Digital halftoning method utilizing diffused weighting and class matrix optimization
AssigneeNational Taiwan University of Science and Technology
InventorsJing-ming Guo, Yun-fu Liu
Life of this patent
Estimating the term…
Abstract
The present invention discloses a digital halftoning method. The method comprises steps of: (a1) dividing an original image into non-overlapping blocks; (a2) obtaining a Least-Mean-Square trained (LMS-trained) filter by comparing at least a training image and a halftone result corresponding to the training image (a3) optimizing a class matrix with the LMS-trained filter, which involves the diffused area and the diffused weightings; and (a4) processing the non-overlapping blocks by performing a dot diffusion procedure with the optimized class matrix and the diffused weightings to generate a halftone image corresponding to the original image. A detailed class matrix optimizing method as in the above-mentioned step (a3) is also disclosed.
Claims17
A digital halftoning method, the method comprising steps of:
- (a1) dividing an original image into non-overlapping blocks;
- (a2) comparing at least a training image and a halftone result corresponding to the training image to obtain diffused weightings of a diffused area;
- (a3) optimizing a class matrix with the diffused weightings of the diffused area to obtain an optimized class matrix;
- and
- (a4) processing the non-overlapping blocks by performing a dot diffusion procedure with the optimized class matrix and the diffused weightings to generate a halftone image corresponding to the original image.
A method for optimizing a class matrix, the method comprising steps of:
- (c1) selecting a member of the class matrix;
- (c2) swapping the selected member of the class matrix with one of the other members;
- (c3) performing a dot diffusion procedure with the class matrix to at least a testing image to obtain a first halftone testing image and performing the same dot diffusion procedure with the swapped class matrix obtained from the swapping step (c2) to the testing image to obtain a second halftone testing image;
- (c4) comparing a peak signal-to-noise ratio (PSNR) of the first halftone testing image and a PSNR of the second halftone testing image to obtain a comparing result;
- and
- (c5) determining an optimized class matrix according to the comparing result.
A method for optimizing a class matrix, the method comprising steps of:
- (d1) selecting a member of the class matrix;
- (d2) swapping the selected member of the class matrix with the other members to obtain a plurality of swapped class matrices;
- (d3) performing a dot diffusion procedure with diffused weightings of a diffused area and the class matrix, and performing the same dot diffusion procedure with the diffused weightings of the diffused area and the swapped class matrices to at least a testing image to obtain corresponding halftone testing images;
- (d4) evaluating peak signal-to-noise ratios (PSNRs) of each halftone testing image;
- and
- (d5) retaining the class matrix which has the maximal PSNR of the halftone testing images.
Dependencies are inferred from the claim text.