Robust Mean-Shift implementation

This software implements the Robust Mean-Shift algoritm. Code is written in CUDA.

Requirements

• CUDA SDK (optional, recommended)
• OpenMP (if CUDA not provided, recommended)
• libpng

Installation

You can define in Makefile if you will use the CUDA framework or not, by default no. To use the CUDA framework add CUDA=1 to make command

Run:

make lib CUDA=1
make main CUDA=1

Usage

Sample usage: ./main_ms_rlsf <reference image { rgb }> <noisy image {rgb}> <block_radius> <alpha> <sigma> <iter> where:

Reference image - original, not noisy image Noisy image - image which will be filtered block_radius - radius of the processing block B alpha - numer of first pixels being choosen from patch sigma - smoothing parameter iter - number of iteration for limiting of fliter execution

Acknowledgment

This code uses parts of the Fourier 0.8 library by Emre Celebi licensed on GPL avalaible here:

http://sourceforge.net/projects/fourier-ipal

http://www.lsus.edu/faculty/~ecelebi/fourier.htm