CN100341709C - Image processing device for combining image enhancement technique and half tone method - Google Patents

Abstract

The present invention discloses an image processing device which combines an image enhancement technique and a half tone method. A half tone method and an image enhancement technique are combined, and simultaneously, the purposes of processing a half tone and improving image quality are reached. The present invention mainly comprises an image input module, an image enhancement processing module and a half tone processing module. By means of the image input module, an image is directly input into the image enhancement processing module with a state of original image data so as to carry out image enhancement operation which takes filtering as the basis. Then, the half tone processing module is used for carrying out error diffusion operation to the enhanced image data. The original two different processing procedures can be simplified to one mechanism so that the hardware architecture is simple and the memory consumption is reduced.

Description

Strengthen the image processor of technology and half tone method in conjunction with image

Technical field

The present invention relates to a kind of image processor, be applied to image output device, particularly relate to the image processor of a kind of while in conjunction with half tone method and image enhancing technology.

Background technology

In general, shown digitized video all is to be mixed according to different ratios by three primary colors-redness (R), green (G), blue (B) on the computer, with 24 images is example, R, G, B are respectively 8, the color range of each primary colors is 0～255, for example R, G, B value are all 0 demonstration black, and R, G, B value are all 255 and represent white.Yet when will exporting, the digitized video on the computer just problem has taken place, because many printings and display unit can only produce the image of binaryzation, therefore polychrome rank video conversion must be become the output of binaryzation image, this conversion method is called halftoning (Halftone).

Halftoning is to utilize human eye the illusion of shade to be caused the sensation on polychrome rank, with the printer is example, with the lattice on the paper is a unit, per unit corresponds to different color range values with the different degree of filling up, the observer sees this grid outside suitable distance, can't see that the light and shade in the grid changes, and lattice can be regarded as an integral body, only can feel the mean flow rate of grid.

According to calculating the number that required processing of pixel of halftoning image is counted, halftone process can be divided into two big classes usually: standalone processes method and neighbor point facture.For the standalone processes method, the output of halftoning is normally passed through with certain shielding (Mask) comparison by each pixel of former image and is got, and this kind method is representative with the method for trembling (Dither Method).For the neighbor point facture, the output of halftoning is not just can be obtained by simple pixel comparison, and needs the processing through filtering (Filter), and this kind method is representative with error-diffusion method (Error Diffusion Method).Because the use error diffusion method can obtain color range performance preferably, when in order to obtain quality preferably during the halftoning image output, error-diffusion method relatively more commonly used, still, the shortcoming of this method is exactly complex calculations, and taking advantage of of several points added just and can finish around the calculating of a pixel needed.

The image on polychrome rank all is just can do halftone process for the characteristic that meets output device usually, because the image output after the halftoning is through downgrading color range, it is good that the quality of image output also just has raw video scarcely, at this moment, if the image quality on original polychrome rank is very poor, as: blur, have noise, the halftoning image of output be it seems will be poorer.Generally speaking, in order to improve this problem, raw video can be strengthened image quality earlier before doing halftone process, and thus, the framework and the computational complexity of algorithm all increase, and required internal memory also increases relatively thereupon.

On the other hand, multifunctional paper feeding machine and photo printer all use the technology of halftoning.In the copy flow process of multifunctional paper feeding machine, a colored file can the directly printout via scanning, this flow process is independently to finish fully, the not processing of process computer, if multifunctional paper feeding machine not image itself strengthens the mechanism of handling, then output quality is dependent on original document fully, in case original document has flaw, printout also just has flaw.Identical situation also occurs in photo printer, general photo printer all has the device that inserts memory card, many image files are arranged in the memory card, the user can select the image that will print from memory card, because this flow process is also without Computer Processing, if the raw video quality is bad, the quality of printout is also just bad.In these cases, the improvement of output quality must be finished at device end, just derives many different technology as for how to improve.

