Electronic high-fidelity screenless conversion system and method using a separable filter

• US 5,335,089 A
• Filed: 07/29/1992
• Issued: 08/02/1994
• Est. Priority Date: 07/29/1992
• Status: Expired due to Fees

• Current Assignee: Graphic Arts Technical Foundation Incorporated A Not-For-Profit Corporation of The State of Delaware
• Original Assignee: RR Donnelley & Sons Company
• Inventors: Xie, Zhenhua, Rodriguez, Michael A.
• Primary Examiner(s): Powell, Mark R.
• Assistant Examiner(s): Cohen, Cheryl
• Application Number:
  US07/921,349
• Time in Patent Office: 734 Days
• Field of Search: 358/448, 358/455, 358/456, 358/457, 358/458, 358/443, 358/296, 358/298, 382/50, 382/54
• US Class Current: 358/3.03
• CPC Class Codes: H04N 1/4052 by error diffusion, i.e. transferring the binarising error to neighbouring dot decisions

Abstract

A system for converting a continuous tone value to an output value having either a first level or a second level develops a first error based upon the application of a separable filter to an output value having a first level assumed for the continuous tone value undergoing conversion. A second error is developed based upon the application of the separable filter to an output value having a second level assumed for the continuous tone value undergoing conversion. If the first error is greater than the second error, the continuous tone value is converted to the output value having the second level. Otherwise, the continuous tone value is converted to the output value having the first level. The resulting error is diffused in a predetermined manner to neighboring continuous tone values yet to be converted.

Claims

1. A system for converting one input value of a series of input values into an output value wherein previous input values have been converted to previous output values and subsequent input values are to be converted to output values, the system comprising:

error generating means for generating first and second errors for the one input value undergoing conversion, wherein the first error is based upon application of a one dimensional separable filter to a selected number of previous output values, to output values predicted for a certain number of subsequent input values, and to an output value having an assumed first level for the one input value undergoing conversion, and wherein the second error is based upon application of the one dimensional separable filter applied to the selected number of previous output values, to the output values predicted for the certain number of subsequent input values, and to an output value having an assumed second level for the one input value undergoing conversion; and

output value setting means for setting the one input value undergoing conversion to an output value having the first level if the first error is less than the second error and to an output value having the second level if the first error is greater than the second error.

2. The system of claim 1 wherein the output value setting means comprises error diffusing means for diffusing the first error to a selected number of the subsequent input values if the one input value undergoing conversion is converted to the output value having the first level and for diffusing the second error to a selected number of the values if the one input value undergoing conversion is converted to the output value having the second level.

3. The system of claim 2 wherein the error generating means includes comparing means for comparing each of the certain number of subsequent input values to a threshold, converting means responsive to the comparing means for converting each of the certain number of subsequent input values to a corresponding predicted output value having the first level or the second level depending upon whether the subsequent input value is above or below the threshold, and difference diffusing means for diffusing a difference between each of the certain number of subsequent input values and the predicted output value corresponding thereto to other subsequent input values.

4. The system of claim 3 wherein the threshold is a random threshold.

5. The system of claim 3 wherein the difference diffusing means diffuses the difference in a serpentine manner.

6. The system of claim 1 further including sampling means for sampling an original at a first resolution to obtain sampled values and wherein the error generating means further includes digital resampling means responsive to the sampling means for increasing the resolution of the sampled values to a second resolution greater than the first resolution to obtain the series of input values.

7. The system of claim 6 wherein the sampled values have a certain bandwidth and wherein the separable filter is of the low pass type that limits the output values to the certain bandwidth.

8. A system for sequentially converting input values into corresponding output values so that, at a given instant in time, an input value undergoing conversion has before it input values previously converted to corresponding output values and after it subsequent input values yet to be converted to output values, the system comprising:

a separable filter;

error generating means for generating first and second errors for the input value undergoing conversion;

the first error being based upon application of the separable filter to a two dimensional array comprising a first set of output values corresponding to the previously converted input values, a second set of output values corresponding to the subsequent input values and an output value having a first level assumed for the input value undergoing conversion and the second error being based upon application of the separable filter to a further two dimensional array comprising the first and second sets of output values and an output value having a second level assumed for the input value undergoing conversion; and

output value setting means for setting the input value undergoing conversion to the output value having the first level if the first error is less than the second error and to the output value having the second level if the first error is greater than the second error.

9. The system of claim 8 wherein the output value setting means comprises error diffusing means for diffusing the first error to a selected number of the input values yet to be converted if the input value undergoing conversion is converted to the output value having the first level and for diffusing the second error to a selected number of the input values yet to be converted if the input value undergoing conversion is converted to the output value having the second level.

10. The system of claim 8 wherein the error generating means includes means for predicting output values for the input values yet to be converted to obtain the second set of output values.

11. The system of claim 10 wherein the predicting means includes comparing means for comparing each input value yet to be converted to a threshold, converting means for converting each input value yet to be converted to a predicted output value having the first level or to a predicted output value having the second level depending upon whether the input value yet to be converted is below or above the threshold, and difference diffusing means for diffusing the difference between each input value yet to be converted and the corresponding predicted output value to a selected number of input values yet to be converted.

