JP3167676B2 - Image processing device - 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 applied to an image processing section such as a copying machine, a facsimile, a printer, etc. using digital image data, and more particularly to an image processing apparatus which is generated from multi-tone data by an error diffusion method. The present invention relates to an image processing apparatus characterized by improving the image quality of a pseudo halftone image.

[0002]

2. Description of the Related Art Halftone processing is performed on input multi-tone image data,
Dither processing is generally used as an image processing method for converting data into data suitable for an output device such as a printer that can output only binary values (pixels are output / not output).

In dither processing, input image data is compared with a dither matrix pattern.
Although this method determines the value output, this method has a drawback that resolution and gradation are not compatible.

[0004] On the other hand, there is an error diffusion method as a halftone process in which both resolution and gradation are relatively well reproduced.

[0005] In the error diffusion method, a density error generated at the time of the binarization processing is stored and used for processing of peripheral pixels.
This processing is characterized in that the density can be preserved even in the image after the gradation processing. Literature includes RW Floyd and
L. Steinberg “An Adaptive Algorithm for Special Gra
y Scale "SID 75 Digest and Journal of the Television Society Vol.4
0, No. 4 (1986), p. 320, "Digitalization of halftone".

Here, the gradation processing means that the number of gradations of input image (pixel) data is determined by using a copier, a facsimile,
This refers to a process of reducing the number of tones that can be output by an image output device such as a printer.

Conventionally, since most image output devices output binary data, they may be expressed as a binarization process. Indicates multi-value processing.

[0008]

As described above, error diffusion processing for processing input multi-tone data to create a pseudo halftone image has been conventionally known. However, conventional error diffusion processing for binarization processing has been known. In the error diffusion method, a process suitable for an output device such as a printer that assigns a multi-level gradation to one pixel cannot be performed.
Therefore, binarization using the conventional error diffusion method is not performed, and multi-leveling is performed using a plurality of threshold values. However, since the optimal number of levels of multi-leveling differs depending on the output printer,
The hardware configuration of the error diffusion circuit had to be changed in order to change the number of stages of multi-value processing in accordance with each output printer.

An object of the present invention is to provide an image processing apparatus capable of changing the number of levels of multi-value processing with the same hardware configuration.

[0010]

To achieve the above object, the present invention provides an image processing apparatus for performing halftone processing on input multi-tone image data by an error diffusion method. Multi-valued by performing error correction based on the image data after error diffusion processing on the image data of the image data, making the error-corrected image data multi-valued by a plurality of threshold values, and performing halftone processing by an error diffusion method Error diffusion processing means and the multi-level error selection signal,
A first bit width having a first bit width output from the
A second bit narrower than the first bit width is obtained from the error data.
And selecting the second error data having a width of
Outputs error data to be added to the input data of the dispersion processing means
And selecting means for performing the selection.

[0011]

Embodiments of the present invention will be described below with reference to the drawings.

First, regarding the outline of the present invention, multi-level error diffusion processing is performed on input pixel data ID (0 to 127) of 128 gradations (7 bits) after gamma correction processing to perform 9-valued pixel data I
An example will be described in which D ′ (0 to 8) is created, and ID ′ is further subjected to multi-value data correction processing to create 10-level pixel data ID ″ (0 to 9).

First, the input pixel data ID is added to the error G of the peripheral pixel that has been processed, as in the conventional error diffusion processing. Next, the values are compared with a plurality of threshold values and multi-valued. Here, since the number is multi-valued to nine values, the number of thresholds becomes eight, and the post-processing data ID ′ is 1 when ID + G is equal to or more than the threshold 1, 2 when the ID + G is more than the threshold 2, and so on. Similarly, when the threshold value is 8 or more, the value becomes 8. Also, ID
When + G is less than the threshold value 1, ID 'becomes 0.

FIG. 2 shows the relationship between ID + G and threshold values 1 to 8 and ID '. Here, as an example, the threshold intervals are made equal, but adjustment is made in accordance with the multi-value output recording characteristics of a printer or the like. For example, if the difference between the densities of the second and third values is larger than the difference between the third and fourth densities in the output characteristics of the printer, the interval between the threshold 2 and the threshold 3 is set to the threshold. It is set to be larger than the interval between 3 and the threshold value 4.

A processing error is calculated simultaneously with the multi-value processing. The processing error is the difference between the input ID + G and the value of ID + G corresponding to the largest threshold value that has exceeded it. For example, when the input ID + G is 74, it exceeds the threshold ID + G of 64, but the threshold 5 ID + G
Since G does not exceed 80, the processing error is 74-64 =
It becomes 10. This processing error is held in the error memory in the same manner as the conventional error diffusion processing, and is used for the subsequent pixel processing.

The details will be described below.

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

In FIG. 1, image data read by a scanner 101 is subjected to corresponding processing by a gamma correction processing unit 102 and a multi-value error diffusion processing unit 103, and the printer 1
05 is output.

FIG. 3 is a block diagram of a main part of the image processing apparatus according to this embodiment.

The continuous image data read in the main scanning direction by the scanner 101 is γ-corrected by the γ-correction circuit 102 and sent to the adder 301 as a 7-bit data ID.

The adder 301 performs an addition of the error of the peripheral pixel having undergone the processing from the selector 307 and the weighted sum G, and the addition result is input to the address input of the ROM 302 for performing the multi-value processing and the error calculation processing. Become. ROM30
2 has a capacity of about 8 bits.times.256, and outputs 4-bit data as a multi-value code ID 'of the processing result and 4-bit data as an output of the processing error G' in response to the address input.