Therefore, existing patent proposes some different solutions, announce the 6th as United States Patent (USP), 424, No. 747 patents, it has designed a smoothing circuit, selects a suitable wave filter by this circuit from the wave filter storage element, comes smoothing in addition by the value that checks in the color transformation table (Color Conversion Table) by wave filter then; Yet this invention directly changes the color representation that color-values probably can have influence on image.Announce the 6th, 201 as United States Patent (USP), No. 613 patents are that the image after the halftoning is handled the effect that reaches smoothing via low pass filter; Because be to carry out smoothing at the image after the halftoning to handle, treatment effect is limited.Announce the 6th as United States Patent (USP), 061, No. 145 patents, doing smoothing at a halftoning image equally handles, it is to detect type sharp keen in the halftoning image (Pattem) earlier, again these sharp keen types are replaced with predefined level and smooth type, the method must be through the processing of two steps: detect and replace, the part that detects can consume many times, and original sharp keen type kind of setting is many more, the time of being detected is also just many more, and required internal memory is also just many more, and impracticable.Announce the 5th as United States Patent (USP), 757, No. 976 patents, its use error diffusion method is reached halftone process, and there is a filter control circuit to select different error diffusion wave filters according to area grayscale values different in the image, reaching different effects, but the change of error-filter only to noise that halftoning produced with to repeat type influential, for the raw video quality just not good situation can't be improved.

Summary of the invention

Technical problem to be solved by this invention provides a kind of image processor in conjunction with image enhancing technology and half tone method, solves the problem that colors of image is performed poor, the internal memory use amount is big of prior art.

For achieving the above object, the invention provides a kind of image processor in conjunction with image enhancing technology and half tone method, its characteristics are, include:

One image input module in order to reading an image, and produces raw video data;

One image strengthens processing module, receives this raw video data, and carries out filtering at these raw video data and handle, and strengthens data and export an image; And

One halftoning processing module receives this image and strengthens data, and carries out error diffusion to handle and export a halftoning image;

This image enhancing processing module is carried out the required pixel of error diffusion processing according to this halftone process module and is carried out image enhancing processing.

The above-mentioned image processor in conjunction with image enhancing technology and half tone method, its characteristics are that it is the pattern of wave filter that this image strengthens processing module.

The above-mentioned image processor in conjunction with image enhancing technology and half tone method, its characteristics are that it is a smoothing processing module that this image strengthens processing module.

The above-mentioned image processor in conjunction with image enhancing technology and half tone method, its characteristics are that it is a sharp keenization processing module that this image strengthens processing module.

The above-mentioned image processor in conjunction with image enhancing technology and half tone method, its characteristics are that this image strengthens processing module and includes:

One multiplicative operator according to a default weighted value of this wave filter, calculates the weighted value of the pixel in these raw video data;

One adder calculator adds up the weighted value of a pending pixel and the pixel of contiguous this pending pixel, adds total value and obtain one; And

One divider adds total value obtains this pending pixel divided by this default weighted value image enhancing data with this.

The above-mentioned image processor in conjunction with image enhancing technology and half tone method, its characteristics are that it is the wave filter of a two dimension that this image strengthens processing module.

Technique effect of the present invention is: the image processor in conjunction with image enhancing technology and half tone method provided by the invention, can handle image by the image input module and to obtain the raw video data, and strengthen processing module by image and directly the raw video data are strengthened processing, carry out halftone process by the halftone process module again after strengthening processing, and then image output.The present invention becomes a mechanism with original two different disposal process simplifications, not only framework is simple, also strengthen processing because of directly the raw video data being carried out image, carry out halftone process again, so, not only can strengthen image, and not change the original contents performance of image, also can finish image simultaneously and strengthen and halftone process, significantly reduce the use amount of internal memory.

Further describe specific embodiments of the invention below in conjunction with accompanying drawing.

Description of drawings

Fig. 1 is that the data of printer are handled schematic diagram; And

Fig. 2 is a configuration diagram of the present invention.

