JP3287707B2 - Image processing apparatus and image processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and an image processing method.

[0002]

2. Description of the Related Art At present, image processing apparatuses include a scanner for inputting image data, a printer for outputting image data, and a digital copying machine for inputting and outputting image data.

This digital copying machine will be described below as a conventional example of an image processing apparatus with reference to FIGS.
As shown in FIG. 5, the digital copying machine 1 includes a scanner 2
And an A / DC (Analog / Digital Converter) 3, an image processing unit 4, and a printer unit 5. The scanner unit 2
Reads image data from a document (not shown), and the A / DC 3 converts analog image data into digital data. The image processing unit 4 performs image processing such as gamma conversion and image quality correction on the image data,
Prints out on image data printing paper (not shown).

[0004] Such a digital copying machine 1 generally stores the fetched image data in a memory so as to repeatedly use the image data and to perform processing such as rotation and inversion of the image. It is possible.
However, since the amount of image data is enormous, simply storing it in a memory requires a large-capacity memory.

In particular, recent digital copiers 1 perform high-quality image processing by treating each pixel data in image data as multi-gradation, but form image data. Each of a large number of pixel data is
If the data is simply converted into multi-gradation density data, the data amount becomes enormous, and an extremely large capacity memory is required.

For this reason, encoding of image data into compressed data to reduce the amount of data and decoding of compressed data into image data have been performed. FIG. 7 shows an algorithm for realizing such encoding and decoding.

When encoding image data into compressed data according to such an algorithm, as shown in FIG.
Each of the multi-gradation pixel data forming the image data is referred to as “4
The data is decomposed into × 4 ”to generate block data, and density data Lij (16 bytes) is calculated from the number of black and white of the block data. The maximum value Lmax of the density data Lij
And the minimum value Lmin, the average data La and the gradation width index Ld
And calculate the average data La and the gradation width index Ld.
Then, quantization thresholds P1 and P2 are calculated from the above. This threshold P
1 and P2, the code φij (16 × 2 bits = 4 bytes) for each pixel is converted to the average data La (1 byte) and the gradation width index Ld (1
Density data Lij (16)
Byte) into compressed data (6 bytes).

[0008] By such encoding, the density data of 16 bytes becomes compressed data of 6 bytes, so that the data amount can be reduced to "3/8".

When the compressed data is decoded into image data, quantized representative values Q1 to Q4 are calculated from the average data La of the compressed data and the gradation width index Ld, and the quantized representative values Q1 to Q4 are calculated. The compressed data is decoded into block data by assigning it to the code φij for each pixel, and image data is reproduced by arranging pixel data corresponding to the density data.

[0010] In the above-mentioned algorithm, "Pi" is a quantization threshold value at the time of encoding, and "Qj" is a representative quantization value at the time of decoding. Generally, the quantization threshold Pi and the quantization representative value Qj are, as shown in the algorithm of FIG. 7, P1 = La + 1 / 2Ld P2 = La-1 / 2Ld Q1 = La + 3 / 4Ld Q2 = La + ／ Ld Q3 = La− ／ Ld Q4 = La− ／ Ld

[0011]

As described above, the image data can be encoded into compressed data by calculating the quantization threshold, and the compressed data can be encoded by calculating the quantization representative value. Data can be decrypted.

However, in the actual digital copying machine 1, the parameters obtained by changing the parameters of the arithmetic expression for calculating the threshold value and the representative quantization value may be changed due to the reading characteristics of the scanner unit 2 and the printing characteristics of the printer unit 5. There are good things.

[0013]

According to the first aspect of the present invention,
Data decomposing means for decomposing each of the multi-gradation pixel data forming the image data into a predetermined number to generate block data; density calculating means for calculating density data of the block data; The average data and the gradation width index are calculated from the minimum data and the minimum value, and the quantization threshold is calculated from the average data and the gradation width index. Data encoding means for encoding density data into compressed data by allocating; threshold switching means for switching a method of calculating a threshold value of the data encoding means; and threshold selection for selecting a calculation method to be switched by the threshold switching means Means were provided.

