Home > Research > Publications & Outputs > Reducing the effects of vegetation phenology on...

Electronic data

  • 061118_GISRS_clean_version

    Rights statement: This is an Accepted Manuscript of an article published by Taylor & Francis in GIScience and Remote Sensing on 28/11/2019, available online: https://www.tandfonline.com/doi/full/10.1080/15481603.2018.1550245

    Accepted author manuscript, 466 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes. / Silveira, Eduarda Martiniano de Oliveira; Espírito-Santo, Fernando Del Bon; Acerbi-Júnior, Fausto Weimar ; Galvão, Lênio Soares ; Withey, Kieran Daniel; Blackburn, George Alan; de Mello, José Márcio; Shimabukuro, Yosio Edemir ; Domingues, Tomas ; Scolforo, José Roberto Soares.

In: GIScience & Remote Sensing, Vol. 56, No. 5, 01.09.2019, p. 699-717.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Silveira, EMDO, Espírito-Santo, FDB, Acerbi-Júnior, FW, Galvão, LS, Withey, KD, Blackburn, GA, de Mello, JM, Shimabukuro, YE, Domingues, T & Scolforo, JRS 2019, 'Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes', GIScience & Remote Sensing, vol. 56, no. 5, pp. 699-717. https://doi.org/10.1080/15481603.2018.1550245

APA

Silveira, E. M. D. O., Espírito-Santo, F. D. B., Acerbi-Júnior, F. W., Galvão, L. S., Withey, K. D., Blackburn, G. A., de Mello, J. M., Shimabukuro, Y. E., Domingues, T., & Scolforo, J. R. S. (2019). Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes. GIScience & Remote Sensing, 56(5), 699-717. https://doi.org/10.1080/15481603.2018.1550245

Vancouver

Silveira EMDO, Espírito-Santo FDB, Acerbi-Júnior FW, Galvão LS, Withey KD, Blackburn GA et al. Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes. GIScience & Remote Sensing. 2019 Sep 1;56(5):699-717. https://doi.org/10.1080/15481603.2018.1550245

Author

Silveira, Eduarda Martiniano de Oliveira ; Espírito-Santo, Fernando Del Bon ; Acerbi-Júnior, Fausto Weimar ; Galvão, Lênio Soares ; Withey, Kieran Daniel ; Blackburn, George Alan ; de Mello, José Márcio ; Shimabukuro, Yosio Edemir ; Domingues, Tomas ; Scolforo, José Roberto Soares. / Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes. In: GIScience & Remote Sensing. 2019 ; Vol. 56, No. 5. pp. 699-717.

Bibtex

@article{a2c819b60a50462b9d5869620c1aa293,
title = "Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes",
abstract = "Tropical seasonal biomes (TSBs), such as the savannas (Cerrado) and semi-arid woodlands (Caatinga) of Brazil, are vulnerable ecosystems to human-induced disturbances. Remote sensing can detect disturbances such as deforestation and fires, but the analysis of change detection in TSBs is affected by seasonal modifications in vegetation indices due to phenology. To reduce the effects of vegetation phenology on changes caused by deforestation and fires, we developed a novel object-based change detection method. The approach combines both the spatial and spectral domains of the normalized difference vegetation index (NDVI), using a pair of Operational Land Imager (OLI)/Landsat-8 images acquired in 2015 and 2016. We used semivariogram indices (SIs) as spatial features and descriptive statistics as spectral features (SFs). We tested the performance of the method using three machine-learning algorithms: support vector machine (SVM), artificial neural network (ANN) and random forest (RF). The results showed that the combination of spatial and spectral information improved change detection by correctly classifying areas with seasonal changes in NDVI caused by vegetation phenology and areas with NDVI changes caused by human-induced disturbances. The use of semivariogram indices reduced the effects of vegetation phenology on change detection. The performance of the classifiers was generally comparable, but the SVM presented the highest overall classification accuracy (92.27%) when using the hybrid set of NDVI-derived spectral-spatial features. From the vegetated areas, 18.71% of changes were caused by human-induced disturbances between 2015 and 2016. The method is particularly useful for TSBs where vegetation exhibits strong seasonality and regularly spaced time series of satellite images are difficult to obtain due to persistent cloud cover.",
keywords = "remote sensing, geostatistics, seasonality, LULCC",
author = "Silveira, {Eduarda Martiniano de Oliveira} and Esp{\'i}rito-Santo, {Fernando Del Bon} and Acerbi-J{\'u}nior, {Fausto Weimar} and Galv{\~a}o, {L{\^e}nio Soares} and Withey, {Kieran Daniel} and Blackburn, {George Alan} and {de Mello}, {Jos{\'e} M{\'a}rcio} and Shimabukuro, {Yosio Edemir} and Tomas Domingues and Scolforo, {Jos{\'e} Roberto Soares}",
note = "This is an Accepted Manuscript of an article published by Taylor & Francis in GIScience and Remote Sensing on 28/11/2019, available online: https://www.tandfonline.com/doi/full/10.1080/15481603.2018.1550245",
year = "2019",
month = sep,
day = "1",
doi = "10.1080/15481603.2018.1550245",
language = "English",
volume = "56",
pages = "699--717",
journal = "GIScience & Remote Sensing",
issn = "1943-7226",
publisher = "Taylor & Francis",
number = "5",

