CN1424694A - Frequency modulation gridding method for improving quality value of F.M. grid medium tone - Google Patents
Frequency modulation gridding method for improving quality value of F.M. grid medium tone Download PDFInfo
- Publication number
- CN1424694A CN1424694A CN 02159181 CN02159181A CN1424694A CN 1424694 A CN1424694 A CN 1424694A CN 02159181 CN02159181 CN 02159181 CN 02159181 A CN02159181 A CN 02159181A CN 1424694 A CN1424694 A CN 1424694A
- Authority
- CN
- China
- Prior art keywords
- pixel
- diffusion
- original image
- image
- frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Facsimile Image Signal Circuits (AREA)
- Image Processing (AREA)
Abstract
This invention relates to a frequency-modulated net hanging method. Various prior frquency-modulated net hanging methods have the problems that ripples exists commonly in the intermediate tone of images, and that the quality of output frequency-modulated net points is worse or the speed of that is lower. The present invention gives out an optimized method and a controllable shaking mechanism of domain value of white noise on the basis of the prior frequency-modulated net hanging method. The method of the present invention can effectively improve the effects of net hanging, while increasing the speed of net hanging. Furthermore, the particle size can be controlled by adjusting shaking amplitude according to requirements.
Description
Technical field
The present invention relates to a kind of image hard copy and duplicate the site generation method in field, be specifically related to a kind of frequency-modulation screening method that improves the numerical value of modulating net middle tone quality.
Background technology
The linked network plate-making technology that relates generally to printer and high-grade plate making equipment is duplicated in the hard copy of image.Be used for the screening technology that the image hard copy duplicates and be called the digital picture halftone technique again.The digital picture halftone technique can be divided into two classes, is respectively amplitude modulation screening and frequency-modulation screening.Amplitude modulation screening is called convergence point ordered dithering technology again, the stain that it is characterized in that the half tone image that generated is adjacent to condense together on the geometric position in twos, thereby formed the pigmented section of cluster cluster, these pigmented section site that is otherwise known as, because convergence point ordered dithering technology adopts the network point Method for Area to reproduce the original image gray scale, thereby is called as the amplitude modulation site.
Opposite with amplitude modulation screening, the frequency modulation half tone image is avoided the gathering of stain on the geometric position as far as possible in the process of its generation.It realizes the gray-scale rendition of original image by the number of stain in the control unit area.Because being the form with non-gathering, the stain in the frequency modulation half tone image distributes, therefore to different original copy gray levels, will be corresponding to the mean distance between the different stain, angle from Digital Image Processing, the frequency that promptly is image changes, the origin of the noun of Here it is frequency-modulation screening.In the process that generates the frequency modulation half tone image, because stain is being distributed on the image of isolating, so for the fixing original image zone of gray scale, the distribution of stain must guarantee very even, that is the distance between each stain and its neighbour's stain must very approaching foregoing mean distance, so just can avoid harmful texture that may occur.
In traditional frequency-modulation screening method, the method that error diffusion method is the most frequently used.Document " Ditheringwith blue noise ", (Robert A.Ulichney, Proceddings of The IEEE, Vol.76, No.1, January 1988, pp56-79) a kind of error diffusion method is disclosed, in the processing procedure of error diffusion algorithm, (m n) is sequentially handled the pixel g of original image line by line, in the process of handling, each pixel carries out threshold operation (quantizer Quantizer) with a fixing threshold value, the result of computing be converted into the corresponding pixel b of half tone image (m, n).Meanwhile, with this as a result pixel b (m, n) and asked the pixel g of threshold value (m n) compares, and (m n) diffuses on the not processed pixel around the pre-treatment pixel with a diffusion filter (Errorfilter) with the difference e of relatively trying to achieve.In processing subsequently, the pixel value of being asked threshold value be original copy pixel g (m, n) and be flooded to error on this pixel and.
There is regular texture in the frequency-modulation halftone dot of the method output above using in medium tone, and there is the worm effect at the place at the high light of shallow mediation.In addition, error spreads to four direction, and for real-time output device, speed is a bottleneck.In actual applications, in order to obtain in all equally distributed stain of each density level, three kinds of improved strategies of general main employing, the first changes the evolving path, waits the raster paths (as shown in Figure 3) that replaces above us as adopting serpentine path, Hilbert curve or random walk; It two is the methods that adopted diffusion filter variable, as changing the weight of range of scatter and each dispersal direction; It three is to carry out thresholding shake, adopts certain algorithm constantly to change thresholding, rather than fixing thresholding.This patent mainly is the breakthrough at second, third aspect.
