JPS63181569A - Picture processing method - Google Patents

Abstract

PURPOSE:To eliminate a characteristic texture in a part monotonous in a density change by binarizing (nXn) picture elements as one block and reallocating error data generated in this binarization processing to adjacent blocks. CONSTITUTION:The sum of the density of picture data is obtained, and then, the number of black picture elements N=[S/Q] and a remainder E=S-NXQ are obtained. In this case, [ ] indicates a Gauss' notation and Q is a prescribed picture signal level. Then, only N picture elements are transformed to the black picture elements in the series of size and a residual elements are decided to white picture elements, thereby, the density of Q is obtained. The error E is weighted to reallocate to the blocks on the circumference of remarkable blocks Bi, j. Thereby, the black picture elements are converged on the higher density to generate an edge emphasizing effect.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to an image processing method.

[Prior Art] When outputting an image having halftones using a dot matrix printer, it is safe to perform binarization processing on the image data. Various binarization processing methods have been announced, but a binarization method called the error diffusion method (Error Diff+ion) is superior to other methods in terms of image quality.

However, this method produces a unique texture,
When the grayscale changes are monotonous, there is a problem in that the texture is extremely noticeable.

[Object of the Invention] The present invention has been made in view of the problems in the conventional methods described above, and an object of the present invention is to provide an image processing method that can eliminate unique textures in areas where density changes are monotonous. It is something to do.

[Embodiments of the Invention] The 1st rA is based on Book 9. FIG. 2 is a block diagram showing one embodiment of the present invention.

In this embodiment, an image input device 2tl, a data addition circuit 2 which adds the image data from the image input device ml and the IIf allocated error from the redistribution circuit 6, and a density summation of the data from the data addition circuit 2 are used. a black pixel number determining circuit 3 that determines the number of black pixels based on; a ranking circuit 4 that sorts the image data from the image input device 1 in descending order of density;
An error correction circuit 5 that calculates and corrects an error based on the sum of the density of data from the data addition circuit 2 and the number of black pixels from the binarization circuit 7; A redistribution circuit 6 for distributing, a binarization circuit 7,
An image output device 8 is provided.

The A1 converting circuit 7 is a circuit that converts data created by one person into black pixels by the determined number of black pixels in descending order of magnitude, and converts the remaining pixels into white to binarize the data.

FIGS. 2(1) and 2(2) are explanatory diagrams of the operation of the redistribution circuit 6 in the above embodiment.

In FIG. 2(1), B indicates a block, Bi,j indicates the block of interest, and FIG. 2(2) shows the pixels of the block of interest Bi,j. Block B is
Generally, it is made up of nXn pixels, but in FIG. 2 it is made up of 4×4 pixels.

Next, the operation of the above embodiment will be explained.

FIG. 3 is a flowchart showing the operation in the above embodiment.

First, calculate the sum S of the density of the image data as shown in FIG. 2 (2) (S t). In the case of the example shown in FIG. The number N and the remainder E are determined (S2).

N= [S/Q]・・・・・・・・・・・・(1) E=
S-NXQ・・・・・・・・・・・・(2) However, [
] is a Gaussian symbol (a symbol that calculates only the integer part within [])
, and Q is a predetermined image signal level.

Next, the order of magnitude of the image data shown in FIG. 2 (2) is examined. At this time, items having the same value are determined according to a predetermined order. In order of size investigated as above, N
are converted to black pixels, and the rest are determined to be white pixels (S
3), that is, the concentration becomes Q.

Then, the error E obtained by the above equation (2) is collected and redistributed to the blocks around the block of interest Bi, j shown in FIG. 2 (1) (S4).

By doing the above, black pixels are concentrated in areas with higher density, thereby producing an edge enhancement effect.

Further, since the average density of the input image and the output image are approximately equal in the block size being scanned, moire does not occur.

FIG. 4 is a block diagram showing another embodiment of the present invention.

