Color image processing device with error-diffusion binarization function

• US 5,708,728 A
• Filed: 05/16/1996
• Issued: 01/13/1998
• Est. Priority Date: 05/16/1995
• Status: Expired due to Term

• Current Assignee: Brother Kogyo Kabushiki Kaisha
• Original Assignee: Brother Kogyo Kabushiki Kaisha
• Inventors: Nomura, Mayumi
• Primary Examiner(s): Coles, Sr., Edward L.
• Assistant Examiner(s): Williams, Kimberly A.
• Application Number:
  US08/648,579
• Time in Patent Office: 607 Days
• Field of Search: 358/429, 358/443, 358/447, 358/448, 358/453, 358/455, 358/456, 358/458, 358/459, 358/460, 358/465, 358/518, 358/529, 358/530, 358/534, 382/162, 382/166, 382/167, 382/251, 382/252, 382/254
• US Class Current: 382/162
• CPC Class Codes: H04N 1/52 Circuits or arrangements for halftone screening

Abstract

Input chromatic densities IC, IM, and IY are first converted into corrected chromatic densities I'"'"'C, I'"'"'M, and I'"'"'Y based on the input density IK. The input density IK is then added with the corresponding weighted error sum EK, so that a modified density IKmod is obtained. Similarly, the corrected chromatic densities I'"'"'C, I'"'"'M, and I'"'"'Y are added with the weighted error sums EC, EM, and EY, respectively, so that modified densities ICmod, IMmod, and IYmod are obtained. Then, the modified density IKmod is compared with the threshold T, and an output OK is determined. When the obtained output OK is zero, the ICmod, IMmod, and IYmod are compared with the threshold T, and output data OC, OM, and OY are obtained. On the other hand, when the obtained output OK is a dot-forming data, the output data OC, OM, and OY are forcibly set to a non-dot forming data.

Claims

1. A color image processing device for binarizing a color continuous tone image to produce color pseudo-continuous tone image data, the device comprising:

modification means for modifying input density data of each pixel for each of a plurality of color components with a corresponding sum of binarization errors which are distributed to the subject pixel from at least one preceding nearby pixel, the plurality of color components including a predetermined color component and at least one reference color component;

predetermined color output determination means for comparing the modified input density of the subject pixel for the predetermined color component with a predetermined threshold value, thereby determining binarized output data for the predetermined color component, the binarized output data being either one of a dot-forming data and an on-dot forming data, the dot-forming data indicating production of a dot of the predetermined color component on the subject pixel, the non-dot forming data indicating production of no dots of the predetermined color component on the subject pixel;

reference color output determination means for comparing, when the binarized output data for the predetermined color component is a non-dot forming data, the modified input density of the subject pixel for the reference color component with the predetermined threshold value, thereby determining binarized output data for the reference color component, the binarized output data being either one of a dot-forming data and a non-dot forming data, the dot-forming data indicating production of a dot of the reference color component on the subject pixel, the non-dot forming data indicating production of no dots of the reference color component on the subject pixel, the reference color output determination means determining binarized output data for the reference color component as the non-dot forming data when the binarized output data for the predetermined color component is a dot-forming data;

binarization error determining means for determining a binarization error for the predetermined color component as a difference between the modified input density data of the predetermined color component and the binarized output data for the predetermined color component and for determining a binarization error for the reference color component as a difference between the modified input density data of the reference color component and the binarized output data for the reference color component; and

means for distributing the binarization errors of the predetermined color component and of the reference color component to at least one subsequent nearby pixel which has not yet been binarized.

2. A color image processing device of claim 1, further comprising reference color conversion means for converting the input density data of the reference color component into a converted input density data based on the input density data of the predetermined color component, the modification means modifying the converted input density data of the subject pixel for the reference color component with the corresponding sum of binarization errors distributed to the subject pixel from the at least one preceding nearby pixel.

3. A color image processing device of claim 2, wherein the reference color conversion means calculates, as the converted input density data, a product of the input density data of the reference color component and a ratio of a difference between a maximum inputtable density data and the input density data of the predetermined color component relative to the maximum inputtable density data.

4. A color image processing device of claim 3, wherein the predetermined color component is black, the reference color conversion means converting the input density data of the reference color component into data which represents a chromatic density degree of the reference color component.

