Shape-based unmixing for vegetation mapping

L. Tits, B. Somers, W. De Keersmaecker, G. P. Asner, J. Farifteh, P. Coppin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Spectral mixture analyses (SMA) is often used as a tool to map complex/mixed (semi-)natural ecosystems.Yet, the performance of SMA, which traditionally uses the amplitudebased RMSE as the objective function, is often hampered by the high spectral similarity among co-occurring plant species. Experiments, based on ray-tracing simulations, in situ measured reflectance data and AVIRIS imagery demonstrated the added value of implementing shape-based error metrics in the unmixing of forests and orchards. The approach allowed to highlight the subtle spectral differences among co-occurring plant species resulting in an overal improvement of species specific mapping (i:e: decrease in MSE ≈ 40%).

Original languageEnglish (US)
Title of host publication2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012
PublisherIEEE Computer Society
ISBN (Print)9781479934065
DOIs
StatePublished - Jan 1 2012
Externally publishedYes
Event2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012 - Shanghai, China
Duration: Jun 4 2012Jun 7 2012

Publication series

NameWorkshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing
ISSN (Print)2158-6276

Conference

Conference2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012
CountryChina
CityShanghai
Period6/4/126/7/12

Fingerprint

Orchards
Ray tracing
Ecosystems
Experiments

Keywords

  • forests
  • Hyperspectral
  • orchards
  • Spectral Mixture Analysis
  • spectral similarity

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Tits, L., Somers, B., De Keersmaecker, W., Asner, G. P., Farifteh, J., & Coppin, P. (2012). Shape-based unmixing for vegetation mapping. In 2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012 [6874222] (Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing). IEEE Computer Society. https://doi.org/10.1109/WHISPERS.2012.6874222

Shape-based unmixing for vegetation mapping. / Tits, L.; Somers, B.; De Keersmaecker, W.; Asner, G. P.; Farifteh, J.; Coppin, P.

2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012. IEEE Computer Society, 2012. 6874222 (Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tits, L, Somers, B, De Keersmaecker, W, Asner, GP, Farifteh, J & Coppin, P 2012, Shape-based unmixing for vegetation mapping. in 2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012., 6874222, Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing, IEEE Computer Society, 2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012, Shanghai, China, 6/4/12. https://doi.org/10.1109/WHISPERS.2012.6874222
Tits L, Somers B, De Keersmaecker W, Asner GP, Farifteh J, Coppin P. Shape-based unmixing for vegetation mapping. In 2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012. IEEE Computer Society. 2012. 6874222. (Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing). https://doi.org/10.1109/WHISPERS.2012.6874222
Tits, L. ; Somers, B. ; De Keersmaecker, W. ; Asner, G. P. ; Farifteh, J. ; Coppin, P. / Shape-based unmixing for vegetation mapping. 2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012. IEEE Computer Society, 2012. (Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing).
@inproceedings{82589dce080f40ebb062e5f642bb6042,
title = "Shape-based unmixing for vegetation mapping",
abstract = "Spectral mixture analyses (SMA) is often used as a tool to map complex/mixed (semi-)natural ecosystems.Yet, the performance of SMA, which traditionally uses the amplitudebased RMSE as the objective function, is often hampered by the high spectral similarity among co-occurring plant species. Experiments, based on ray-tracing simulations, in situ measured reflectance data and AVIRIS imagery demonstrated the added value of implementing shape-based error metrics in the unmixing of forests and orchards. The approach allowed to highlight the subtle spectral differences among co-occurring plant species resulting in an overal improvement of species specific mapping (i:e: decrease in MSE ≈ 40{\%}).",
keywords = "forests, Hyperspectral, orchards, Spectral Mixture Analysis, spectral similarity",
author = "L. Tits and B. Somers and {De Keersmaecker}, W. and Asner, {G. P.} and J. Farifteh and P. Coppin",
year = "2012",
month = "1",
day = "1",
doi = "10.1109/WHISPERS.2012.6874222",
language = "English (US)",
isbn = "9781479934065",
series = "Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing",
publisher = "IEEE Computer Society",
booktitle = "2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012",

}

TY - GEN

T1 - Shape-based unmixing for vegetation mapping

AU - Tits, L.

AU - Somers, B.

AU - De Keersmaecker, W.

AU - Asner, G. P.

AU - Farifteh, J.

AU - Coppin, P.

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Spectral mixture analyses (SMA) is often used as a tool to map complex/mixed (semi-)natural ecosystems.Yet, the performance of SMA, which traditionally uses the amplitudebased RMSE as the objective function, is often hampered by the high spectral similarity among co-occurring plant species. Experiments, based on ray-tracing simulations, in situ measured reflectance data and AVIRIS imagery demonstrated the added value of implementing shape-based error metrics in the unmixing of forests and orchards. The approach allowed to highlight the subtle spectral differences among co-occurring plant species resulting in an overal improvement of species specific mapping (i:e: decrease in MSE ≈ 40%).

AB - Spectral mixture analyses (SMA) is often used as a tool to map complex/mixed (semi-)natural ecosystems.Yet, the performance of SMA, which traditionally uses the amplitudebased RMSE as the objective function, is often hampered by the high spectral similarity among co-occurring plant species. Experiments, based on ray-tracing simulations, in situ measured reflectance data and AVIRIS imagery demonstrated the added value of implementing shape-based error metrics in the unmixing of forests and orchards. The approach allowed to highlight the subtle spectral differences among co-occurring plant species resulting in an overal improvement of species specific mapping (i:e: decrease in MSE ≈ 40%).

KW - forests

KW - Hyperspectral

KW - orchards

KW - Spectral Mixture Analysis

KW - spectral similarity

UR - http://www.scopus.com/inward/record.url?scp=84906545680&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906545680&partnerID=8YFLogxK

U2 - 10.1109/WHISPERS.2012.6874222

DO - 10.1109/WHISPERS.2012.6874222

M3 - Conference contribution

AN - SCOPUS:84906545680

SN - 9781479934065

T3 - Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing

BT - 2012 4th Workshop on Hyperspectral Image and Signal Processing, WHISPERS 2012

PB - IEEE Computer Society

ER -