Solutionware: Hyperspectral Image Analysis Toolbox
The Hyperspectral Image Analysis Toolbox (HIAT) is intended for the
analysis of hyperspectral and multispectral data. HIAT is a collection
of functions that extend the capabilities of the MATLAB numerical
computing environment. It has been implemented for the MacIntosh and
PC-Windows systems using MATLAB. The purpose of this toolbox is to
provide the user with an environment where can utilize
different image processing methods for hyperspectral and multispectral
data. HIAT provides standard image processing methods such as
discriminant analysis, principal component, euclidean distance, and
maximum likelyhood. In addition, image processing methods developed by
research done at the Laboratory
of Applied Remote Sensing and Image Processing at the University of Puerto Rico at Mayagüez is also included.
- Image formats for loading and saving: .mat, .bsq, .bil, .bip,
.jpg, with ENVI header info and .tiff.
- Pre-Processing algorithms: Resolution Enhancements and Principal
Component Analysis Filter.
- Feature Extraction/Selection Algorithms: Principal
Components Analysis, Discriminant Analysis, Singular Value
Band Subset Selection, Information Divergence Band. Subset Selection,
Information Divergence Projection Pursuit, Optimized Information
Divergence Projection Pursuit.
- Classifiers: Euclidean Distance, Fisher's Linear Discriminant,
Mahalanobis Distance, Maximum Likelihood, Angle Detection.
- Post-Processing Algorithms: supervised and unsupervised ECHO
classifier. ECHO (2x2, 3x3, 4x4).
- Supported Platforms: UNIX/Linux, MS-Windows 2000 and
XP, Macintosh (OS X 10.1.4 and higher).
- Online help documentation and a hyperspectral data set.
- MATLAB 6.5, R13 or higher
- Image Processing Toolbox of MATLAB
- 512MB RAM.
Download and Install the Hyperspectral Image
The software is available for downloading to members of the
hyperspectral image analysis and related research communities for non-commercial
purposes only. It is distributed as a 17.1MB zip file for
MATLAB and 135MB for the standard alone version.
to download for the MATLAB version.
Click here to download the standard alone version.
for MATLAB version
- Create a directory in which you would like to place the software
(e.g., C:\HIAT or ~/HIAT).
- Move the zipped toolbox to this directory.
- Unzip the software. This will create a directory called
HIAT 2.0 which contains the precompiled matlab pcode. The
documentation is in a directory called hyper_help; to view the
documentation locally, load the file
toolbox_folder/hyper_help/hyper_product_page.html in your browser.
- After following the steps of Installation Instructions open
a session of MATLAB.
- In the MATLAB current directory
path change to the path where you install the toolbox.
- To run HIAT type in the MATLAB command prompt: start.
Instructions for standard alone version
- Unzip the file HIAT_V2_SAV.zip.
- Double click in the file MCRInstaller.exe included downloaded
(this file include the necessary information from MATLAB to run
HIAT) an install the files in a place where you have access
- Double click in the file HIAT_V2B.exe to run the toolbox (the
first time it will extract some information into the folder
- Double click in HIAT_V2B.exe to run the toolbox.
Documentation and software use guidelines can be viewed online,
and are also included in the software download.
Main GUI with RGB Image Loaded
Sample of Classification Map
Sample of HSI Cube
- Luís O. Jiménez, Professor, Electrical
and Computer Eng'g, University of Puerto Rico at Mayagüez
- Miguel Vélez-Reyes, Professor,
Electrical and Computer Eng'g, University of Puerto Rico at Mayagüez
- Shawn Hunt,
Professor, Electrical and Computer Eng'g, University of Puerto Rico at
- David Kaeli,
Associate Professor, Electrical and Computer Engineering, Northeastern
Arzuaga-Cruz, Ph.D. student, Electrical and Computer Engineering,
- Samuel Rosario-Torres,
CenSSIS-UPRM Software Engineer, University of Puerto Rico at Mayagüez
Castrodad, Graduate Student, Electrical and Computer Eng'g,
University of Puerto Rico at Mayagüez
For more information, support, bug reports, and suggestions, please