The spectral image processing system (SIPS)—interactive visualization and analysis of imaging spectrometer data

Kruse, Fred A.; Lefkoff, A. B.; Boardman, Joseph W.; Heidebrecht, Kathleen B.; Shapiro, A. T.; Barloon, P. J.; Goetz, Alexander

doi:10.1016/0034-4257(93)90013-n

Cited by 2,455 publications

(655 citation statements)

References 11 publications

Supporting

Mentioning

550

Contrasting

Unclassified

Order By: Relevance

“…2a) and ground truths (to identify the spectral signatures of relevant soil uses and crop types) has allowed a spatially-distributed and updated description of the soil uses which exerts a fundamental control on the soil mosture dynamics and the ensuing runoff production. The classification procedure adopted was based on the spectral angle mapper (SAM) algorithm (Kruse et al, 1993) and its application is shown in Fig. 3.…”

Section: Source Area Geomorphic Transitions Legendmentioning

confidence: 99%

Transport at basin scales: 2. Applications

Rinaldo

Botter

Bertuzzo

et al. 2006

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. In this paper, the second of a series, we apply the models discussed in Part 1 to a significant case study. The nature of the catchment under study, the transport phenomena investigated (i.e. nitrates moving as solutes within runoff waters) and the scales involved in space and time, provide an elaborate test for theory and applications. Comparison of modeling predictions with field data (i.e. fluxes of carrier flow and solute nitrates) suggests that the framework proposed for geomorphic transport models is capable to describe well large-scale transport phenomena driven and/or controlled by spatially distributed hydrologic fields (e.g. rainfall patterns in space and time, drainage pathways, soil coverage and type, matter stored in immobile phases). A sample MonteCarlo mode of application of the model is also discussed where hydrologic forcings and external nitrate applications (through fertilization) are treated as random processes.

show abstract

Section: Source Area Geomorphic Transitions Legendmentioning

confidence: 99%

Transport at basin scales: 2. Applications

Rinaldo

Botter

Bertuzzo

et al. 2006

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

show abstract

“…Distance-based classifiers (EMD and SAM) confirmed their underperformance compared to statistical and non-parametric algorithms (MLC and NN). This is probably due to EMD and SAM lacking capabilities in handling intra-class variance into the classification decision rules [South et al, 2004], and in their original design being based on spectral rather than multi-temporal information [Kruse et al, 1993;South et al, 2004]. Moreover, SAM invariance to relative magnitude of input features is possibly adding some confusion when inputs are multi-temporal VIs profiles [Kruse et al, 1993].…”

Section: Discussionmentioning

confidence: 99%

“…This is probably due to EMD and SAM lacking capabilities in handling intra-class variance into the classification decision rules [South et al, 2004], and in their original design being based on spectral rather than multi-temporal information [Kruse et al, 1993;South et al, 2004]. Moreover, SAM invariance to relative magnitude of input features is possibly adding some confusion when inputs are multi-temporal VIs profiles [Kruse et al, 1993]. MCL requires the availability of training data, which could limit its operational implementation; alternative approaches could be rule-based classifiers exploiting spectraltemporal profiles and synthetic features as input.…”

Section: Discussionmentioning

confidence: 99%

“…before water demand peak of major summer crops in the study area (rice and maize) and last top-dressing fertilization. Four supervised classification algorithms, three parametric and one non-parametric, were used: Maximum Likelihood Classification (MLC) [Richards, 1999], Euclidean Minimum Distance (EMD) [Richards, 1999], Spectral Angle Mapper (SAM) [Kruse et al, 1993] and Neural Networks (NN) [Richards, 1999]. All of these algorithms are coded in IDL and embedded in ENVI 4.7 software package.…”

Section: Methodsmentioning

confidence: 99%

“…early season and end of season maps) over Lombardy, one of the most important extensive agricultural region in Italy. To this aim, we tested four common supervised classification algorithms (Maximum Likelihood, Minimum Distance, Spectral Angle Mapper and Neural Networks) [Kruse, 1993;Richards, 1999], applied to multi-temporal VIs derived from medium resolution satellite data (i.e. Landsat 5, 7 and 8).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Assessing in-season crop classification performance using satellite data: a test case in Northern Italy

Azar

Villa

Stroppiana

et al. 2016

European Journal of Remote Sensing

View full text Add to dashboard Cite

This study investigated the feasibility of delivering a crop type map early during the growing season. Landsat 8 OLI multi-temporal data acquired in 2013 season were used to classify seven crop types in Northern Italy. The accuracy achieved with four supervised algorithms, fed with multi-temporal spectral indices (EVI, NDFI, RGRI), was assessed as a function of the crop map delivery time during the season. Overall accuracy (Kappa) exceeds 85% (0.83) starting from mid-July, five months before the end of the season, when maximum accuracy is reached (OA=92%, Kappa=0.91). Among crop types, rice is the most accurately classified, followed by forages, maize and arboriculture, while soybean or double crops can be confused with other classes.

show abstract

Near-infrared Raman imaging microscope based on fiber-bundle image compression

Gift

Haber

et al. 1999

J. Raman Spectrosc.

View full text Add to dashboard Cite

The design and performance of a near-infrared Raman imaging microscope (NIRIM) is described. This new instrument utilizes fiber-bundle image compression (FIC) to collect simultaneously a 3-D Raman spectral imaging data cube. Key NIRIM design features are discussed, including the FIC fiber-bundle, excitation laser, optical coupling to the microscope and fiber-bundle, holographic filtering, spectrograph imaging requirements, CCD parameters and chemical image processing. The theoretical collection efficiency and image quality of the NIRIM instrument are compared with those of tunable filter and line scanning Raman imaging methods. The performance of the NIRIM is demonstrated using a white light image of a bar-target and Raman chemical images of samples containing fructose-sucrose and Pb(NO 3 ) 2 -K 2 SO 4 microcrystalline mixtures. A Raman image collection time as fast as 1 s (total detector integration time) is demonstrated, for a 3-D data cube containing 322 image resolution elements and 900 Raman shift wavelengths.

show abstract

The spectral image processing system (SIPS)—interactive visualization and analysis of imaging spectrometer data

Cited by 2,455 publications

References 11 publications

Transport at basin scales: 2. Applications

Transport at basin scales: 2. Applications

Assessing in-season crop classification performance using satellite data: a test case in Northern Italy

Near-infrared Raman imaging microscope based on fiber-bundle image compression

Contact Info

Product

Resources

About