--- Start of Idrisi List message
Dear Colleague,

Idrisi32 Release 2 is now shipping, and ordering the software has never been
easier! Merely visit the Ordering page on our web site at www.clarklabs.org
and select the appropriate product. You can have the CD shipped directly to
your mailing address or you can purchase a digital download.

As always, you can still place your order via mail, email, fax or phone.  If
you have any questions or concerns, a customer service representative will
be happy to assist you.

With this release, you will receive a CD containing the Idrisi32 Release 2
software along with our new interactive manual set in Adobe's Portable
Document Format (pdf).

Below please find some of the new features of Idrisi32 Release 2.

Macro Modeler 
Release 2 enhances system functionality through the addition of a graphical
modeling environment. Users develop models by constructing graphical flow
diagrams using a simple drag and drop interface and over 100 analytical
functions. Dynamic models are also supported by linking outputs as inputs in
later iterations. Repetition structures also allow the processing of
multiple modules through a model and the development of iterative
procedures. Users can also save models as submodels, thereby extending the
functionality of the modeling system. Users can also extend the
functionality of IDRISI32 through the addition of their own modular
programs. Idrisi32 is fully COM compliant, offering comprehensive access to
the system in a manner that is simple to access from programming
environments such as Visual Basic for Applications, Delphi, or Visual C++.
The complete Idrisi32 API documentation in pdf format can be accessed from
the new Modeling menu. 

Image Processing
Release 2 includes extensive additions to the image processing suite and new
tools for change and time series analysis. For image processing, facilities
have been added for automatic mosaicking (MOSAIC), atmospheric correction
(ATMOSC), local affine transformation (LOCAL AFFINE), and the import of a
variety of new image data formats including SPOT Cap format (SPOT , LANDSAT
). RADIANCE has been updated to allow for user defined specifications. In
addition, a set of new multi-spectral classification routines has been
added, including linear discriminant analysis (Fisher Classifier) (FISHER)
and linear spectral unmixing (UNMIX). 

The Fisher Classifier establishes an optimal set of linear functions for the
discrimination between classes in training site data. We have found the
results to be superior to Maximum Likelihood classification in many
instances.

Linear Spectral Unmixing complements the other soft classifiers in Idrisi32
by offering superior capabilities for sub-pixel classification. Sub-pixel
classification provides estimates of the proportional composition of mixed
pixels. However, the underlying regression procedure usually limits the
number of potential constituents to no more than the number of bands. To
avoid this, Idrisi32 offers several procedures, including one that combines
the Bayesian probability calculation of Maximum Likelihood in order to
isolate likely constituents and linear spectral unmixing to estimate their
proportional composition. 

Release 2 supports a special procedure for the delineation of training
sites: flood polygon digitizing (DIGITIZE). A flood polygon is one that
comprises all pixels that are adjacent to a designated pixel and which fall
within a specified multi-spectral tolerance of the reflectance associated
with the designated pixel. For example, suppose that one can see evidence of
a stand of red pines in a 24-bit composite. To isolate a training site for
red pine, one would activate the flood polygon feature while on-screen
digitizing and then click on the single pixel that seems most representative
of that stand. One would then be prompted for a multi-spectral tolerance
(e.g., 4 Dn) and a maximum distance over which the flood polygon should be
constructed. The procedure then constructs this polygon as a vector feature.

Hyperspectral Analysis
Release 2 also provides major enhancements to the Idrisi32 suite of tools
for hyperspectral image analysis. These include absorption spectra analysis
using continuum removal for estimation of the degree of support for members
of a library of spectral response curves developed in a laboratory setting
(HYPERABSORB), a procedure commonly used in prospecting applications.
Hyperspectral classifier development includes an unsupervised classifier
(HYPERUSP), orthogonal subspace projection (HYPEROSP), and linear spectral
unmixing (HYPERUNMIX). Other additions to the Image Processing suite include
automatic signature extraction (HYPERAUTOSIG), end-member signature creation
(ENDSIG), and a screening procedure for atmospheric noise (SCREEN).

Change and Time Series Analysis
Idrisi has long had a distinctive set of facilities for change and time
series analysis. With Release 2, this capability has been streamlined and
expanded with special tools for image differencing (IMAGEDIFF), image
ratioing (IMAGERATIO), change vector analysis (CVA), and regression-based
calibration (for regression differencing) (CALIBRATE). For time series data,
a temporal resonance tool has been developed to determine the degree of
correlation between each pixel, over time, and a designated temporal index
(CORRELATE). For example, one could map the degree of resonance, over time,
with the El Nino phenomenon by supplying a time series of NDVI images and a
temporal series of the Southern Oscillation Index.

Special attention has also been directed to the problem of land cover change
modeling. Release 2 provides a tool for Markov Chain analysis (MARKOV) and
the modeling of change based on Cellular Automata (CA_MARKOV). The most
ambitious of these features is a land cover change simulation model that
incorporates the strengths of multi-criteria/multi-objective decision
support with cellular automata modeling. Special focus has also been
directed to the problem of model validation, with a set of tools for
comparison of categorical map data (ROC, VALIDATE).

Other Additions
Other important additions to Release 2 include a wizard for
multi-criteria/multi-objective decision making (WIZARD), improved logistic
regression modeling (LOGISTICREG), and import/export to the Statistica
statistical analysis software system (v 5.5) (STATIDRIS). New tools for
surface analysis include  a new flow direction (FLOW) procedures,  a revised
(and fast) VIEWSHED operator that can calculate the proportion of a feature
in view, and a streamlined set of TIN operators. RECLASS, ASSIGN, CONVERT,
and CONCAT now also work with vector data. In addition, RECLASS now has the
option to assign real number values.

A Windows List menu item has been added for open window management including
an option to Close All open windows. Shortcut (SHORTCUT), a utility from
which any module may be launched from an alphabetical list of modules, has
also been reintroduced with this version. Toggle it on and off under the
File menu. 

Finally, all users now have complete access to our documentation on-line.
>From the Help menu, you can access both the manuals and the tutorials in pdf
format. You can also link directly to our web site to check for the latest
patch releases and to report technical problems. 

With these additions to Idrisi32, Clark Labs continues to set the standard
for excellence in analytical geographic software.

To contact Clark Labs, our address is:

Clark Labs
Clark University
950 Main Street
Worcester, MA  01610-1477  USA

To contact us by phone, fax, electronic mail, or to visit our web site:

Customer Support:       +1.508.793.7526
Technical Inquiries:    +1.508.793.7528
Fax:                    +1.508.793.8842
Internet:                       [EMAIL PROTECTED]
Web Site:               www.clarklabs.org


Sincerely,

The Clark Labs



--- End of Idrisi List message

IDRISI-L is an unmoderated list. As a service to list users, please post a 
summary of useful responses to your questions.
To post a message to IDRISI-L, send it to [EMAIL PROTECTED] To 
subscribe to the list, send an email from the address to be subscribed to 
[EMAIL PROTECTED] with "subscribe" (without quotes) as the first line 
of the message body. To unsubscribe, do the same, only with "unsubscribe" 
as the first line of the message body.
The list is archived at
http://www.mail-archive.com/idrisi-l%40clarklabs.org/.

Reply via email to