Optimizing mixed spectra generation for regression-based unmixing of land cover in urban areas

Frederik Priem; Frank Canters; Sebastian Van der Linden; Akpona Okujeni

doi:10.1109/JURSE.2017.7924554

Optimizing mixed spectra generation for regression-based unmixing of land cover in urban areas

Frederik Priem, Frank Canters, Sebastian Van der Linden, Akpona Okujeni

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › Research

7 Citations (Scopus)

Abstract

Regression-based unmixing for quantifying urban land cover at the subpixel scale requires mixed training spectra for model calibration. In this paper optimization and synthetic mixing of hyperspectral image endmember libraries for the calibration of unmixing models are investigated. APEX and HyMap airborne hyperspectral transects respectively covering Brussels and Berlin are used to produce an endmember library for unmixing, as well as reference land cover fractions for validation. The library is spectrally resampled, optimized and synthetically mixed to produce quantitative training data for unmixing of a Sentinel-2 surface reflectance image of Brussels (Belgium). Support Vector Regression models are developed for vegetation-impervious-soil land cover mapping. Findings may contribute to the use of multi-sensor data and to the demonstration of Sentinel-2's added value for quantitative urban land cover assessment.

Original language	English
Title of host publication	2017 Joint Urban Remote Sensing Event (JURSE)
Place of Publication	Dubai, United Arab Emirates
Publisher	IEEE
Pages	1-4
Number of pages <span style="color:red"p> <font size="1.5"> ✽ </span> </font>	4
Edition	2017
ISBN (Electronic)	978-1-5090-5808-2
ISBN (Print)	978-1-5090-5809-9
DOIs	https://doi.org/10.1109/JURSE.2017.7924554
Publication status	Published - 2017
Event	Joint Urban Remote Sensing Event - Ritz-Carlton Hotel Dubai, Dubai, United Arab Emirates Duration: 6 Mar 2017 → 8 Mar 2017 http://www.jurse2017.com/

Conference

Conference	Joint Urban Remote Sensing Event
Abbreviated title	JURSE
Country/Territory	United Arab Emirates
City	Dubai
Period	6/03/17 → 8/03/17
Internet address	http://www.jurse2017.com/

Keywords

endmember libraries, land cover, Sentinel-2, sub-pixel mapping, Support Vector Regression, synthetic mixing, urban remote sensing

Access to Document

10.1109/JURSE.2017.7924554Licence: Unspecified

Cite this

@inproceedings{0f2a54afe3e14a5ebc5be44a9be92bc9,

title = "Optimizing mixed spectra generation for regression-based unmixing of land cover in urban areas",

abstract = "Regression-based unmixing for quantifying urban land cover at the subpixel scale requires mixed training spectra for model calibration. In this paper optimization and synthetic mixing of hyperspectral image endmember libraries for the calibration of unmixing models are investigated. APEX and HyMap airborne hyperspectral transects respectively covering Brussels and Berlin are used to produce an endmember library for unmixing, as well as reference land cover fractions for validation. The library is spectrally resampled, optimized and synthetically mixed to produce quantitative training data for unmixing of a Sentinel-2 surface reflectance image of Brussels (Belgium). Support Vector Regression models are developed for vegetation-impervious-soil land cover mapping. Findings may contribute to the use of multi-sensor data and to the demonstration of Sentinel-2's added value for quantitative urban land cover assessment.",

keywords = "endmember libraries, land cover, Sentinel-2, sub-pixel mapping, Support Vector Regression, synthetic mixing, urban remote sensing",

author = "Frederik Priem and Frank Canters and {Van der Linden}, Sebastian and Akpona Okujeni",

year = "2017",

doi = "10.1109/JURSE.2017.7924554",

language = "English",

isbn = "978-1-5090-5809-9 ",

pages = "1--4",

booktitle = "2017 Joint Urban Remote Sensing Event (JURSE)",

publisher = "IEEE",

edition = "2017",

note = "Joint Urban Remote Sensing Event, JURSE ; Conference date: 06-03-2017 Through 08-03-2017",

url = "http://www.jurse2017.com/",

}

Priem, F, Canters, F, Van der Linden, S & Okujeni, A 2017, Optimizing mixed spectra generation for regression-based unmixing of land cover in urban areas. in 2017 Joint Urban Remote Sensing Event (JURSE). 2017 edn, IEEE, Dubai, United Arab Emirates , pp. 1-4, Joint Urban Remote Sensing Event, Dubai, United Arab Emirates, 6/03/17. https://doi.org/10.1109/JURSE.2017.7924554

Optimizing mixed spectra generation for regression-based unmixing of land cover in urban areas. / Priem, Frederik; Canters, Frank; Van der Linden, Sebastian; Okujeni, Akpona.

2017 Joint Urban Remote Sensing Event (JURSE). 2017. ed. Dubai, United Arab Emirates : IEEE, 2017. p. 1-4.

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › Research

TY - GEN

T1 - Optimizing mixed spectra generation for regression-based unmixing of land cover in urban areas

AU - Priem, Frederik

AU - Canters, Frank

AU - Van der Linden, Sebastian

AU - Okujeni, Akpona

PY - 2017

Y1 - 2017

N2 - Regression-based unmixing for quantifying urban land cover at the subpixel scale requires mixed training spectra for model calibration. In this paper optimization and synthetic mixing of hyperspectral image endmember libraries for the calibration of unmixing models are investigated. APEX and HyMap airborne hyperspectral transects respectively covering Brussels and Berlin are used to produce an endmember library for unmixing, as well as reference land cover fractions for validation. The library is spectrally resampled, optimized and synthetically mixed to produce quantitative training data for unmixing of a Sentinel-2 surface reflectance image of Brussels (Belgium). Support Vector Regression models are developed for vegetation-impervious-soil land cover mapping. Findings may contribute to the use of multi-sensor data and to the demonstration of Sentinel-2's added value for quantitative urban land cover assessment.

AB - Regression-based unmixing for quantifying urban land cover at the subpixel scale requires mixed training spectra for model calibration. In this paper optimization and synthetic mixing of hyperspectral image endmember libraries for the calibration of unmixing models are investigated. APEX and HyMap airborne hyperspectral transects respectively covering Brussels and Berlin are used to produce an endmember library for unmixing, as well as reference land cover fractions for validation. The library is spectrally resampled, optimized and synthetically mixed to produce quantitative training data for unmixing of a Sentinel-2 surface reflectance image of Brussels (Belgium). Support Vector Regression models are developed for vegetation-impervious-soil land cover mapping. Findings may contribute to the use of multi-sensor data and to the demonstration of Sentinel-2's added value for quantitative urban land cover assessment.

KW - endmember libraries, land cover, Sentinel-2, sub-pixel mapping, Support Vector Regression, synthetic mixing, urban remote sensing

U2 - 10.1109/JURSE.2017.7924554

DO - 10.1109/JURSE.2017.7924554

M3 - Conference paper

SN - 978-1-5090-5809-9

SP - 1

EP - 4

BT - 2017 Joint Urban Remote Sensing Event (JURSE)

PB - IEEE

CY - Dubai, United Arab Emirates

T2 - Joint Urban Remote Sensing Event

Y2 - 6 March 2017 through 8 March 2017

ER -

Optimizing mixed spectra generation for regression-based unmixing of land cover in urban areas

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this