According to a second aspect of the present invention, a quantized representative value is calculated from the average data of compressed data including a code for each pixel, average data, and a gradation width index and a gradation width index, and is assigned to a code for each pixel. Data decoding means for decoding compressed data into block data, and representative value switching means for switching a method of calculating a quantized representative value of the data decoding means. A representative for selecting a calculation method to be switched by the representative value switching means. Value selection means is provided.

According to a third aspect of the present invention, there is provided a data decomposing means for decomposing each of the multi-tone pixel data forming the image data into a predetermined number to generate block data, and calculating density data of the block data. Providing density calculating means, calculating average data and a gradation width index from the maximum value and the minimum value of the density data, calculating a quantization threshold from the average data and the gradation width index, and calculating the average data by the threshold value. Data encoding means for encoding density data into compressed data by assigning a code for each pixel to the gray scale index and a gradation width index; and threshold switching means for switching a method of calculating a threshold value of the data encoding means. Threshold value selection means for selecting a calculation method to be switched by the switching means is provided, and a quantization representative value is calculated from the average data of the compressed data and the gradation width index, and is divided into a code for each pixel. Data decoding means for decoding the compressed data into block data by applying the data, a representative value switching means for switching a method of calculating a quantized representative value of the data decoding means, and a calculation method to be switched by the representative value switching means is selected. A representative value selecting means is provided.

According to a fourth aspect of the present invention, each of the multi-gradation pixel data forming the image data is decomposed into a predetermined number by data decomposing means to generate block data, and the density data of the block data is calculated. Means, calculate the average data and the gradation width index from the maximum value and the minimum value of the density data, calculate the quantization threshold from the average data and the gradation width index, and calculate the average data using the threshold value. By assigning a code for each pixel to the gray scale index and the gradation width index, the density data is encoded into compressed data by the data encoding means, and the threshold value calculation method in this encoding is made switchable by the threshold value switching means. The calculation method can be freely selected by the threshold value selecting means.

According to a fifth aspect of the present invention, a representative quantization value is calculated from the average data of the compressed data including the code, the average data, and the gradation width index for each pixel and the gradation width index, and is assigned to the code for each pixel. Thus, the compressed data is decoded into block data by the data decoding means, and the method of calculating the quantized representative value in the decoding is made switchable by the representative value switching means, and this switchable calculation method is made selectable by the representative value selecting means. did.

According to a sixth aspect of the present invention, each of the multi-gradation pixel data forming the image data is decomposed into a predetermined number by data decomposing means to generate block data, and the density data of the block data is calculated. Means, calculate the average data and the gradation width index from the maximum value and the minimum value of the density data, calculate the quantization threshold from the average data and the gradation width index, and calculate the average data using the threshold value. By assigning a code for each pixel to the gray scale index and the gradation width index, the density data is encoded into compressed data by the data encoding means, and the threshold value calculation method in this encoding is made switchable by the threshold value switching means. The calculation method is freely selectable by the threshold value selection means, and is calculated from the average data of the compressed data including the code, the average data, and the gradation width index for each pixel and the gradation width index The compressed representative data is decoded into block data by the data decoding means by calculating the child representative value and assigning it to a code for each pixel, and the method of calculating the quantized representative value in this decoding is made switchable by the representative value switching means. The switchable calculation method can be freely selected by the representative value selection means.

[0019]

According to the first and fourth aspects of the present invention, when the method of calculating the threshold value in the encoding of the data encoding means is selected by the threshold value selecting means, the threshold value switching means calculates the threshold value in the encoding of the data encoding means. Since the method is switched, the quantization parameter of the coding can be changed.

According to the second and fifth aspects of the present invention, when the method of calculating the quantized representative value in the decoding by the data decoding means is selected by the representative value selecting means, the quantization in the decoding of the data decoding means is performed by the representative value switching means. Since the calculation method of the representative value is switched, the parameter of quantization in decoding can be changed.