Wherein, description of reference numerals is as follows:

100 print data processing apparatus

110 color conversion mechanism

120 halftone process devices

130 data rrangement devices

140 printing control models

10 image input modules

20 images strengthen processing module

21 multiplicative operators

22 adder calculators

23 dividers

30 halftone process modules

31 quantizers

32 error-filters

The specific embodiment

Disclosed image processor in conjunction with image enhancing technology and half tone method, be mainly used in image output device, such as printer, multifunctional paper feeding machine etc., see also Fig. 1,100 is the data processing equipment of printer or multifunctional paper feeding machine, includes color conversion mechanism 110, halftone process device 120, data rrangement device 130 and printing control model 140.The image of wanting to print can exist with the data kenel of three primary colors (red bluish-green RGB), at first, image inputs to color conversion mechanism 110 conversion colour gamut coordinates, be converted to the printing color from three primary colors, halftone process device 120 becomes the halftoning image output with the image depression of order on polychrome rank then, image output is arranged in via data rrangement device 130 and prints required data kenel, with the ink-jet printer is example, this step is exactly the form that the halftoning image output is arranged in jet orifice, at last, printing control model 140 control printouts are to media.

Disclosed image processor promptly is to replace original halftone process device 120, see also Fig. 2, include image input module 10, image enhancing processing module 20 and halftone process module 30, at first be converted into raw video data I [m, n] and directly be delivered to the processing that image enhancing processing module 20 is carried out the image enhancing by the image of image input module 10 with input.

Image enhancing processing module 20 mainly is the kenel with wave filter, and its algorithm summary is as follows:

$O[m,n]=\underset{k,r}{\mathrm{\Σ}}I[m-k,n-r]\×a[k,r]$

I[m wherein, n] be raw video data, O[m, n] be that image strengthens data, a[k, r] be wave filter, its situation can be handled (seeing the following form) for smoothing

1/9	1/9	1/9
			1/9	1/9	1/9
1/9	1/9	1/9

Or sharp keenization processing (seeing the following form)

0	1	0
			1	1	-1
0	-1	0

No matter be the wave filter of which kind of tupe, centre italics person is a pending pixel in the raw video data in principle, utilize multiplicative operator 21 that pending pixel and adjacent pixels be multiply by default weighted value (numeral in the table) and obtain weighted value, by adder calculator 22 weighted value of pending pixel and neighbor is added up then and obtain adding total value, utilize divider 23 will add total value divided by all default weighted value summations at last, the image that just can obtain pending pixel strengthens data.

After handling all pixels in regular turn, again image is strengthened data O[m, n] deliver to halftone process module 30 and carry out halftone process.Shown in following algorithm:

$O*[m,n]=O[m,n]+\underset{k,r}{\mathrm{\Σ}}E[m-k,n-r]\×a[k,r]$

E[m，n]＝O*[m，n]-B[m，n]

$B[m,n]=\left\{\begin{array}{c}1,O*[m,n]\&GreaterEqual;\frac{2(D-1)-1}{2(D-1)}\\ \frac{D-2}{D-1},\frac{2(D-1)3}{2(D-1)}\&le;O*[m,n]<\frac{2(D-1)-1}{2(D-1)}\\ .\\ .\\ .\\ \frac{2}{D-1},\frac{3}{2(D-1)}\&le;O*[m,n]<\frac{5}{2(D-1)}\\ \frac{1}{D-1},\frac{1}{2(D-1)}\&le;O*[m,n]<\frac{3}{2(D-1)}\\ 0,O*[m,n]<\frac{1}{2(D-1)}\end{array}\right.$

Wherein image strengthens data O[m, n], its scope is 0 (White)～1 (Black) normally, and B[m, n] be the output of quantizer 31, it has D in various degree, be respectively 0,

... 1, the critical value in the quantizer 31 (Threshold) can be fixed on particular value, if critical value is a five equilibrium, then is

E[m, n] be error signal, the error of Liang Huaing just, this value is get error signal E[m, n by the signal subtraction before and after quantizing] and producing correction signals through error-filter 32 backs, this correction signal is the input that is used for proofreading and correct future.O*[m, n] be the halftoning image output after proofreading and correct, a[k, r] be error-filter 32 (value in the wave filter is the weight of error signal, the transmission direction of [k, r] expression error signal).