Summary of the invention
The objective of the invention is at traditional error diffusion method reality in the second-rate and slow-footed defective of output frequency-modulation halftone dot middle tone, propose a numerical method that can be good at improving modulating net middle tone quality, and taken this to propose to optimize accordingly and improve the method for modulating net network point distribution.Both improve output speed, improved output effect again.
For reaching above purpose, the technical solution used in the present invention is:
A kind of frequency-modulation screening method that improves the numerical value of modulating net middle tone quality may further comprise the steps:
1) input original image, the pixel g of original image (m n) is sequentially handled line by line, and each pixel carries out threshold operation with a controlled threshold value nThreshold, the result of computing be converted into the corresponding pixel b of half tone image (m, n);
2) with b (m, n) and asked threshold value pixel g (m n) compares, and try to achieve difference e (m, n);
3) (m n) diffuses on the not processed pixel around the pre-treatment pixel with diffusion filter, promptly carries out error diffusion with difference e;
4) repeat above step, up to all pixel g of original image (m, n) processed intact, form the half tone image of original image at last;
It is characterized in that:
(1) carry out the thresholding computing before, determine optimum coefficient of diffusion and jitter amplitude earlier: elder generation is to each gray level of original image, set initial error diffusion coefficient (three directions) and a jitter amplitude, carry out frequency-modulation screening then, the frequency spectrum of image after the analysis linked network, and with the degree of correlation of the radiation spectra of its three directions as optimization aim, utilize the simple form evolutionary method to be optimized, thereby search out optimum coefficient of diffusion and jitter amplitude nThresholdScale;
(2) utilize the jitter amplitude nThresholdScale that finds to calculate thresholding nThreshold by following formula:
nThreshold＝127+(rand()％256-127)*nThresholdScale/100
Here rand () is the random number generator;
When (3) carrying out error diffusion, utilize thresholding nThreshold that previous calculations comes out and the optimum coefficient of diffusion that finds to carry out error diffusion.
That supposes the original image G that treats linked network widely is W, and height is H, and this algorithm is converted into the frequency modulation shadow tone binary map B of same size, g[i with this image, j] be original image (i, the pixel value of j) locating, b[i, j] be on the half tone image (specific algorithm is as follows for i, the pixel value of j) locating:
Original image scans along serpentine path, get (i, the original image pixel value g[i that j) locates, j]
if(g[i，j]＞nThreshold)
b[i，j]＝255；
else
b[i，j]＝0；
error＝b[i，j]-g[i，j]；
This error (error) be multiply by the coefficient of the optimum diffusion filter that search comes out respectively, then respectively with the original image pixel addition of relevant position, and the result replaced the corresponding pixel value of original image.
Effect of the present invention is: by method of the present invention, can obtain optimum coefficient of diffusion and jitter amplitude very effectively on the prior art basis, thereby produce the very uniform frequency-modulation halftone dot of distribution.Shake strategy simultaneously provided by the invention, difference that can be as requested obtains the network point distribution of desired particle size by the adjusting jitter amplitude.The present invention is based on digital signal processing and optimum theory scheduling theory basis, realize simply easy operating.
Description of drawings
Fig. 1 is an error diffusion schematic diagram in the frequency-modulation screening method, and wherein (m n) is the input original copy to g, and (m n) is output original text after the error diffusion to b, and quantizer is a quantizer, and Error filter is the error diffusion wave filter, and error is the error that will spread;
Fig. 2 is traditional error diffusion filter schematic;
Fig. 3 is a various error diffusion path synoptic diagram, from left to right is followed successively by raster paths, Hilbert path and serpentine path;
Fig. 4 is the error diffusion filter schematic that this method is used;
Fig. 5 is the frequency spectrum of image after the desirable error diffusion;
Fig. 6 is the optimization aim synoptic diagram, with the degree of correlation of the radiation spectra of three directions of frequency spectrum as optimization aim;
Fig. 7 is that gray scale is that 127 original image uses linked network image and the spectrogram after the initial propagations coefficient carries out linked network
Fig. 8 is that gray scale is that 127 original image uses the coefficient of diffusion after optimizing, and jitter amplitude is 60%, carries out linked network image and spectrogram after the linked network.
Fig. 9 is that gray scale is that 127 original image uses the coefficient of diffusion after optimizing, and jitter amplitude is 100%, carries out linked network image and spectrogram after the linked network.
Embodiment
Below in conjunction with accompanying drawing the specific embodiment of the present invention is done to describe further.