According to the third and sixth aspects of the present invention, when the method of calculating the threshold value in the encoding of the data encoding means is selected by the threshold value selecting means, the threshold value switching means calculates the threshold value in the encoding of the data encoding means. The method is switched,
When the method of calculating the quantized representative value in the decoding of the data decoding means is selected by the representative value selecting means, the method of calculating the quantized representative value in the decoding of the data decoding means is switched by the representative value switching means. The parameters for decoding and quantization can be changed respectively.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIGS. Regarding the digital copying machine 6 shown as an image processing apparatus in the present embodiment, the same portions as those of the digital copying machine 1 described above as a conventional example are denoted by the same names and reference numerals, and detailed description thereof will be omitted.

First, in the digital copying machine 6 of this embodiment,
As shown in FIG. 1, an image storage unit 7 is connected to the image processing unit 4, and a printer unit 5 is connected to a changeover switch 8 that selects one of the image storage unit 7 and the image processing unit 4. I have. The image storage unit 7 is connected to a quantizer selecting unit 9, which outputs a quantizer selecting signal according to a manual operation.

Next, the structure of the image storage unit 7 will be described below with reference to FIG. First, the image processing unit 4
A 4-line FIFO (First In First Out) 10 which is a data decomposing means for converting pixel data into block data is connected. This 4-line FIFO 10 is a density calculating means for converting block data into density data.
An encoding unit 11 that is also a data encoding unit that encodes the density data into compressed data is connected. The encoding unit 11 has a random access memory (RAM) for storing compressed data.
Memory 12 is connected.

The memory 12 is connected to a decoding unit 13 as data decoding means for decoding compressed data into block data. The decoding unit 13 includes a 4-line FIFO 14 for converting block data into image data. It is connected. The four-line FIFO 14 is connected to the changeover switch 8, and the encoder 11 and the decoder 1
3 is connected to the quantizer selector 9.

Next, the structure of the encoding unit 11 will be described below with reference to FIG. First, the 4-line FIFO
10, a maximum / minimum calculator 15 for calculating the maximum value and the minimum value of the density data, and a delay buffer 16 including a flip-flop are connected. The maximum / minimum calculator 1
5 includes an average index calculating unit 17 for calculating average data La and gradation width index Ld from the maximum value and the minimum value of the density data.
Is connected. The average index calculator 17 includes a threshold calculator 18 as threshold switching means for calculating quantization thresholds P1 and P2 from the memory 12, the average data La and the gradation width index Ld, and thresholds P1 and P2. Average data L
A code allocating unit 19 that allocates a code φij for each pixel to a and the gradation width index Ld is connected. This code allocating unit 1
The delay buffer 16, the threshold calculator 18, and the memory 12 are connected to 9.

Next, the structure of the decoding unit 13 will be described below with reference to FIG. First, the representative quantization values Q1 to Q1 are stored in the memory 12 based on the average data La and the gradation width index Ld.
Representative value calculation unit 20 that is a representative value switching unit that calculates Q4
The quantizer selector 9 is connected to the representative value calculator 20. The representative value calculation unit 20
The memory 12 is connected to a quantized value ratio unit 21 for calculating block data from the quantized representative values Q1 to Q4 and the code φij for each pixel.

In the threshold value calculating section 18, a plurality of arithmetic expressions for calculating the threshold values P1 and P2 from the average data La and the gradation width index Ld are set in a switchable manner. Is set such that a plurality of arithmetic expressions for calculating the quantization representative values Q1 to Q4 from the average data La and the gradation width index Ld are switchable. The arithmetic expressions of the threshold value calculation unit 18 and the representative value calculation unit 20 are as shown in Table 1 below.

[0029]

[Table 1]

Switching is performed by a 2-bit quantizer selection signal output from the quantizer selector 9.