5. A color image processing device of claim 3, wherein the modification means addes, to input density data of each pixel for the predetermined color component, with a sum of binarization errors for the predetermined color component which are distributed to the subject pixel, the modification means adding, to the value calculated by the reference color conversion means for the reference color component of each pixel, with a sum of binarization errors for the reference color component which are distributed to the subject pixel.

6. A color image processing device of claim 1, wherein the predetermined color component is black, and the at least one reference color component is cyan, magenta, and yellow.

7. A color image processing device of claim 1, wherein the binarization error distributing means distributes the binarization errors of the predetermined color component and of the reference color component, according to a predetermined weight set, to the at least one subsequent nearby pixel.

8. A color image processing device of claim 7, further comprising accumulating means assigned to each pixel for each color component, the accumulating means accumulating the binarization errors produced at the at least one preceding nearby pixel and distributed to the subject pixel, thereby obtaining the sum of the binarization errors for each color component of each pixel.

9. A color image processing device of claim 1, wherein the modification means addes, to input density data of each object pixel for each of a plurality of color components, a corresponding sum of binarization errors which are distributed to the subject pixel from at least one preceding nearby pixel.

10. A color image processing device of claim 1, further comprising means for receiving input density data for the plurality of color components of each pixel.

11. A method of binarizing a color continuous tone image to produce color pseudo-continuous tone image data, the method comprising the steps of:

modifying input density data of one pixel for each of a plurality of color components with a corresponding sum of binarization errors which are distributed to the subject pixel during error-diffusion binarization processes previously performed for at least one preceding nearby pixel, the plurality of color components including a predetermined color component and at least one reference color component;

comparing the modified input density of the subject pixel for the predetermined color component with a predetermined threshold value, thereby determining binarized output data for the predetermined color component, the binarized output data being either one of a dot-forming data and non-dot forming data, the dot-forming data indicating production of a dot of the predetermined color component on the subject pixel, the non-dot forming data indicating production of no dots of the predetermined color component on the subject pixel;

comparing, when the binarized output data for the predetermined color component is a non-dot forming data, the modified input density of the subject pixel for the reference color component with the predetermined threshold value, thereby determining binarized output data for the reference color component, the binarized output data being either one of a dot-forming data and a non-dot forming data, the dot-forming data indicating production of a dot of the reference color component on the subject pixel, the non-dot forming data indicating production of no dots of the reference color component on the subject pixel;

determining binarized output data for the reference color component as the non-dot forming data when the binarized output data for the predetermined color component is a dot-forming data;

determining a binarization error for the predetermined color component as a difference between the modified input density data of the predetermined color component and the binarized output data for the predetermined color component and determining a binarization error for the reference color component as a difference between the modified input density data of the reference color component and the binarized output data for the reference color component; and

distributing the binarization errors of the predetermined color component and of the reference color component to at least one subsequent nearby pixel which has not yet been binarized.

12. A method of claim 11, further comprising the step of converting the input density data of the reference color component into a converted input density data based on the input density data of the predetermined color component, the converted input density data of the subject pixel for the reference color component being modified with the corresponding sum of binarization errors distributed to the subject pixel from the at least one preceding nearby pixel.

13. A method of claim 12, wherein the converted input density data is calculated as a product of the input density data of the reference color component and a ratio of a difference between a maximum inputtable density data and the input density data of the predetermined color component relative to the maximum inputtable density data.

14. A method of claim 13, wherein the predetermined color component is black, the converted input density data of the reference color component represents a chromatic density degree of the reference color component.

15. A method of claim 13, wherein the input density data of the subject pixel for the predetermined color component is added with a sum of binarization errors for the predetermined color component which are distributed to the subject pixel, and wherein the converted input density data for the reference color component of the subject pixel is added with a sum of binarization errors for the reference color component which are distributed to the subject pixel.

16. A method of claim 11, wherein the predetermined color component is black, and the at least one reference color component is cyan, magenta, and yellow.

17. A method of claim 11, wherein the binarization errors of the predetermined color component and of the reference color component are distributed, according to a predetermined weight set, to the at least one subsequent nearby pixel.

18. A method of claim 17, further comprising the step of adding the binarization errors produced at the subject pixel and distributed to at least one subsequent pixel, thereby obtaining the sum of the binarization errors for each color component of the subject pixel.