Method for processing FM-AM mixed halftone images on a multi-bit depth imaging apparatus

• US 8,035,860 B2
• Filed: 04/29/2006
• Issued: 10/11/2011
• Est. Priority Date: 10/26/2005
• Status: Active Grant

• Current Assignee: New Founder Holdings Development Limited Liability Company, Beijing Founder Electronics Company Limited (Peking University), Peking University
• Original Assignee: Peking University Founder Group Company Limited (Beijing Peking University Asset Management Company Limited), Beijing Founder Electronics Company Limited (Peking University), Peking University
• Inventors: Yang, Bin, Li, Haifeng
• Primary Examiner(s): Park; Chan S
• Assistant Examiner(s): Nguyen; Allen H
• Application Number:
  US12/091,290
• Publication Number:
  US 20080278765A1
• Time in Patent Office: 1,991 Days
• Field of Search: 358/3.06, 358/3.16, 358/3.14, 358/3.19
• US Class Current: 358/3.16
• CPC Class Codes: H04N 1/40087 Multi-toning, i.e. converting a continuous-tone signal for reproduction with more than two discrete brightnesses or optical densities, e.g. dots of grey and black inks on white paper
• CPC Class Codes: H04N 1/4053 with threshold modulated relative to input image data or vice versa

Abstract

Disclosed is a method for processing FM-AM mixed halftone images on a multi-bit depth imaging apparatus, which relates to a method for producing halftone dots in the field of image hard copying. In the prior art, since it is hard to avoid the impact of the error diffusion for the output apparatus to control the mixed dots with multi-bit imaging depth based on the error diffusion, the output of the mixed dots with multi-bit imaging depth cannot satisfy requirements of the apparatus. According to the method of the present invention, the dynamic algorithm for controlling the multi-bit mixed dots is used for screening based on the existing mixed screening process using dual-feedback error diffusion. Furthermore, multi-bit halftone images with high quality and rich gradations can be output by the multi-bit depth imaging apparatus. The method of the present invention can solve the phenomenon of sawtooth in the margins of the mixed dots output by the conventional single-bit apparatus and obtain the FM-AM mixed dots with the effect of high resolution and continuous gradations, which are output under low resolution.

Claims

1. A method for processing FM-AM mixed halftone images on a multi-bit depth imaging apparatus, comprising:

1) dividing averagely an interval [0, 255] into 2ⁿ+1 gradations in light of a bit depth n of the apparatus;

[0, R₁], (R₁,R₂], . . . , (R_i-1, R_i], . . . (R_2-2ⁿ, 255], wherein

i is a positive integer and less than 2ⁿ−

2, corresponding ranges of a dot-matrix of bit-outputting are

(0,Out₁), (Out₁,Out₂), . . . , (Out_i-1,Out_i), . . . (Out_2-2ⁿ,11 . . .

     1),

the Out_i being a binary representation of the n-bit depth, and

a threshold M_i of a central point of each of the gradations is sampled as a threshold comparison parameter for the gradation;

2) setting output probability thresholds with a n-bit imaging depth in the interval [0, 255];

3) processing respectively the dots in the 2ⁿ−

1 gradations (R_i-1, R_i) based on an FM-AM mixed screening process using a dual-feedback error diffusion; and

4) computing dynamically output dot-matrix data, by using a dynamic gradation-changeable output mechanism with adjacent output gray levels in light of the probability thresholds F_i and an accumulated value ShapeCur for controlling the shape of a current dot, when mixed screening in the gradations (R_i-1, R_i) is achieved.

2. The method of claim 1, wherein the step 2) further comprises

setting 2ⁿ−

2 integers L₁, L₂, . . . , L_i, . . . , L_2-2ⁿ as the output probability thresholds with the n-bit imaging depth, and

adjusting the probability thresholds based on requirements of the apparatus.