A kind of frequency-modulation screening method that improves the numerical value of modulating net middle tone quality may further comprise the steps:
At first to each gray level, set initial error diffusion coefficient (three directions), with a jitter amplitude, carry out frequency-modulation screening, analyze the frequency spectrum of image after the linked network, with the degree of correlation of the radiation spectra of its three directions as optimization aim, utilize the simple form evolutionary method to be optimized, as Fig. 5, shown in Figure 6, thus search out optimum coefficient of diffusion (d
10, d
-11, d
01) and jitter amplitude nThresholdScale.
Utilize the jitter amplitude nThresholdScale that finds previously to calculate thresholding nThreshold:nThreshold=127+ (rand () %256-127) * nThresholdScale/100 then by following formula
Here rand () is the random number generator.
Carry out error diffusion at last according to the following steps, try to achieve the half tone image of original image:
1) input original image, (m n) sequentially handles by serpentine path the pixel g of original image, and each pixel carries out the thresholding computing with the thresholding nThreshold that utilizes jitter amplitude to calculate, the result of computing be converted into half tone image corresponding pixel value b (m, n);
2) with this as a result pixel value b (m, n) and asked the pixel value g of threshold value (m n) compares, and tries to achieve error e rror;
3) carry out error diffusion, (m is n) with the error diffusion coefficient (d that searches previously with error e
10, d
-11, d
01) diffuse on the not processed pixel around the pre-treatment pixel (g[i, j+1], g[i ,+1, j-1], g[i+1, j]);
4) repeat above step, up to all pixel g of original image (m, n) processed intact, form the half tone image of original image at last;
Suppose the wide W of being of the original image G that treats linked network, height is H, and this algorithm is converted into this image the frequency modulation shadow tone binary map B of same size
FOR(i＝0；i＜H；i++)
IF (i%2==0) // even number line data processing
{
FOR(j＝0；j＜W；j++)
{
Scan former pictorial data from left to right along this row, get (i, the image pixel value g[i that j) locates, j]
if(g[i，j]＞nThreshold)
b[i，j]＝255；
else
b[i，j]＝0；
error＝b[i，j]-g[i，j]；
Optimize above this error (error) be multiply by respectively the coefficient of diffusion filter,
This coefficient is the function of gray level.Then respectively with the original image pixel addition of relevant position,
And the result replaced the corresponding pixel value of original image
g[i，j+1]＝g[i，j+1]+d
10*error；
g[i，+1，j-1]＝g[i+1，j-1]+d
-11*error；
g[i+1，j]＝g[i+1，j]+d
01*error；
ELSE{ // odd-numbered line processing FOR (j=W-1; J＞=0; J--)
if(g[i，j]＞nThreshold)
b[i，j]＝255；
else
b[i，j]＝0；
error＝b[i，j]-g[i，j]；
g[i，j+1]＝g[i，j+1]+d
10*error；
g[i，+1，j-1]＝g[i+1，j-1]+d
-11*error；
g[i+1，j]＝g[i+1，j]+d
01*error；
From Fig. 8, Fig. 9 as can be seen, the distributing very evenly of site, its Fourier frequency spectrum low frequency part constitute one just round.In addition, by regulating jitter amplitude (nThresholdScale), can obtain varigrained network point distribution, be 60% as the jitter amplitude of Fig. 8, and the granularity of its site is fine and smooth.The jitter amplitude of Fig. 9 is 100%, the granularity little rough of its site.
Get 127 gray scales below and illustrate as an example this patent is how to improve middle tone.At first from before diffusion parameter, search out the coefficient of diffusion and the jitter amplitude of three optimum directions according to optimized Algorithm.Utilize error diffusion algorithm to carry out frequency-modulation screening then.The parameter and the result of 127 gray scales have been listed in the table 2.Illustrate referring to Fig. 7 Fig. 8, Fig. 9 accordingly.
