Extended Abstract (276K)5A.6
ADaM-IVICS: A software tool to mine satellite data
Todd Berendes, Univ. of Alabama, Huntsville, AL; and R. Ramachandran, J. Rushing, and S. Graves
The Algorithm Development and Mining System (ADaM) is a data mining toolkit designed for use with scientific and image data. It includes classification, clustering, feature selection, model validation, data cleaning, image processing, optimization, and association rule mining capabilities. The system consists of a set of individual algorithms or components that can be put together to perform complex tasks. Components are packaged as stand alone executables and as Python modules for easy scripting. The Interactive Visualizer and Image Classifier (IVICS) was developed as a visualization tool to facilitate selection of training samples from satellite images. It has evolved into a general purpose visualization system which supports several common satellite and remote sensing data formats. Recently, IVICS has been integrated with the ADaM toolkit. This integration now provides end users not only the capability to interactively visualize and analyze satellite data but also exploit the large suite of mining algorithms available in ADaM. This paper will describe this integrated tool. The paper will also describe the process of performing both unsupervised and supervised classification using this tool
Session 5A, Satellite IIPS and Applications
Wednesday, 17 January 2007, 8:30 AM-10:00 AM, 216AB
Previous paper
Recorded presentation