JPH0318177A - Picture processor - Google Patents

Abstract

PURPOSE:To attain error spread processing suitable for a multi-value printer capable of outputting a multi-value picture with several gradation degrees from one picture element by using plural threshold levels so as to apply multi-value processing to a data after the processing. CONSTITUTION:A picture data consecutive in the main scanning direction read by a scanner 101 is subjected to gamma correction by a gamma correction circuit 102 and the result is fed to an adder 103. The adder 103 adds the data with an error G of surrounding picture elements to which the processing by a divider 108 is finished and the result becomes an address input to a ROM 104. The ROM 104 according to the address input outputs a 3-bit data for a multi-value code ID' for the result of processing and a 5-bit data for the output of a processing error G'. The multi-value code ID' being the result of processing is inputted to a printer 109, from which a picture is outputted. Moreover, the processing error G' is stored in an error memory 105. The error memory 105, a multiplier 106, an adder 107 and a divider 108 calculate the error quantity to an input picture element.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a copying machine that uses halftone digital image data. The present invention relates to an image processing apparatus applied to an image processing unit of a facsimile, a printer, etc., and particularly relates to an image processing apparatus characterized by improving the image quality of a pseudo-halftone binary image generated from multi-tone data using an error diffusion method.

[Conventional technology]

Dither processing is generally used as an image processing method to perform halftone processing on multi-tone image data and convert it into data suitable for output devices such as printers that can only output 2I direct (pixel output or not). There is.

Dither processing compares input image data with a dither matrix pattern and determines the binary output of one pixel based on the magnitude relationship between the two, but this method has the drawback that resolution and gradation are not compatible.

On the other hand, there is an error diffusion method as a halftone process that reproduces relatively well both resolution and gradation.

The error diffusion method is a process in which the density error that occurs during binarization processing is saved and used when processing surrounding pixels, and the density can be preserved even in images after gradation processing. It is.

In addition, as a literature regarding the error diffusion method, R. W F
roydand L. Steinberg”An
Adaptive Algorithm for Sp
ecial Gray Scale”S TD
75 Digest and Television Society Journal Vol.
40. No. 4 (1986). p. 320R halftone digitization.

[Problem to be solved by the invention]

In this way, error diffusion processing methods that process input multivalued image data to create a binary image have been known for a long time. It could not be said to be suitable for a printer capable of outputting gradations on the order of values. The purpose of the present invention is to obtain a number of pixels Il1gJ
An object of the present invention is to provide an image processing device that performs error diffusion processing suitable for a multivalue printer capable of outputting an I multivalue image.

[Means to solve the problem]

The above object is to provide an addition means for adding errors of input pixel data and peripheral pixels, and an addition result, in an image processing device that binarizes image data having multiple gradations obtained by an image input device using an error diffusion method. This is accomplished by providing a processing means that inputs an address, compares it with a stored threshold value, and outputs multivalued data.

[Effect]

In order to take advantage of the characteristics of halftone processing using the conventional error diffusion method and to obtain output image data with several gradations, we eliminated the binarization processing in the conventional error diffusion processing and replaced it with multiple threshold values. Using this method, the processed data is multi-valued.

〔Example〕

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

First, the outline of the present invention will be explained with input pixel data 10 (0
to 64) to process the 8-value pixel data [D' (0 to 7
) will be explained using an example.

First, the input pixel data ID is added to the error G of peripheral pixels for which processing has been completed, similar to conventional error diffusion processing. Next, it is compared with multiple threshold values and multivalued. Here, the number of thresholds is 7 because it is multi-valued to 8 values, and the 8-value pixel data ID' after processing is ID + G, 1 when the threshold is 1 or more, and 1 when the threshold is 2 or more. When the value is 2, it becomes 7 when the threshold value is 7 or more. Further, when ID+G is less than the threshold value 1, ID' becomes 0.

Figure 2 shows TD+G. ? 、． This is a relationship diagram between threshold value and ID', and here, as an example, the threshold interval is made equal;
Usually, it is adjusted according to the multi-value output characteristics of the printer. For example, if the difference in density between the 2nd value and the 3rd value is larger than the difference between the 3rd value and the 4th value due to the output characteristics of the printer, the interval between the threshold value 2 and the threshold value 3 may be changed. Threshold 3 and Threshold 4
be larger than the interval between.

Furthermore, processing errors are also calculated at the same time as this 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 it exceeds. For example, when the input TD+G is 37, it exceeds 32, which is ID+G of threshold 4, but does not exceed 40, which is TD+G of threshold 5, so the processing error is 37-32=5. This processing error is held in an error memory in the same way as in conventional error diffusion processing, and is used for subsequent pixel processing.

This will be explained in detail below.

FIG. 1 is a block diagram of an embodiment of an image processing apparatus according to the present invention, in which continuous image data in the main scanning direction read by a scanner 101 is subjected to γ correction in a γ correction step &S102, and sent to an adder 103. Sent. Adder 10
3, addition is performed with the error G of the peripheral pixels whose processing has been completed from the divider 108, and the addition result is stored in the ROM 104.
This is the address input. ROM104 is 8 bits x 12
It has a capacity of about 8 and produces a multi-value code (8-value pixel data) as a processing result in response to address input! Three bits of data are output for D' and five bits of data are output for processing error G'.

FIGS. 3 and 4 are explanatory diagrams of ROM address contents.

The multivalued code ID' resulting from the processing is sent to the printer 109 and output as an image. Furthermore, the processing error G' is held in the error memory 105 and used for subsequent pixel processing.

The error memory 105, multiplier 106, adder 1o7, and divider 10B calculate the amount of error to be added to the input pixel.

The closer the distance to the input pixel to be processed, the larger the weight matrix data, and the stronger the influence of errors. The adder 107 adds errors around the input pixel, and the divider 108 divides the result by a constant value and adds it to the input pixel. This part is the same as conventional error diffusion processing, so detailed explanation will be omitted.

Note that the addition means described in the claims is the adder 10.
3, and a ROM 104 constitutes the processing means.

〔Effect of the invention〕

As explained above, according to the present invention, it is possible to output multivalued image data from 3 to 8 values while performing error diffusion processing. It is possible to provide an image processing device suitable for a multilevel printer that can output .

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of an image processing apparatus according to the present invention, FIG. 2 is an explanatory diagram showing the correspondence between error correction input data, threshold values, and output multi-value data, and FIGS. 3 and 4 are It is an explanatory diagram of ROM address contents. 103... Adder, link. 4... ROM, 7... Figure 3 Figure 4

Claims (1)

[Claims] An image processing device that binarizes multi-gradation image data obtained by an image input device using an error diffusion method includes an adding means for adding errors between input pixel data and peripheral pixels, and an address input for the addition result. 1. An image processing apparatus, comprising: a processing means for performing multivalued data, comparing the data with a stored threshold value, and outputting multivalued data.