FIGS. 4A, 4B and 4C show the relationship of FIG.
The example realized in 302 is shown.

The multi-value code ID 'of the processing result is stored in the printer 1
05 and output as an image. The processing error G 'is stored in the error buffer memory 304,
The data is sent to the diffusion matrix creation unit 305 and used for the subsequent pixel processing.

An error buffer memory 304, a diffusion matrix creation unit 305, and a weighting adder 306 calculate an error amount G to be added to an input pixel. The closer the distance to the input pixel to be processed is, the larger the data of the diffusion matrix is, and the error in the surrounding pixels is weighted to increase the influence on the input pixel.

FIG. 5 shows an example of the data of the diffusion matrix. Here, * represents a processing pixel (target pixel).

The error amount calculated by the weighting adder 306 is 7-bit data. The selector 307 selects 4 bits of the data and sets it as error data G.
Send to There are four types of methods for selecting 4-bit data from 7-bit data, as described below, and are selected by a 2-bit multi-level level selection signal.

Bit 0 to Bit 3... 9-value multi-value conversion Bit 1 to Bit 4... 5-value multi-value conversion Bit 2 to Bit 5... 3-value multi-value conversion Bit 3 to Bit 6. The formula for calculating the amount of error performed by the weighting adder is shown below. Since this formula is generally used for error diffusion processing, a detailed description thereof will be omitted. Note that this
Is handled as described in the previous topic.
The error diffusion method is not binarized as in
If the value is multivalued, the optimal number of multilevel
Output printers are different.
Error diffusion circuit to change the number of multilevel
This is because it is necessary to change the hardware configuration of the road.
By doing so, the same hardware configuration can be used.
It is possible to change the number of stages of multi-value processing. Further details
To clarify, the above or the multi-valued level (9, 5,.
Which bits of error data of 3 and 4 values) and 7 bits
4 indicates whether to select a 4-bit error from.
As the level of multi-value increases, the error naturally decreases
Therefore, the error selection bit must be selected from the lower
Become. In this embodiment, as described above,
9 is explained, so it is explained along with this.
Then, in the selection of bit data, as can be seen from FIG.
When the multi-valued level = 9, the threshold value is 8 (2 of 8)
Power = 128), and ID + G = 144.
The difference value (G) = 144−128 = 16 (2 to the fourth power).
The error value is in the range of 0 to 3 which is the lower 4 bits.
You. For this reason, the selector 307 operates from bit 1 to bit 3
Is selected, and the selector 30
The hardware configuration can be changed only by selecting the bit data by 7
It becomes unnecessary. That is, the present invention provides a binary processing printer.
Halftone image by multi-valued data and performing error diffusion processing
The present invention relates to an image processing apparatus for processing image data.
However, the multi-level
Feedback with error diffusion when processing with different
Multi-valued bit range to take 4-bit error data
The selector 307 is switched according to the level to change the 4-bit
It is those you so that you to select the error data. This
By selecting the error data as described above,
And the same model changes, the level of multi-level changes
Error diffusion processing can be performed well with the same hardware.
I can.

G = Σ (Gij × Wij) Here, Gij is a processing error of a peripheral pixel stored in the error buffer memory 304, and Wij is diffusion matrix data as shown in FIG.

[0029]

As described above, according to the present invention,
In an image processing device that performs halftone processing by the error diffusion method and outputs multi-valued image data, even if the number of multi-valued stages is changed , an error value suitable for that can be selected. With this configuration, it is possible to select an appropriate number of levels of multi-value processing for each printer and output an image.

[Brief description of the drawings]

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

FIG. 2 is an explanatory diagram showing a relationship among input image data, a weighted sum of errors, processed data, and the like.

FIG. 3 is a block circuit diagram of a main part of the image processing apparatus according to the embodiment of the present invention.

FIG. 4 is a diagram showing an example in which the contents shown in FIG. 2 are implemented in a ROM.

FIG. 5 is a diagram illustrating an example of a diffusion matrix coefficient.

[Explanation of symbols]

 Reference Signs List 101 Scanner 102 Gamma correction processing unit 103 Multi-level error diffusion processing unit 105 Printer 301 Adder 307 Selector

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-3-68083 (JP, A) JP-A-3-286680 (JP, A) JP-A-3-62783 (JP, A) 140061 (JP, A) JP-A-63-35091 (JP, A) JP-A-58-156266 (JP, A) JP-A-57-125579 (JP, A) (58) Fields investigated (Int. ⁷ , DB name) H04N 1/40-1/40 104 G06T 5/00-5/30 Patent file (PATOLIS)

Claims (1)

(57) [Claims] An image processing apparatus for performing halftone processing on input multi-tone image data by an error diffusion method, based on the image data after error diffusion processing on the input multi-tone image data. Multi-level error diffusion processing means for performing multi-level error correction, multi-leveling the error-corrected image data by a plurality of threshold values, and performing halftone processing by an error diffusion method, and the multi-level level selection signal for multi-level error diffusion. Value error diffusion processing means
Error data having a first bit width output from
Has a second bit width narrower than the first bit width from
Selecting the second error data, the multi-level error diffusion processing means
An image processing apparatus, comprising: selecting means for outputting error data to be added to the input data .