switching. Switching of the selector 122 is based on a switching signal 1000 from the outside. Hereinafter, image data from the image scanner unit 120 is referred to as scanner image data, and image data from the page description language rendering unit 121 is referred to as PDL image data.

The input image data is, for example, in the case of a color signal, three colors of red (R), green (G) and blue (B).
This is a signal of 256 gradations.

The encoding unit 101 divides an image into tiles,
For example, discrete cosine transform (DCT) coding (for example, according to JPEG) is performed on image data for each tile of M × M pixels. Also, the image area flag data is separated and sent to the image area flag data encoding / conversion unit 102.

Image area flag data encoding / conversion section 102
Performs run-length encoding on the image area flag data added to the image data, and in a certain case, changes the image area flag data before encoding.

The image area flag data is attached to each pixel. However, as in the present embodiment, for the data processed by DCT for each 8 × 8 block, for each block,
The image area flag data is used as a representative.

The quantization / encoding unit 103 includes the encoding unit 10
In step 1, the DCT-transformed image data is quantized, and run-length encoding is performed.

The inverse quantization / decoding section 107 performs inverse quantization and decoding corresponding to the quantization / encoding section 103. The image area flag decoding unit 126 performs decoding corresponding to the encoding in the image area flag encoding / changing unit 102. Decoding section 109 performs decoding such as inverse DCT transform corresponding to encoding section 101.

The switching signal 1000 is supplied to the selector 122 and the quantization / encoding unit 103, and is used as a reference for selecting image data and selecting a quantization parameter as described later.

FIG. 2 is a diagram illustrating a part of each block shown in FIG. 1 in detail.

The encoding unit 101 has a color conversion unit 11 and a DCT
The color conversion unit 11 including the unit 12 converts the RGB signal into a luminance / color difference signal (YCbCr). The DCT unit 12 performs spatial frequency transform (discrete cosine transform:
DCT transformation).

The quantization / encoding unit 103 includes a quantization unit 13
, A VLC unit 14, a quantization switching unit 203, and a code amount monitoring unit 202.

The quantization unit 13 reduces the data amount by quantizing the DCT coefficients.

The VLC unit 14 performs variable length coding,
The data is further reduced by the Huffman encoding process of the quantized value.

The code amount monitoring unit 202 monitors the generated code amount and transmits the generated code amount to the quantization switching unit 203.

The quantization switching unit 203 outputs a switching signal 1000
Follow the steps below.

First, when the switching signal 1000 indicates that the image data to be processed is PDL image data, the quantization parameter is switched according to the image area flag data from the encoding unit 101.

When the image data to be processed is designated as the scanner image data by the switching signal 1000, it is handled by the rate control, and the code amount monitoring unit 2
Focusing on the amount of generated code from S.02, if the amount of code is expected to exceed a predetermined set value, the quantization unit 13
Is instructed to switch the quantization parameter.

According to this instruction, the quantization of the image data of the subsequent block is roughly performed. If the code amount is expected to be within the range of the set value, the quantization parameter returns to the original value.

Further, the quantization switching unit 203 sends the corresponding image area flag data to the image area flag encoding / change unit 102 in order to clarify the switching point of the quantization parameter. An instruction to change the data to the data corresponding to the parameter is issued, and the encoding / change unit 102 of the image area flag executes this.

In this embodiment, the switching of the quantization parameter is performed, for example, by switching between two steps as the quantization step. For example, the quantization step is performed by Q1 <Q2.
Accordingly, the code amount is L1> L2, and the image quality is worse in the latter case.

FIG. 3 shows an example of area division. In the case of PDL image data, small and large quantization steps can be made to correspond to a character area and a photograph area, respectively. Further, in the case of the scanner image data, it indicates that the region is divided into a region having a relatively small quantization step before rate control and a region having a relatively large quantization step after rate control.

FIG. 4 shows a tile image of 32 × 32 pixels.
FIG. 4 is an explanatory diagram showing that the image is divided into blocks for 8 × 8 DCT and divided into a region having a fine quantization step (white) and a region having a large quantization step (hatching). It is the role of the image area flag data to identify each area.

FIG. 6 shows an example of a basic quantization matrix as a quantization parameter for 8 × 8 DCT coefficients. For the matrix after the change, the elements of the basic matrix are Q
Multiply and divide by 50. By changing the value of Q, the quantization step can be easily changed.

FIG. 5 shows the encoding / changing unit 10 for the image area flag.
FIG. 3 is a diagram showing, in particular, blocks related to encoding among blocks constituting the second embodiment.

The determination unit 310 determines whether the data is the same as the previous value flag data, and switches the data to the RL unit if the data is the same, and switches to the LT unit if the data is different.

The RL section 311 counts the number of times the same as the previous value data until different data comes out, and finally outputs the repeated data.

The LT unit 312 counts the number of cases where the code is different from the previous value, and outputs the code word corresponding to the count and the minimum number of constituent bits of the actual data for the count.

The combining section 313 combines the output data of the RL section and the output data of the LT section and outputs the result as a code 315.

Next, the inverse quantization / decoding unit 107 determines whether the VL
D unit 16, inverse quantization unit 17, quantization switching unit 207
And

The VLD unit 16 performs Huffman decoding by a variable length decoder (VLD) corresponding to the VLC unit 14.
The inverse quantization unit 17 performs inverse quantization corresponding to the quantization unit 13 and restores a DCT coefficient.