This embodiment is connected to an image input device 9, a data addition circuit 10 that adds input data from the image input device 9, and an error redistributed from a redistribution circuit 13, and is connected to the data addition circuit 10. A black pixel number provisional determination circuit 11 that temporarily determines the number of black pixels in a block, a black pixel number determination circuit 12 that officially determines the number of black pixels in a block, and a redistribution circuit 13 that redistributes the corrected error. D: An error correction circuit 14 that corrects the error between the data from the addition circuit IO and the binarized data from the binarization circuit 16, and a ranking circuit that ranks the input data from the image processing device 9 in descending order of magnitude. 15, a binarization circuit 16 that determines input data to be black or white, and an image output device 17 that outputs the binarized data.

Next, the operation of the embodiment shown in FIG. 4 will be explained.

FIG. 5 is a flowchart showing the operation of the embodiment shown in FIG.

First, calculate the sum S of the density of the image data in the block (S l 1). In the above case, since the image data is 4 x 4, next calculate the number N of black pixels and the remainder E (S 12 ).

N= [S/Ql・・・・・・・・・・・・(3) E=
S-QXN・・・・・・・・・・・・(4) However, [
] is a Gaussian symbol, which calculates only the integer part of the symbol. Further, Q is a predetermined image signal level and usually takes the same value as the number of gradations of the original image, but is not limited thereto.

Next, using the remainder E obtained by the above equation (4), it is compared with the threshold Z, and as a result, if the remainder E is greater than or equal to the threshold (I 2), then (5
13), Increment the number of black pixels N by 1, (Si2)
, the remainder E is replaced with (E-Q) (S15).

In this case, the threshold Z may be fixed, but it may be variable, such as decreasing the threshold Z if the average concentration in the block is high, and increasing the threshold Z if the average concentration in the block is high. You can also set the interval to 1.

Next, N pieces of data in the block are determined to be black pixels in descending order (S16).However, if the data sizes are equal, a predetermined rank is taken, and according to that rank, black pixels Determine.

Then, the error E when binarized is weighted and redistributed to the surrounding blocks (S l 7). An example of the distribution at this time is between blocks that are adjacent to each other in the main scanning and sub-scanning directions. , the distribution ratio with the blocks that are in contact with the diagonal direction,
There is a way to make it 2:1.

However, other allocation methods may be adopted.

In this way, when the binarization in one block is completed and the next block is to be binarized, the sum of the redistributed error and the C degree data in the block is set to S, and the process described above is performed. Return green. By doing this for all blocks,
All data can be converted to ray 1.

As mentioned above, the image data is divided into blocks, and the error between before and after binarization is adjusted so that the average density within the block is almost the same before and after binarization. By distributing it to the blocks, moiré can be suppressed and an edge enhancement effect can be achieved.

[Effects of the Invention] According to the present invention, there is an effect that a unique texture can be eliminated in a monotonous portion of density change.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention. FIG. 2(1) is a diagram showing where the error of the block of interest is allocated. FIG. 2(2) is a diagram showing the density of each pixel in the block of interest Bi,j. FIG. 3 is a flowchart showing the operation of the above embodiment. FIG. 4 is a block diagram showing another embodiment of the present invention. It is a diagram. FIG. 5 is a flowchart showing the operation of the embodiment shown in FIG. l, 9... Image human power device, 3.12... Black pixel number determination circuit, ll... Black pixel number provisional determination circuit. 4.15... Ranking circuit, 5.14... Error correction circuit, 6.13... Redistribution circuit, 7.16... Stingray 1 circuit. Patent applicant Canon Co., Ltd. Agent Arata Yokubo - Figure 1 Figure 2 (2) B1. j Figure 3 Figure 4

Claims (1)

[Claims] Image data having halftones obtained by an image input device,
In an image processing method that performs binarization processing based on the error diffusion method, nXn pixels are binarized as one block, and the error data generated in this binarization processing is redistributed to adjacent blocks. Featured image processing method.