| ??p | Before the shake | After the shake | |||||||||
| Embodiment 1 | ?127 | d’ 10(p) | d’ -11(p) | d’ 01(p) | Jitter amplitude | Legend | ?d 10(p) | ?d -11(p) | ?D 01(p) | Jitter amplitude | Legend |
| ??4/6 | ?1/6 | 1/6 | Do not have | Fig. 7 | ?2/4 | ?1/4 | ?1/4 | ?60％ | Fig. 8 | ||
| Embodiment 2 | ?127 | ??4/6 | ?1/6 | 1/6 | Do not have | Fig. 7 | ?2/4 | ?1/4 | ?1/4 | ?100％ | Fig. 9 |
Claims (1)
1. frequency-modulation screening method that improves the numerical value of modulating net middle tone quality may further comprise the steps:
1) input original image, the pixel g of original image (m n) is sequentially handled line by line, and each pixel carries out threshold operation with a controlled threshold value nThreshold, the result of computing be converted into the corresponding pixel b of half tone image (m, n);
2) with b (m, n) and asked threshold value pixel g (m n) compares, and try to achieve difference e (m, n);
3) (m n) diffuses on the not processed pixel around the pre-treatment pixel with diffusion filter, promptly carries out error diffusion with difference e;
4) repeat above step, up to all pixel g of original image (m, n) processed intact, form the half tone image of original image at last;
It is characterized in that:
(1) carry out the thresholding computing before, determine optimum coefficient of diffusion and jitter amplitude earlier: elder generation is to each gray level of original image, set initial error diffusion coefficient (three directions) and a jitter amplitude, carry out frequency-modulation screening then, the frequency spectrum of image after the analysis linked network, and with the degree of correlation of the radiation spectra of its three directions as optimization aim, utilize the simple form evolutionary method to be optimized, thereby search out optimum coefficient of diffusion and jitter amplitude nThresholdScale;
(2) utilize the jitter amplitude nThresholdScale that finds to calculate thresholding nThreshold by following formula:
nThreshold＝127+(rand()％256-127)*nThresholdScale/100
Here rand () is the random number generator;
When (3) carrying out error diffusion, utilize thresholding nThreshold that previous calculations comes out and the optimum coefficient of diffusion that finds to carry out error diffusion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021591814A CN1144153C (en) | 2002-12-30 | 2002-12-30 | Frequency modulation gridding method for improving quality value of F.M. grid medium tone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021591814A CN1144153C (en) | 2002-12-30 | 2002-12-30 | Frequency modulation gridding method for improving quality value of F.M. grid medium tone |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1424694A true CN1424694A (en) | 2003-06-18 |
| CN1144153C CN1144153C (en) | 2004-03-31 |
Family
ID=4753258
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB021591814A Expired - Lifetime CN1144153C (en) | 2002-12-30 | 2002-12-30 | Frequency modulation gridding method for improving quality value of F.M. grid medium tone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1144153C (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006114031A1 (en) * | 2005-04-27 | 2006-11-02 | Peking University Founder Group Co., Ltd | Frequency modulated screening method based on the dual feedback error diffusion |
| CN1326384C (en) * | 2005-04-07 | 2007-07-11 | 北京北大方正电子有限公司 | Method and apparatus capable of producing frequency-modulation halftone dot in high speed |
| CN100385909C (en) * | 2005-02-03 | 2008-04-30 | 凌阳科技股份有限公司 | Error-diffusion half-tone image treatment system and method |
| CN100411419C (en) * | 2004-05-28 | 2008-08-13 | 大日本网目版制造株式会社 | Threshold matrix generation method, method and apparatus for creating halftone dot image, and recording medium |
| US7782493B2 (en) | 2004-05-28 | 2010-08-24 | Dainippon Screen Mfg. Co., Ltd. | Method of generating threshold matrix for creating halftone dot image and method and apparatus for creating halftone dot image and recording medium |
| CN102023475A (en) * | 2010-10-13 | 2011-04-20 | 北京方嘉彩色印刷有限责任公司 | High-precision frequency-modulated screen dot printing process |
| CN102263881A (en) * | 2010-05-24 | 2011-11-30 | 北大方正集团有限公司 | Method and device for processing halftone image frequency modulation mesh hanging |
| CN101237516B (en) * | 2007-02-02 | 2012-12-19 | 株式会社东芝 | System and method for controlling expansion of sector half-tone adjusting network point |
| CN104417063A (en) * | 2013-08-21 | 2015-03-18 | 北大方正集团有限公司 | Method and apparatus for reducing influences of sleeving alignment errors |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1317604C (en) * | 2004-12-20 | 2007-05-23 | 北京北大方正电子有限公司 | Generating method of linear circular shaped lattice point for printing machine |
-
2002
- 2002-12-30 CN CNB021591814A patent/CN1144153C/en not_active Expired - Lifetime