12. The system of claim 11 wherein the threshold is a random threshold.

13. The system of claim 11 wherein the output value setting means comprises error diffusing means for diffusing the first error to a selected number of the input values yet to be converted if the input value undergoing conversion is converted to the output value having the first level and for diffusing the second error to a selected number of the input values yet to be converted if the input value undergoing conversion is converted to the output value having the second level.

14. The system of claim 11 wherein the system includes means for providing a series of sampled values at a first resolution and wherein the error generating means includes digital resampling means for increasing the resolution of the sampled values to a second resolution greater than the first resolution to obtain the input values.

15. The system of claim 14 wherein the sampled values have a certain bandwidth and wherein the separable filter is of the low pass type that limits the output values to the certain bandwidth.

16. The system of claim 14 further including a film plotter coupled to the output value setting means.

17. The system of claim 14 further including means coupled to the output value setting means for producing a printing plate.

18. The system of claim 14 wherein the providing means comprises a scanner that scans a continuous tone original on a pixel-by-pixel basis.

19. The system of claim 14 wherein the providing means comprises a descreener that converts a halftone original into a continuous tone representation and means for sampling the continuous tone representation to obtain the series of sampled values.

20. A method of converting input values to corresponding output values wherein an input value undergoing conversion has before it previously converted input values and after it input values yet to be converted and wherein each of the input values has one of at least M possible levels and each of the output values has one of N possible levels where M is greater than N, the method comprising the steps of:

(a) using a separable filter to generate N errors wherein each of the N errors is based upon an output value having a corresponding one of the N possible levels assumed for the input value undergoing conversion and wherein at least one of the N errors is smallest;

(b) converting the input value undergoing conversion to an output value having a level associated with the smallest error;

(c) diffusing the smallest error to a predetermined number of input values yet to be converted; and

(d) repeating steps (a)-(c) for each of the input values yet to converted.

21. The method of claim 20 wherein the step of generating N errors includes the steps of applying the separable filter to output values corresponding to a selected number of previously converted input values, applying the separable filter to output values corresponding to a selected number of input values yet to be converted, and applying the separable filter to an output value having a corresponding one of the N possible levels assumed for the input value undergoing conversion.

22. The method of claim 21 wherein the step of generating N errors includes the step of predicting each of the output values corresponding to the selected number of input values yet to be converted by comparing each such input value to the N possible levels, converting each such input value to a predicted output value having one of the N possible levels dependent upon which of the N possible levels is closest to such input value, and diffusing a difference between each such input value and its corresponding predicted output value to a predetermined, subsequent number of input values yet to be converted.

23. The method of claim 22 wherein the step of predicting each of the output values comprises the step of diffusing the difference in a serpentine manner.

24. The method of claim 21 wherein the step of generating N errors includes the step of predicting each of the output values corresponding to the selected number of input values yet to be converted by comparing each such input value to a threshold, converting each such input value to a predicted output value as a result of such comparison, and diffusing a difference between each such input value and its corresponding predicted output value to a predetermined, subsequent number of input values yet to be converted.

25. The method of claim 24 wherein the step of predicting each of the output values includes the step of generating random thresholds against which the input values are compared.

26. The method of claim 25 wherein the step of predicting each of the output values comprises the step of diffusing the difference in a serpentine manner.

27. The method of claim 21 further comprising the step of sampling a continuous tone original image to obtain sampled values at a first resolution and increasing the resolution of the sampled values to obtain the input values.

28. The method of claim 27 wherein the step of sampling comprises the step of using a scanner to obtain the sampled values.

29. The method of claim 27 wherein the step of sampling comprises the steps of descreening a halftone original image and sampling the descreened image to obtain the sampled values.

30. The method of claim 20 wherein the separable filter comprises a plurality of filter values arranged in a linear array.

31. The method of claim 30 wherein the step of converting includes the steps of (i) predicting output values for a selected number of input values yet to be converted, wherein such predicted output values and output values corresponding to previously converted input values are arranged together in a two dimensional array and (ii) applying the linear array of filter values to the output values of the two dimensional array along first and second dimensions thereof.

32. The method of claim 31 wherein the step of converting includes the steps of determining intermediate T values based upon application of the linear array of filter values along one dimension of the two dimensional array and using the T values to compute the N errors.

33. The method of claim 32 wherein the step of determining intermediate T values comprises the step of applying the filter values horizontally to the two dimensional array.

34. The method of claim 33 wherein the step of using the T values includes the step of determining a variable u according to the following equation:

##EQU11## where i and n identify the T values, W_m is a filter value to be applied to a corresponding T value, and p is the number of filter values.

35. The method of claim 34 wherein the T values are arranged in a further two dimensional array and wherein the step of determining the variable u includes the step of applying the filter values vertically to the further two dimensional array.

36. The method of claim 33 wherein the T values are arranged in a further two dimensional array and wherein the step of converting includes the step of applying the filter values vertically to the further two dimensional array.

37. The method of claim 36 wherein the step of converting includes the step of recomputing those T values based upon an output value corresponding to a just converted input value for later use during conversion of subsequent input values.

38. The method of claim 32 wherein the step of converting includes the step of recomputing those T values based upon an output value corresponding to a just converted input value for later use during conversion of subsequent input values.