In such a configuration, the digital copying machine 6 of the present embodiment encodes image data read from a manuscript (not shown) by the scanner unit 2 into compressed data and stores it in the memory 12. Is decoded into image data, and reproduced by the printer unit 5 on a print sheet (not shown).

The image data read from the original by the scanner unit 2 can be directly copied to printing paper by the printer unit 5. When copying image data directly in this manner, the image processing unit 4 is connected to the printer unit 5 by the changeover switch 8, and a copying operation is executed in the same manner as in the digital copying machine 1 described above as a conventional example. .

Then, the image data is encoded and stored in the memory 1.
When decoding is performed after the image data is stored in the storage unit 2, the image storage unit 7 is connected to the printer unit 5 by the switch 8. In this case, in the image storage unit 7, the line data output by the image processing unit 4 for one pixel data is
By accumulating four lines by the IFO 10, each of the multi-gradation pixel data is decomposed into “4 × 4” to generate block data. The encoding unit 11 has a 4-line FIFO1
0, the density data Lij is calculated by counting the number of black and white for each block data output, and the density data Lij is converted into a code φ for each pixel by a predetermined algorithm.
It is encoded into compressed data consisting of ij, average data La and gradation width index Ld.

More specifically, when the maximum value Lmax and the minimum value Lmin are calculated from the density data Lij by the maximum / minimum calculation unit 15, the average index calculation unit 17 calculates La = (Lmax + Lmin) / 2 Ld = (Lmax -Lmin) / 2, the average data La and the gradation width index Ld are calculated from the maximum value Lmax and the minimum value Lmin. Next, the threshold calculator 18 calculates the average data as P1 = La + 1 / 2Ld P2 = La-1 / 2Ld, for example, using an arithmetic expression switched by the quantizer selection signal output from the quantizer selector 9. Threshold value P from La and gradation width index Ld
1 and P2 are calculated.

Since the block data delayed by the delay buffer 16 is transmitted to the code allocating unit 19, the code allocating unit 19 compares the average data La with the threshold values P1 and P2 with the block data. Since a 2-bit code φij is assigned to each pixel, the code φ
Compressed data including ij, average data La, and gradation width index Ld is stored.

Next, the decoding unit 13 reads out the compressed data stored in the memory 12 as described above,
3 decodes the read compressed data into density data Lij. More specifically, the representative value calculation unit 20 calculates, for example, Q1 = La + 3 / 4Ld Q2 = La + 1 / 4Ld Q3 = La by an arithmetic expression switched by the quantizer selection signal output from the quantizer selection unit 9. Assuming −１ ／ Ld Q4 = La− ／ Ld, the quantization representative values Q1 to Q4 are calculated from the average data La and the gradation width index Ld. Compressed data is decoded into block data by the quantization value ratio section 21 allocating a code φij for each pixel to the quantization representative values Q1 to Q4.

The block data consisting of "4.times.4" pixels is stored in the 4-line FIFO 14 and then output as 4-line image data to the printer unit 5. Is printed out on printing paper.

As described above, the image data read from the original by the scanner unit 2 is encoded into compressed data and
2, the compressed data is decoded into image data, and can be reproduced on printing paper by the printer unit 5. However, due to the reading characteristics of the scanner unit 2 and the printing characteristics of the printer unit 5, it may be better to change the method of calculating the threshold value in encoding and the method of calculating the quantization representative value in decoding in some cases. In such a case, the threshold value calculation method and the quantization representative value calculation method can be selected by manual operation of the quantizer selection unit 9.

When a method of calculating a threshold value or a method of calculating a quantization representative value is selected by manual operation of the quantizer selecting unit 9, the quantizer selecting unit 9 sends the threshold value calculating unit 18 and the representative value calculating unit 20 two bits. Is output, the threshold value calculating unit 18 switches the arithmetic expression of the threshold value according to the quantizer selection signal, and the representative value calculating unit 20 switches the arithmetic expression of the quantized representative value. By repeating the switching of the arithmetic expression and the copying operation, and selecting the arithmetic expression in a state where the result is the best, the digital copying machine 6 can be set in the best state.