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7782493B2 (en) | 2004-05-28 | 2010-08-24 | Dainippon Screen Mfg. Co., Ltd. | Method of generating threshold matrix for creating halftone dot image and method and apparatus for creating halftone dot image and recording medium |
| CN100411419C (en) * | 2004-05-28 | 2008-08-13 | 大日本网目版制造株式会社 | Threshold matrix generation method, method and apparatus for creating halftone dot image, and recording medium |
| CN100385909C (en) * | 2005-02-03 | 2008-04-30 | 凌阳科技股份有限公司 | Error-diffusion half-tone image treatment system and method |
| CN1326384C (en) * | 2005-04-07 | 2007-07-11 | 北京北大方正电子有限公司 | Method and apparatus capable of producing frequency-modulation halftone dot in high speed |
| US8023152B2 (en) | 2005-04-27 | 2011-09-20 | Peking University Founder Group Co., Ltd. | Method for frequency-modulation screening using error diffusion based on dual-feedback |
| WO2006114031A1 (en) * | 2005-04-27 | 2006-11-02 | Peking University Founder Group Co., Ltd | Frequency modulated screening method based on the dual feedback error diffusion |
| CN101237516B (en) * | 2007-02-02 | 2012-12-19 | 株式会社东芝 | System and method for controlling expansion of sector half-tone adjusting network point |
| CN102263881A (en) * | 2010-05-24 | 2011-11-30 | 北大方正集团有限公司 | Method and device for processing halftone image frequency modulation mesh hanging |
| CN102263881B (en) * | 2010-05-24 | 2013-08-07 | 北大方正集团有限公司 | Method and device for processing halftone image frequency modulation mesh hanging |
| CN102023475A (en) * | 2010-10-13 | 2011-04-20 | 北京方嘉彩色印刷有限责任公司 | High-precision frequency-modulated screen dot printing process |
| CN102023475B (en) * | 2010-10-13 | 2012-07-04 | 北京方嘉彩色印刷有限责任公司 | High-precision frequency-modulated screen dot printing process |
| CN104417063A (en) * | 2013-08-21 | 2015-03-18 | 北大方正集团有限公司 | Method and apparatus for reducing influences of sleeving alignment errors |
| CN104417063B (en) * | 2013-08-21 | 2016-03-23 | 北大方正集团有限公司 | A kind ofly reduce the method and device of overlapping neat error effect |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1144153C (en) | 2004-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100338936C (en) | Double-feedback error diffusion based frequency-modulation screening method | |
| CN100352258C (en) | Method of generating threshold matrix and recording medium storing data of threshold matrix | |
| CN1239262A (en) | Image processing method and apparatus | |
| CN1144153C (en) | Frequency modulation gridding method for improving quality value of F.M. grid medium tone | |
| Li et al. | Tone-dependent error diffusion | |
| CN1114889C (en) | Texture classification apparatus employing coarseness and directiviety of patterns | |
| CN1172263C (en) | Frequency modulation Internet access method for copying images on multiple position imaging depth equipment | |
| CN103402042A (en) | Halftone image generation method based on sectioned error diffusion | |
| CN1207893C (en) | Image processing method and apparatus | |
| CN1774029A (en) | Frequency-modulated and amplitude modulation mixed half tone image processing method on multiposition imaging depth equipment | |
| CN100377567C (en) | FM and AM mixed net point net shape controlling method in multi-position imaging depth device | |
| WO2008028409A1 (en) | Frequency modulation amplitude modulation mixed net point hierarchy continuous adjustment control method | |
| CN1199434C (en) | Multiple stage rasterization with rasters of any raster width and raster angle | |
| Liu et al. | Clustered-dot screen design for digital multitoning | |
| CN108810314A (en) | Unordered error based on the enhancing of multi-grey image edge spreads digital halftoning method | |
| CN1668062A (en) | Method and apparatus capable of producing frequency-modulation halftone dot in high speed | |
| CN100344146C (en) | Signal processing method, signal processing apparatus, signal output apparatus, image processing apparatus, and image forming apparatus | |
| CN1167981C (en) | Frequency-medulating screening method for printer with different length and breadth resolution | |
| US8289577B2 (en) | Method for halftone image transformation, printing and halftone mask generation | |
| CN102158630B (en) | Image-tone-based adaptive green-noise screening method | |
| KR100425886B1 (en) | Two-dimensional modulation for line screen printing | |
| Abedini et al. | 3D halftoning based on iterative method controlling dot placement | |
| CN1909592A (en) | Image scan processing method and system for error diffusion image screening technology | |
| CN1941840A (en) | Method and device for generating multi-site modulating web site simultaneouslly | |
| CN1651253A (en) | Method of performing image grating by random using multi primitire object |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CX01 | Expiry of patent term | ||
| CX01 | Expiry of patent term |
Granted publication date: 20040331 |