Linear unmixing is a commonly used technique for the analysis of imaging spectrometer data. This procedure requires the use of 'endmember' spectra which represent pure target species. One of the methods under investigation at Canada Centre for Remote Sensing for the automatic extraction of endmembers from the image cube, is called the 'Iterative Error Analysis', (IEA) method. This procedure is described, and the results of its application to a data set acquired over Nevada are presented.
RESUMECet article decrit le systeme ISDAS (Imaging Spectrometer Data Analysis System). WI systeme de traitement d'images developpe par le Centre canadien de teledetection en collaboration avec l'industrie afin de fournir WI systeme efficace de traitement et d'analyse des donnees hyperspectrales. Le systeme ISDAS. confu II partir d'un ensemble commercial de logiciels graphiques, le systeme AVS (Application Yisualization System), implante sur line station de travail SPARC de Sun Microsystems, rencontre les normes pour le prototypage rapide des algorithmes et des produits d'information. La structure modulaire du systeme permet l'ajout facile de nouveaux reseaux de traitement de meme que des modifications aux reseaux existants. De nombreux outils de visualisation ont ite developpes pour l'analyse exploratoire de meme que des outils de traitement des donnees, de pre-traitement et d'extraction de l'information pour l'analyse numerique. Des liens avec une bibliotheque spectraIe implantee dans une base relationnelle de domaine public. PostgreSQL, ont he etablis en soutien II l'analyse des donnees. Afin de permettre l'utilisation d'outils de traitement additionnels. le systeme permet l'acces II d'autres systemes de traitement d'imuges disponibles sur Ie marche, Ie systeme EASIIPACE de pC! par exemple. via les outils d'entree/ sortie de donnees. Le systeme ISDAS traite les donnees de tout capteur hyperspectral aeroporte ou satellitaire en autant que Ie capteur effectue l'acquisition des donnees dans les regions du visible, du proche-infrarouge et de l'infrarouge iJ ondes courtes. L'environnement ISDAS est confu pour faire face aux defis de l'application de la teledetection hyperspectrale II differentes disciplines. Le developpement des applications iJ l'heure actuelle vise des domaines tels que l'exploration geologique, la geologie environnementale et l'agriculture de precision. SUMMARY This paper describes the Imaging Spectrometer DataAnalysis System (ISDAS), an image analysis system developed by the Canada Centre for Remote Sensing in conjunction with industry to provide efficient processing and analysis of hyperspectral data. ISDAS, built on top of a commercial off-the-shelfgraphics software package, the Application Visualization System (AVS), and implemented 011 Sun Microsystems SPARC workstations, meets the requirements for rapid prototyping of algorithms and information products. The modular framework allows easy addition of new processing networks as well as modifications to existing networks. Numerous visualization tools have been developed for exploratory analysis together with data handling, preprocessing, and information extraction tools for numerical analysis. Linkages to a spectral library which was implemented 011 a public domain relational database, PostgreSQL. were established to support the data analysis. In order to make use of additional processing tools, other imaging analysis system such as PCI:~EASIIPACE, a commercially available system call be accessed via the data input/output tools. ISDAS processes...
Hyperspectral imagery from the Compact Airborne Spectrographic Imager (<I>casi</I>) was used to characterize sulphide mine tailings at the Copper Cliff mine in Sudbury, Ontario. The objective of the study was to evaluate the usefulness fo high spatial and spectral resolution data and their analysis techniques for characterizing mine tailings sitees and monitoring their restoration. Flight data were acquired in late August 1996 in 72 continguous 9 nm wide spectral bands from 400 nm to 950 nm along with a detailed ground survey. Image data were processed and analysed using the Imaging Spectrometer Data Anaysis System (ISDAS) developed at the Canada Centre for Remote Sensing. The image data were classified using unconstrained linear spectral unmixing. Validation of the unmixing results was achieved by correlating image fractions to fractions measured on the ground. Five end-members were selected: oxidized tailings, lime, green vegetation, fresh and contaminated water. Results show that image fractions of green vegetation and lime are well correlated with the ground fractions (r-square = 0.96 and 0.92). The lower r-squaare (0.65) of the oxidized tailings endmember could be attributed to the variable concentrations of oxidizing agents (various degrees of oxidation) in the tailings which are difficult to assess visually.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.