[0040]

According to the first aspect of the present invention, there is provided a threshold value switching means for switching a method of calculating a threshold value of the data encoding means,
By providing the threshold value selection means for selecting the calculation method to be switched by the threshold value switching means, when the threshold value calculation method is selected by the threshold value selection means, the threshold value calculation method is switched by the threshold value switching means. Can change the parameters of the quantization in the encoding of
The result when encoding and decoding image data can be adjusted to a favorable state.

According to a second aspect of the present invention, there is provided a representative value switching means for switching the calculation method of the quantized representative value of the data decoding means, and a representative value selection means for selecting the calculation method to be switched by the representative value switching means. Thereby, when the method of calculating the quantized representative value is selected by the representative value selecting means,
Since the calculation method of the quantization representative value is switched by the representative value switching means, the quantization parameter in the decoding of the image data can be changed, and the result of encoding and decoding the image data is adjusted to a good state. can do.

According to a third aspect of the present invention, there is provided a threshold value switching means for switching a calculation method of a threshold value of the data encoding means, a threshold value selection means for selecting a calculation method to be switched by the threshold value switching means, The representative value switching means for switching the calculation method of the generalized representative value is provided, and the representative value selection means for selecting the calculation method to be switched by the representative value switching means is provided. Since the threshold value calculation method is switched by the threshold value switching means, it is possible to change the quantization parameter in the encoding of the image data, and the representative value selection means selects the quantization representative value calculation method. Since the calculation method of the quantization representative value is switched by the value switching unit, it is possible to change the quantization parameter in the decoding of the image data. Can, the results of the case of decoding image data encoded can be adjusted in a good state.

According to a fourth aspect of the present invention, the method of calculating the threshold value in the encoding by the data encoding means is switchable by the threshold value switching means, and this switchable calculation method is freely selectable by the threshold value selection means. When the threshold value calculation method is selected by the threshold value selection means, the threshold value calculation method is switched by the threshold value switching means, so that it is possible to change the quantization parameter in the encoding of the image data, and to encode the image data. Thus, the result of decoding can be adjusted to a good state.

According to a fifth aspect of the present invention, the method of calculating the quantized representative value in the decoding by the data decoding means can be switched by the representative value switching means, and this switchable calculation method can be selected by the representative value selecting means. Accordingly, when the method of calculating the quantized representative value is selected by the representative value selecting means, the method of calculating the quantized representative value is switched by the representative value switching means, so that the quantization parameter in the decoding of the image data is changed. The result of encoding and decoding image data can be adjusted to a favorable state.

According to a sixth aspect of the present invention, the method of calculating the threshold value in the encoding by the data encoding means is made switchable by the threshold value switching means, and this switchable calculation method is made freely selectable by the threshold value selection means. The method of calculating the quantized representative value in the decoding of (i) is made switchable by the representative value switching means, and this switchable calculation method is made selectable by the representative value selecting means, so that the threshold calculating method is selected by the threshold selecting means. Then, the calculation method of the threshold value is switched by the threshold value switching means, so that the quantization parameter in the encoding of the image data can be changed, and the calculation method of the quantization representative value is selected by the representative value selection means. Since the calculation method of the quantization representative value is switched by the representative value switching unit, the quantization parameter in the decoding of the image data is changed. It is possible, the results of the case of decoding image data encoded can be adjusted in a good state.

[Brief description of the drawings]

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

FIG. 2 is a block diagram illustrating an image storage unit.

FIG. 3 is a block diagram illustrating an encoding unit that is a data encoding unit.

FIG. 4 is a block diagram illustrating a decoding unit that is a data decoding unit.

FIG. 5 is a block diagram showing a digital copying machine as a conventional example of an image processing apparatus.

6A is a schematic diagram showing a state where image data is decomposed into pixel data, and FIG. 6B is a schematic diagram showing a state where pixel data is decomposed into block data.

