UDAT

UDAT is a pattern recognition tool for mass analysis of various types of data, including image, audio, and text. UDAT computed a large set of numerical content descriptors from each file it analyzes, and selects the most informative features using statistical analysis. The tool can perform automatic classification of galaxy images by training with annotated galaxy images. It also has unsupervised learning capabilities, such as query-by-example of galaxies based on morphology. That is, given an input galaxy image of interest, the tool can search through a large database of images to retrieve the galaxies that are the most similar to the query image. The downside of the tool is its computational complexity, which in most cases will require a small or medium cluster.

Shamir, L. (2017). UDAT: A multi-purpose data analysis tool. Astrophysics Source Code Library, 1704.002



Download UDAT source code: UDAT.tar.gz


Important:
UDAT requires the installation of two publicly available libraries - LibTIFF and FFTW. The libraries cab downloaded from:

LibTIFF: http://www.libtiff.org     http://download.osgeo.org/libtiff/
FFTW 3.1: http://www.fftw.org/download.html
LibSndFile: http://www.mega-nerd.com/libsndfile/

Linux compiled binary: udat


Systems requirements:
Recommended Hardware: 512 MB RAM, 10 MB HD space, 2 GHZ CPU. Software: GCC compiler must be installed. Udat was tested using Linux (Fedora, CentOS), but should also run under other Linux distributions.


MS-Windows binary udat.exe

The executable file needs the following run-time DLLs:
fftw3.dll
libfftw3-3.dll
libtiff3.dll
jpeg62.dll
zlib1.dll
libsndfile-1.dll


It might also require Visual Studio 2013 run-time libraries.