AC-ARPENOS
A-Contrario Automated Removal of quasi-PEriodic NOise using
frequency domain Statistics
The following codes can be used for academic research only:
MATLAB function:
ACARPENOS.m v1.2
Needs symmetrically padded convolution:
sym_conv2.m
Example script for additional experiments:
ACARPENOS_example.m
Additional experiments (canvas pattern removal in paintings):
canvas.m. Images:
Rockwell - Carry On - 002.jpg (from
here),
Rockwell - Carry On - 003.jpg (from
here),
Rockwell - The Adventure Trail - 003.jpg (from
here),
Rockwell - Our Heritage - 002.jpg (from
here), and
one_inclusions.sm_.jpg (from
here).
Description of the
Total Variation minimization postprocessing:
note_TVmin.pdf
Images:
mandril_gray.tif;
boat.png;
Striping_Noise_Sample.jpg (comes from
here);
mariner6_close.jpg (comes from
here);
mariner04_07b.gif (comes from
here);
vicking1_f603a14a_crop.pgm (comes from
here);
halftone.pgm and
halftone-eye.pgm (comes from
here);
scanline.tif (comes from
here);
lena512.tif; images from
Gonzalez and Woods textbook can be found
here and
here.
Slides:
talk_ACARPENOS.pdf
Please
cite:
F. Sur.
An a-contrario approach to quasi-periodic noise removal. Proceedings of the
IEEE International Conference on Image Processing (ICIP), p. 3841-3845, Québec City (Canada), September 2015.
Link.
Pdf file.
Bibtex.
See also
ARPENOS.
History:
May 2015: v1.0, first release.
June 2015: v1.1, spectrum interpolation with TV minimization.
January 2016: experiments on canvas pattern removal.
August 2016: v1.2, numerical scheme for constrained TV minimization corrected.
Last modified: August 26th, 2016.