This directory contains programs and images used in paper:
H. Y. Kim, "Binary Halftone Image Resolution Increasing by Decision Tree Learning," IEEE T. Image Processing, 2003.

The files are divided in four ZIP files:

program.zip - contains source programs, executable programs and 8x8 window definition.
cl-set.zip - Halftone images generated by clustered-dot ordered dithering.
cs-set.zip - Halftone images generated by HP LaserJet Driver option "coarse dot."
fn-set.zip - Halftone images generated by HP LaserJet Driver option "fine dot."

The easiest way to perform the tests is:
1) Download 4 zip files, uncompress them in the same directory.
2) Run cl-run.bat to run the first test (clustered-dot ordered dithering). It consists on two programs:
   a) bldz cl-ax.bmp cl-ay.bmp e88.mat cl-filt.ctz
      Builds WZ-operator cl-filt.ctz using 8x8 window (e88.mat) and sample images cl-ax.bmp and cl-ay.bmp.
   b) appz cl-qx.bmp cl-filt.ctz cl-qp.bmp
      Applies the filter cl-filt.ctz to zoom cl-qx.bmp obtaining cl-qp.bmp.
      The processed image cl-qp.bmp is similar to the ideal image cl-qy.bmp.
3) Run cs-run.bat to run the second test (HP coarse dot).
   The processed image is cs-qp.bmp and the ideal image is cs-qy.bmp.
4) Run fn-run.bat to run the third test (HP fine dot).
   The processed image is fn-qp.bmp and the ideal image is fn-qy.bmp.

All programs are written in C++ and they must be compiled using DJGPP compiler and IMG library ( http://www.lps.usp.br/~hae/software ). All programs will run on Windows 98, ME and XP. Probably, the programs also run on Windows 95, 2000 and NT.