Then, in conjunction with above-mentioned algorithm, can obtain:

$E[m,n]=O[m,n]-B[m,n]+\underset{k,r}{\mathrm{\&Sigma;}}E[m-k,n-r]\&times;a[k,r]$

Then again it is converted into frequency domain representation, can obtains:

E[z ₁，z ₂]＝[O[z ₁，z ₂]-B[z ₁，z ₂]]H[z ₁，z ₂]

Therefore, this error-filter 32 is linear systems of full limit as can be known.

In addition, the situation of common error-filter 32 has Floyd and Steinberg (seeing the following form):

	*	7/16
			3/16	5/16	1/16

Jarvis, Judice and Ninke (seeing the following form)

		*	7/48	5/48
					3/48	5/48	7/48	5/48	3/48
1/48	3/48	5/48	3/48	1/48

Stucki (seeing the following form)

		*	8/42	4/42
					2/42	4/42	8/42	4/42	2/42
1/42	2/42	4/42	2/42	1/42

Stevenson and Arce (seeing the following form)

			*		32/200
							12/200		26/200		30/200		16/200
	12/200		26/200		12/200
							5/200		12/200		12/200		5/200

Wherein required pixel of carrying out error diffusion is represented at the * place.

Therefore, the algorithm in conjunction with above-mentioned image enhancing processing module 20 and halftone process module 30 can obtain:

$O*[m,n]=\underset{p,q}{\mathrm{\&Sigma;}}O[m-p,n-q]\&times;a[p,q]+\underset{k,r}{\mathrm{\&Sigma;}}E[m-k,n-r]\&times;c[k,r]$

E[m，n]＝O*[m，n]-B[m，n]

$B[m,n]=\left\{\begin{array}{c}1,O*[m,n]\&GreaterEqual;\frac{2(D-1)-1}{2(D-1)}\\ \frac{D-2}{D-1},\frac{2(D-1)3}{2(D-1)}\&le;O*[m,n]<\frac{2(D-1)-1}{2(D-1)}\\ .\\ .\\ .\\ \frac{2}{D-1},\frac{3}{2(D-1)}\&le;O*[m,n]<\frac{5}{2(D-1)}\\ \frac{1}{D-1},\frac{1}{2(D-1)}\&le;O*[m,n]<\frac{3}{2(D-1)}\\ 0,O*[m,n]<\frac{1}{2(D-1)}\end{array}\right.$

Below lift an application examples result of the present invention is described.The data of supposing image are 8, and in other words, the excursion of image input value is between 0-255 (seeing the following form).

120 101 105 101 96 94 80 72 77 79 84 86 83 72 102 118 131 166 189 186 110 73 102 121 106 92 88 57 61 130 114 77 138 56 53 88 167 184 143 192 127 97 107 124 87 80 88 118 132 173 182 120 184 204 162 165 198 162 94 187 129 120 120 94 84 78 85 140 167 172 206 209 200 192 230 203 192 177 86 182 183 130 123 87 78 78 72 69 72 76 100 162 112 209 215 200 192 185 105 136 192 179 108 95 82 75 74 75 76 77 86 108 205 217 202 187 166 188 154 79 102 203 220 167 80 80 90 67 75 77 85 209 199 211 197 166 92 181 181 107 87 109 126 190 178 174 168 65 43 80 135 216 191 210 203 206 208 185 167 161 98 70 105 90 130 121 71 90 111 139 207 210 213 216 188 161 188 184 156 154 122 62 87 132 150 174 183 178 165 181 217 186 208 186 145 147 169 191 170 156 108 91 129 112 124 99 72 76 80 213 194 201 186 168 149 157 200 162 59 143 44 100 185 86 61 56 56 72 138 224 175 203 161 160 151 180 178 62 105 179 63 83 147 172 70 54 41 121 209 204 201 185 172 163 220 164 52 93 129 143 80 56 70 152 173 122 167 186 113 112 208 174 180 183 224 178 119 129 116 116 106 69 62 76 139 186 116 74 65 64 147 156 167 164 166 199 229 166 125 106