The quantization switching unit 207 performs the decoding corresponding to the encoding in the encoding / conversion unit 102 of the image area flag based on the image area flag data decoded by the image area flag decoding unit 126. The inverse quantization is executed by switching the quantization matrix for the inverse quantization.

The decoding section 109 performs the DCT of the DCT section 12
It includes an IDCT unit 18 that performs DCT inverse transform corresponding to the transform and returns the luminance / color difference signal to a luminance / color difference signal, and a color conversion unit 19 that returns the luminance / color difference signal to an RGB signal.

Although the preferred embodiment of the present invention has been described above, in the case of PDL image data, a plurality of image quality controls for controlling a plurality of image quality under the rate control by the code amount monitoring unit 202 and the quantization switching unit 203 are described. A configuration in which the quantization step group is switched may be employed. FIG.
The parameter Q1 = 50 in the region where the quantization step is small before the code amount control, and the parameter Q2 = 75 in the region where the quantization step is large is the parameter Q1 = 60 in the region where the quantization step is small after the code amount control, and the quantization step is large. It is shown that by changing the area parameter Q2 to 100, it is possible to control the image quality for each area while controlling the code amount.

An object of the present invention is to supply a storage medium (or a recording medium) in which program codes of software for realizing the functions of the above-described embodiments are recorded to a system or an apparatus, and to provide a computer (a computer) of the system or the apparatus. Or a CPU or MPU) reads out and executes the program code stored in the storage medium,
Needless to say, this is achieved. In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.
In addition, by the computer executing the readout program code, not only the functions of the above-described embodiments are realized, but also based on the instructions of the program code,
The operating system (OS) running on the computer performs part or all of the actual processing,
It goes without saying that a case where the functions of the above-described embodiments are realized by the processing is also included.

Further, after the program code read from the storage medium is written into the memory provided in the function expansion card inserted into the computer or the function expansion unit connected to the computer, the program code is read based on the instruction of the program code. , The CPU provided in the function expansion card or the function expansion unit performs part or all of the actual processing,
It goes without saying that a case where the functions of the above-described embodiments are realized by the processing is also included.

[0047]

As described above, according to the present invention,
Appropriate encoding can be performed according to the type of image.

[Brief description of the drawings]

FIG. 1 is a block diagram of a digital image processing apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a part of each block shown in FIG. 1 in detail.

FIG. 3 is a diagram showing an example of area division.

FIG. 4 shows a tile image of 32 × 32 pixels converted to an 8 × 8 DC
FIG. 9 is an explanatory diagram showing that the area is divided into blocks for T and divided into an area having a fine quantization step (white) and an area having a large quantization step (hatching).

FIG. 5 is a diagram showing, in particular, blocks related to encoding among blocks constituting an image area flag encoding / changing unit 102;

FIG. 6 is an example of a basic quantization matrix as a quantization parameter for an 8 × 8 DCT coefficient.

FIG. 7 is a diagram illustrating a change in a parameter Q of a quantization matrix during rate control.

Claims (6)

[Claims] 1. A means for dividing image data into a plurality of blocks; a means for discretely cosine transforming the image data for each block; and a quantizing means for quantizing the discrete cosine transformed image data. When the image data is image data obtained by rendering a page description language, the quantization unit selects a quantization parameter based on image area flag data attached to the image data. If the image data is image data obtained by an image reading device, the quantization is performed by selecting a quantization parameter based on the code amount of the block that has been previously quantized. Image processing apparatus. Means for inputting the image data, whether the input image data is image data obtained by rendering a page description language, or image data obtained by an image reading device; Means for receiving a switching signal indicating that the image data is image data obtained by rendering a page description language based on the switching signal. The image processing apparatus according to claim 1, wherein the image processing apparatus determines whether the data is image data obtained by the method. 3. When the image data is image data obtained by rendering a page description language, the quantization unit further performs a quantization based on a code amount of the block that has been previously quantized. The image processing apparatus according to claim 1, wherein quantization is performed by switching a quantization parameter. 4. When the image data is image data obtained by an image reading device, attaching image area flag data to the image data based on a quantization parameter selected in the quantization. An image processing apparatus characterized by the above-mentioned. 5. A step of dividing the image data into a plurality of blocks, a step of performing discrete cosine transform of the image data for each block, and a step of quantizing the discrete cosine transformed image data. In the quantization step, when the image data is image data obtained by rendering a page description language, a quantization parameter is selected based on image area flag data attached to the image data. If the image data is image data obtained by an image reading device, the quantization is performed by selecting a quantization parameter based on the code amount of the block that has been previously quantized. Image processing method. 6. A computer, comprising: means for dividing image data into a plurality of blocks; means for performing discrete cosine transform of the image data for each of the blocks; and quantizing means for quantizing the discrete cosine transformed image data. A recording medium storing a program to be operated, wherein the quantizing means, when the image data is image data obtained by rendering a page description language, an image area flag attached to the image data When the image data is image data obtained by an image reading device, a quantization parameter is selected based on the code amount of the block previously quantized. And quantizing the selected recording medium.