3. The method of claim 1, wherein the step 3) further comprises:

(a) carrying out an operation T of threshold comparison on a final input value g″

(m, t) of a current pixel of an original image, and then converting a result of the operation to a corresponding value b(m, t) of the current pixel for a halftone image;

(b) comparing the value b(m, t) of the current pixel with an input value g′

(m, t) of the current pixel to obtain a difference between b(m, t) and g′

(m, t), wherein the difference is an error value e(m, t), and the input value g′

(m, t) is used for the operation T of threshold comparison;

(c) multiplying the error value e(m, t) by preset weight distribution coefficients through an error diffusion filter e and then diffusing results of the multiplying to unprocessed pixels around the current pixel, wherein each of the diffused results to the unprocessed pixels around the current pixel is weightedly added to an original input value g(m, t) of the corresponding pixel of the original image to obtain an input value g′

(m, t) of the corresponding pixel of the original image;

(d) diffusing processed results to corresponding unprocessed pixels surrounding the current pixel, respectively, and weightedly adding each of the diffused processed results to the original input value g(m, t) of the corresponding pixel of the original image to obtain the final input value g″

(m, t) of the corresponding pixel, wherein the processed results are obtained by implementing a multiplying operation on the output value b(m, t) of the current pixel using a diffusion filter w which has been processed with a dithering algorithm, and the step (d) is implemented in parallel with the steps (b) and (c); and

(e) repeating the steps (a)-(d) until the original input values g(m, t) of all pixels are processed.

4. The method of claim 3, wherein the step (a) uses a process of bidirectional scanning when the original image is scanned, wherein, when a certain row is scanned from left to right, a next row is subsequently scanned from right to left.

5. The method of claim 4, wherein the error diffusion filter e uses a diffusion principle and a weight distribution mode as below:

6. The method of claim 5, wherein a diffusion mode of the diffusion filter w is set as:

7. The method of claim 6, wherein the dithering algorithm for the diffusion filter w in the step (d) are as below:

fRand=(R(m,t)/R_MAX−

0.5)×

cDither

dw₀=w₀−

fRand

dw₂=w₂+fRand

dw₁=w₁+fRand

dw₃=w₃−

fRand

wherein, fRand is a parameter for fine adjusting dithering;

R(m, t) is a parameter with random value for the current dot;

being scanned;

R_MAX is a maximum of a random parameter R(i);

cDither is a parameter for adjusting amplitude of dithering; and

dw₀˜

dw₃ are the diffusion weight coefficients of the diffusion filter w in different directions after dithering.

8. The method of claim 1, wherein the dynamic gradation-changeable output mechanism in the step 4) comprises:

generating a pseudo-random value of the current dot in light of ShapeCurr;

F_i=random(ShapeCur) 

 

Formula 1

wherein, the pseudo-random function random is to be generated automatically in a compiling environment, F_iε

[0, 255]; and

the output dot-matrix data are computed dynamically;

9. The method of claim 2, wherein the dynamic gradation-changeable output mechanism in the step 4) comprises:

generating a pseudo-random value of the current dot in light of ShapeCur;

F_i=random(ShapeCur) 

 

Formula 1

wherein, the pseudo-random function random is to be generated automatically in a compiling environment, F_iε

[0, 255]; and

the output dot-matrix data are computed dynamically;

10. The method of claim 3, wherein the dynamic gradation-changeable output mechanism in the step 4) comprises:

generating a pseudo-random value of the current dot in light of ShapeCur;

F_i=random(ShapeCur) 

 

Formula 1

wherein, the pseudo-random function random is to be generated automatically in a compiling environment, F_iε

[0, 255]; and

the output dot-matrix data are computed dynamically;

11. The method of claim 7, wherein the dynamic gradation-changeable output mechanism in the step 4) comprises:

generating a pseudo-random value of the current dot in light of ShapeCur;

F_i=random(ShapeCur) 

 

Formula 1

wherein, the pseudo-random function random is to be generated automatically in a compiling environment, F_iε

[0, 255]; and

the output dot-matrix data are computed dynamically;