General pattern recognition in images using WND-CHARM.|
Please see our new GitHub repository to access our source code repository, issue tracker, and documentation.
The current release is v1.52 available here: wndchrm-1.52.775.tar.gz.
The wndchrm command-line utility takes a root directory path as a parameter and computes up to 2600 image features for each image in the subdirectories that define the image classes.
The utility then splits the images into training and test sets, classifies them, and produces html reports.
The code was tested on MacOS X and Linux.
Use "./configure", "make", "sudo make install" after untaring the archive. Then type "wndchrm -h" for instructions.
A full description of the wndchrm utilitycan be found at:
Shamir L, Orlov N, Eckley DM, Macura T, Johnston J, Goldberg IG, Wndchrm - an open source utility for biological image analysis.
BMC Source Code for Biology and Medicine. 3: 13, 2008.
The wndchrm utility is an implementation of the WND-CHARM algorithm described here:
WND-CHARM: Multi-purpose image classification using compound image transforms.
Orlov N, Shamir L, Macura T, Johnston J, Eckley DM, Goldberg IG
Pattern Recognition Letters. 29(11): 1684-93, 2008.
System Requirements --
Latest wndchrm source code: wndchrm-1.52.774.tar.gz
The original version as published in BMC: wndchrm.tar.gz
Dependencies: Wndchrm requires the installation of two open-source libraries - LibTIFF and FFTW. The libraries can downloaded from:
- CentOS/RedHat: sudo yum install libtiff-devel
- Ubuntu/Debian: sudo apt-get libtiff4-dev
FFTW 3.x: http://www.fftw.org/download.html
- CentOS/RedHat: sudo yum install fftw-static fftw-devel
- Ubuntu/Debian: sudo apt-get libfftw3-dev
Optional for dendrograms: PHYLIP
- Some X11 libraries must be installed prior to compiling/installing PHYLIP ("make install" in the src dir.)
- CentOS/RedHat: sudo yum install libX11-devel libXt-devel libXaw-devel
- Ubuntu/Debian: sudo apt-get install libX11-dev libxt-dev libxaw7-dev
MS Windows Binaries
Please note that these binaries are provided as a convenience and are not necessarily the latest version of wndchrm.
Executable file wndchrm.exe
DLLs necessary to run wndchrm on MS Windows:
Recommended Hardware: 2 GB RAM, 10 GB HD space, 2 GHZ CPU.
Please be aware that this utility is very computationally intensive. Multiple instances of wndchrm can work concurrently on the same dataset on multi-core/multi-processor CPUs.
Simply use the same command line to launch as many instances of wndchrm as there are CPU cores available.
Software: GCC compiler must be installed, LibTIFF and FFTW.
Wndchrm was tested using Linux (Fedora Core, Red Hat) and MacOS X, but should also run with other Unix and Linux distributions.
Installation difficulties are expected if Libtool is compiled with a different compiler version than the one used for compiling LibTIFF (this situation is more likely with upgraded operating systems).
This problem can be solved by installing a newer gcc version.