FIG. 7 is a program list showing an algorithm by which data encoding means encodes density data into compressed data.

[Explanation of symbols]

 Reference Signs List 1 image processing device 9 threshold value selecting means, representative value selecting means 10 data decomposing means 11 density calculating means, data encoding means 13 data decoding means 18 threshold value switching means 20 representative value switching means La average data Ld gradation width index Lij density data φij Sign

Claims (6)

(57) [Claims] 1. A data decomposing means for decomposing each of multi-gradation pixel data forming image data into a predetermined number to generate block data; a density calculating means for calculating density data of the block data; The average data and the gradation width index are calculated from the maximum value and the minimum value, and the quantization threshold is calculated from the average data and the gradation width index. Data encoding means for encoding the density data into the compressed data by assigning a code, a threshold switching means for switching the calculation method of the threshold value of the data encoding means, and a threshold value for selecting the calculation method to be switched by the threshold switching means An image processing apparatus comprising: a selection unit. 2. Compressed data is blocked by calculating a quantization representative value from average data of compressed data including a code for each pixel, average data, and a gradation width index and a gradation width index, and assigning it to a code for each pixel. Data decoding means for decoding into data, representative value switching means for switching a method of calculating a quantized representative value of the data decoding means, and representative value selecting means for selecting a calculation method for switching the representative value switching means. Characteristic image processing device. 3. A data decomposing means for decomposing each of multi-gradation pixel data forming image data into a predetermined number to generate block data; a density calculating means for calculating density data of the block data; The average data and the gradation width index are calculated from the maximum value and the minimum value, and the quantization threshold is calculated from the average data and the gradation width index. Data encoding means for encoding the density data into the compressed data by assigning a code, a threshold switching means for switching the calculation method of the threshold value of the data encoding means, and a threshold value for selecting the calculation method to be switched by the threshold switching means Selecting means for calculating the quantized representative value from the average data of the compressed data and the gradation width index and assigning it to a code for each pixel, thereby compressing the compressed data into block data; Data decoding means for decoding data, a representative value switching means for switching a method of calculating a quantized representative value of the data decoding means, and a representative value selecting means for selecting a calculation method to be switched by the representative value switching means. Characteristic image processing device. 4. A multi-tone pixel data forming image data is decomposed into a predetermined number by data decomposing means to generate block data, and density data of the block data is calculated by density calculating means. Average data and a gradation width index are calculated from the maximum value and the minimum value of the density data, and a quantization threshold is calculated from the average data and the gradation width index. By assigning a code for each pixel, the data encoding means encodes the density data into compressed data, and the threshold value calculation method in this encoding is made switchable by the threshold value switching means. An image processing method characterized in that the image processing method can be freely selected by the user. 5. A data decoding means which calculates a quantization representative value from average data of compressed data including a code for each pixel, average data, and a gradation width index and a gradation width index and assigns the quantization representative value to a code for each pixel. The compressed data is decoded into block data, and a method of calculating a quantized representative value in the decoding is made switchable by a representative value switching means, and this switchable calculation method is made selectable by a representative value selecting means. Image processing method. 6. A multi-tone pixel data forming image data is decomposed into a predetermined number by data decomposing means to generate block data, and density data of the block data is calculated by density calculating means. Average data and a gradation width index are calculated from the maximum value and the minimum value of the density data, and a quantization threshold is calculated from the average data and the gradation width index. By assigning a code for each pixel, the data encoding means encodes the density data into compressed data, and the threshold value calculation method in this encoding is made switchable by the threshold value switching means. And a quantization representative value is calculated from the average data and the gradation width index of the compressed data including the code, the average data, and the gradation width index for each pixel. The compressed data is decoded into block data by the data decoding means by assigning the code to each pixel code, and the method of calculating the quantized representative value in this decoding is made switchable by the representative value switching means. An image processing method characterized by being selectable by a selection means.