115 103 181 108 161 52 50 62 61 72 116 159 132 121 126 132 163 213 170 117 112 78 152 206 201 66 41 65 65 80 109 200 159 100 98 115 132 135 132 130 117 68 65 120 189 174 77 56 72 97 121 149 204 110 43 126 126 133 126 129 126 102 103 187 127 118 208 165 123 93 116 123 121 49 67 105 102 106 108 143 118 176 100 153 56 55 53 133 190 89 93 114 118 110 98 83 85 91 95 143

The output of halftoning then is 1 position: 0 or 1, and the critical value in the quantizer is decided to be 128, just imports less than 128 o'clock, and quantizer is output as 0; Input was more than or equal to 128 o'clock, and the output of quantizer then is 1.Image strengthens wave filter situation such as the following table of handling:

1/13	1/13	1/13	1/13	1/13
					1/13	1/13	1/13	1/13	1/13
1/13	1/13	1/13

And halftoning is reached with error-diffusion method, and the error weight wave filter that uses is Jarvis, Judice andNinke error-filter, and pixel (3,3) is as follows as calculated:

$O^{*}\left[\mathrm{3,3}\right]=\frac{1}{13}\left\{O\right[\mathrm{1,1}]+O[\mathrm{2,1}]+O[\mathrm{3,1}]+O[\mathrm{4,1}]+O[\mathrm{5,1}]+O[\mathrm{1,2}]+O[\mathrm{2,2}]$

$+O\left[\mathrm{3,2}\right]+O\left[\mathrm{4,2}\right]+O\left[\mathrm{5,2}\right]+O\left[\mathrm{1,3}\right]+O\left[\mathrm{2,3}\right]+O\left[\mathrm{3,3}\right]\}+E[\mathrm{1,1}]$

$+E\left[\mathrm{2,1}\right]+E\left[\mathrm{3,1}\right]+E\left[\mathrm{4,1}\right]+E\left[\mathrm{5,1}\right]+E\left[\mathrm{1,2}\right]+E\left[\mathrm{2,2}\right]+E\left[\mathrm{3,2}\right]$

$+E\left[\mathrm{4,2}\right]+E\left[\mathrm{5,2}\right]+E\left[\mathrm{1,3}\right]+E\left[\mathrm{2,3}\right]$

$=\frac{1}{13}\{120+101+105+101+96+110+73+102+121+106+127$

$+97+107\}+(120-0)+(119-0)+(135-255)+(96-0)$

$+(98-0)+(132-255)+(78-0)+(119-0)+(173-255)$

$+(142-255)+(142-255)+(101-0)$

$=105+120+119-120+96+98-123+78+119-82-113-113$

$+101$

$=116<128$

∴B[3，3]＝0

Calculate the halftoning image data (seeing the following form) that can obtain whole image in regular turn.

0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 1 1 0 0 1 1 0 1 0 0 1 1 0 1 0 0 0 1 1 0 1 1 0 0 1 0 0 1 1 0 1 1 0 1 1 0 1 0 0 0 1 0 1 1 0 0 0 0 0 1 0 0 1 1 0 1 1 1 1 0 1 1 0 1 0 1 1 0 1 1 0 1 1 0 1 1 1 1 1 0 0 1 1 0 0 0 1 0 0 0 1 0 1 1 1 0 1 0 1 1 0 0 1 1 1 0 0 0 0 0 0 0 0 1 1 1 0 1 0 1 0 0 0 1 1 0 1 1 1 0 1 1 0 1 1 1 1 1 1 1 1 0 0 1 0 1 0 0 0 1 0 1 1 1 0 0 1 0 0 1 0 1 0 0 1 0 1 1 0 1 0 1 1 1 1 1 1 1 1 1 0 0 0 1 0 0 1 1 0 1 1 0 1 1 0 1 1 0 1 0 1 0 1 1 0 1 0 0 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 1 0 0 0 1 0 1 1 1 1 1 1 1 1 0 1 1 0 0 0 1 0 1 1 0 1 0 1 0 0 1 0 1 0 1 0 0