}

RIS

TY - JOUR

T1 - Reducing the effects of vegetation phenology on change detection in tropical seasonal biomes

AU - Silveira, Eduarda Martiniano de Oliveira

AU - Espírito-Santo, Fernando Del Bon

AU - Acerbi-Júnior, Fausto Weimar

AU - Galvão, Lênio Soares

AU - Withey, Kieran Daniel

AU - Blackburn, George Alan

AU - de Mello, José Márcio

AU - Shimabukuro, Yosio Edemir

AU - Domingues, Tomas

AU - Scolforo, José Roberto Soares

N1 - This is an Accepted Manuscript of an article published by Taylor & Francis in GIScience and Remote Sensing on 28/11/2019, available online: https://www.tandfonline.com/doi/full/10.1080/15481603.2018.1550245

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Tropical seasonal biomes (TSBs), such as the savannas (Cerrado) and semi-arid woodlands (Caatinga) of Brazil, are vulnerable ecosystems to human-induced disturbances. Remote sensing can detect disturbances such as deforestation and fires, but the analysis of change detection in TSBs is affected by seasonal modifications in vegetation indices due to phenology. To reduce the effects of vegetation phenology on changes caused by deforestation and fires, we developed a novel object-based change detection method. The approach combines both the spatial and spectral domains of the normalized difference vegetation index (NDVI), using a pair of Operational Land Imager (OLI)/Landsat-8 images acquired in 2015 and 2016. We used semivariogram indices (SIs) as spatial features and descriptive statistics as spectral features (SFs). We tested the performance of the method using three machine-learning algorithms: support vector machine (SVM), artificial neural network (ANN) and random forest (RF). The results showed that the combination of spatial and spectral information improved change detection by correctly classifying areas with seasonal changes in NDVI caused by vegetation phenology and areas with NDVI changes caused by human-induced disturbances. The use of semivariogram indices reduced the effects of vegetation phenology on change detection. The performance of the classifiers was generally comparable, but the SVM presented the highest overall classification accuracy (92.27%) when using the hybrid set of NDVI-derived spectral-spatial features. From the vegetated areas, 18.71% of changes were caused by human-induced disturbances between 2015 and 2016. The method is particularly useful for TSBs where vegetation exhibits strong seasonality and regularly spaced time series of satellite images are difficult to obtain due to persistent cloud cover.

AB - Tropical seasonal biomes (TSBs), such as the savannas (Cerrado) and semi-arid woodlands (Caatinga) of Brazil, are vulnerable ecosystems to human-induced disturbances. Remote sensing can detect disturbances such as deforestation and fires, but the analysis of change detection in TSBs is affected by seasonal modifications in vegetation indices due to phenology. To reduce the effects of vegetation phenology on changes caused by deforestation and fires, we developed a novel object-based change detection method. The approach combines both the spatial and spectral domains of the normalized difference vegetation index (NDVI), using a pair of Operational Land Imager (OLI)/Landsat-8 images acquired in 2015 and 2016. We used semivariogram indices (SIs) as spatial features and descriptive statistics as spectral features (SFs). We tested the performance of the method using three machine-learning algorithms: support vector machine (SVM), artificial neural network (ANN) and random forest (RF). The results showed that the combination of spatial and spectral information improved change detection by correctly classifying areas with seasonal changes in NDVI caused by vegetation phenology and areas with NDVI changes caused by human-induced disturbances. The use of semivariogram indices reduced the effects of vegetation phenology on change detection. The performance of the classifiers was generally comparable, but the SVM presented the highest overall classification accuracy (92.27%) when using the hybrid set of NDVI-derived spectral-spatial features. From the vegetated areas, 18.71% of changes were caused by human-induced disturbances between 2015 and 2016. The method is particularly useful for TSBs where vegetation exhibits strong seasonality and regularly spaced time series of satellite images are difficult to obtain due to persistent cloud cover.

KW - remote sensing

KW - geostatistics

KW - seasonality

KW - LULCC

U2 - 10.1080/15481603.2018.1550245

DO - 10.1080/15481603.2018.1550245

M3 - Journal article

VL - 56

SP - 699

EP - 717

JO - GIScience & Remote Sensing

JF - GIScience & Remote Sensing

SN - 1943-7226

IS - 5

ER -