IPOL-BSDE

TITLE

Implementation of Image Denoising based on Backward Stochastic Differential Equations

VERSION

2.1

AUTHOR

Dariusz Borkowski dbor@mat.umk.pl

CONTENTS

./bsde.cpp - This program reads PNG images, runs BSDE
	 method and writes PNG images. Only 8bit (grayscale
	 and RGB) PNG images are handled. 

./libdenoising.cpp, ./libdenoising.h - Implementation of BSDE
	method: Algorithm 1, Algorithm 2, Algorithm 3.
	This source code provides an implementation of the BSDE
	denoising algorithm.

./lib.cpp, ./lib.h - Implementation of classes dbImageRGB
	and dbImage representing RGB and greyscale images and
	operations on them.

./io_png.c, ./io_png.h - PNG read/write interface by Nicolas
	Limare <nicolas.limare@cmla.ens-cachan.fr>.

./lena.png, ./dice.png, ./traffic.png, ./computer.png,
	./book.png, book10.png - Test PNG images.

The code is written in C/C++ and should compile on any system with an g++ compiler.

The libpng header and libraries are required on the system for compilation and execution.

The implementation uses OPENMP which not supported by old versions of g++.

Ubuntu 22.04 LTS installation:

sudo apt-get install make
sudo apt-get install g++
sudo apt-get install libpng-dev

COMPILATION INSTRUCTIONS

Simply use the provided makefile, with the command make.

USAGE

usage: ./ipol-bsde original.png sigma noisy.png denoised.png add_noise c

BSDE takes 6 parameter:

• original.png : initial noise free image (PNG image)
• sigma : the noise standard deviation (non negative real number)
• noisy.png : noisy image used by the denoising algorithm (PNG image)
• denoised.png : denoised image (PNG image)
• add_noise : the add noise option (integer number from the set {0,1})
• c : the enhancing parameter c (non-negative real number from the range [0, 1])

EXAMPLES OF USE

Use the provided makefile, with the command make tests

Example 1.

We add noise with a standard deviation sigma=20 to lena.png original image and reconstruct it without the enhancing effect c=0.0. We save the noisy and reconstructed image as in.png and out.png, respectively. The program displays the PSNR value of the original and reconstructed image.

./ipol-bsde lena.png 20 in.png out.png 1 0.0

Example 2.

We reconstruct lena.png noisy image with a sigma=10 and the enhancing effect c=0.4. We save the noisy and reconstructed image as in.png and out.png, respectively. In this case, we omit the display of the PSNR value.

./ipol-bsde lena.png 10 in.png out.png 0 0.4

COPYRIGHT AND LICENSE INFORMATION

Copyright (c) 2023 Dariusz Borkowski

This program is free software: you can redistribute it and/or modify it under the terms of the GNU Affero General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more details.

You should have received a copy of the GNU Affero General Public License along with this program. If not, see http://www.gnu.org/licenses/.