1 0 1 0 0 1 1 0 1 0 1 1 1 1 0 1 1 0 1 0 0 1 0 1 1 0 0 0 0 1 0 0 1 1 0 1 1 0 1 1 0 0 1 0 1 0 0 1 0 0 1 0 0 1 0 1 0 1 0 0 1 0 1 0 0 1 1 0 0 1 1 0 1 1 0 0 1 1 0 1 0 0 1 1 0 1 0 0 1 0 0 1 0 1 0 0 1 0 0 1 1 1 0 1 0 1 0 1 1 0 0 1 0 0 1 0 0 1 0 1

The calculating of pixel has just been finished smoothing processing and halftoning in a program, and pixel and halftoning that the smoothing processing uses are identical, at most only need deposit the pixel values of 13 pixels in the wave filter in the internal memory.Therefore significantly reduce the use amount of internal memory.

The above is preferred embodiment of the present invention only, is not to be used for limiting practical range of the present invention; Be that all equalizations of doing according to content of the present invention change and modification, be all scope of the present invention and contain.

Claims (5)

1, a kind of image processor in conjunction with image enhancing technology and half tone method is characterized in that, includes:

One image input module in order to reading an image, and produces raw video data;

One image strengthens processing module, receives this raw video data, and carries out filtering at these raw video data and handle, and strengthens data and export an image, and this image strengthens processing module and is filter form and includes:

One multiplicative operator, an adder calculator and a divider, this multiplicative operator is according to the default weighted value of this wave filter, calculate the weighted value of the pixel in these raw video data, the weighted value that this adder calculator is used to add up a pending pixel and is close to the pixel of this pending pixel, add total value and obtain one, this divider is used for this is added total value obtains this pending pixel divided by the summation of all these default weighted values image enhancing data; And

One halftoning processing module, receive this image and strengthen data, and carry out error diffusion to handle and export a halftoning image, this halftone process module comprises a quantizer and an error-filter, this error-filter receives that this quantizer subtracts each other before and after quantizing and the error signal that obtains, and produce a correction signal according to this error signal, this quantizer receives this image to be strengthened data and this correction signal and carries out quantification treatment, to export this halftoning image, this quantizer has a plurality of different output levels according to a critical value of setting;

Wherein this image enhancing processing module is to carry out the required pixel of error diffusion processing according to this halftone process module to carry out image enhancing processing.

2, the image processor in conjunction with image enhancing technology and half tone method according to claim 1 is characterized in that it is a smoothing processing module that this image strengthens processing module.

3, the image processor in conjunction with image enhancing technology and half tone method according to claim 1 is characterized in that it is a sharp keenization processing module that this image strengthens processing module.

4, the image processor in conjunction with image enhancing technology and half tone method according to claim 1 is characterized in that it is the wave filter of a two dimension that this image strengthens processing module.

5, the image processor in conjunction with image enhancing technology and half tone method according to claim 1 is characterized in that the following formula of described halftone process module utilization quantizes:

$O*[m,n]=O[m,n]+\underset{k,r}{\mathrm{\&Sigma;}}E[m-k,n-r]\&times;a[k,r]$

E[m，n]＝O*[m，n]-B[m，n]

$B[m,n]=\left\{\begin{array}{c}1,O*[m,n]\&GreaterEqual;\frac{2(D-1)-1}{2(D-1)}\\ \frac{D-2}{D-1},\frac{2(D-1)-3}{2(D-1)}\&le;O*[m,n]<\frac{2(D-1)-1}{2(D-1)}\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ \frac{2}{D-1},\frac{3}{2(D-1)}\&le;O*[m,n]<\frac{5}{2(D-1)}\\ \frac{1}{D-1},\frac{1}{2(D-1)}\&le;O*[m,n]<\frac{3}{2(D-1)}\\ 0,O*[m,n]<\frac{1}{2(D-1)}\end{array}\right.$

Wherein, O[m, n] be that described image strengthens data, B[m, n] be the halftoning image data of described quantizer output, D is the critical value of described quantizer, E[m, n] be described error signal, O*[m, n] be the halftoning image data after proofreading and correct, a[k, r] be described error-filter.

Images