File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Conference Paper: Fire risk assessment in the Brazilian Amazon using MODIS imagery and change vector analysis
Title | Fire risk assessment in the Brazilian Amazon using MODIS imagery and change vector analysis |
---|---|
Authors | |
Keywords | Brazilian Amazon Change vector analysis Decision tree Forest fire |
Issue Date | 2009 |
Citation | Proceedings, 33rd International Symposium on Remote Sensing of Environment, ISRSE 2009, 2009, p. 575-578 How to Cite? |
Abstract | This research applied the Change Vector Analysis (CVA) in areas with high occurrence of forest fires in the Brazilian Amazon, aiming to recognize changes that could: 1) Identify areas with high risk of being burnt and 2) Improve current fire scars mapping methods by allowing the discrimination of fires in primary forests and fires in previously burnt areas. A Decision trees (DT) was designed and evaluated through the C 4.5 algorithm to classify sample pixels extracted from four selected classes inside the change vector images: A) Forest; B) Agricultural areas; C) Forests to be burnt and D) Already degraded areas to be re-burnt. The DT achieved a global accuracy of 90.21%. Samples from classes B and D were the main responsible for the DT confusion, with omission errors of 9.5% and 24.5%, respectively. |
Persistent Identifier | http://hdl.handle.net/10722/309208 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Maeda, E. E. | - |
dc.contributor.author | Arcoverde, G. F.B. | - |
dc.contributor.author | Pellikka, P. | - |
dc.contributor.author | Shimabukuro, Y. E. | - |
dc.date.accessioned | 2021-12-15T03:59:45Z | - |
dc.date.available | 2021-12-15T03:59:45Z | - |
dc.date.issued | 2009 | - |
dc.identifier.citation | Proceedings, 33rd International Symposium on Remote Sensing of Environment, ISRSE 2009, 2009, p. 575-578 | - |
dc.identifier.uri | http://hdl.handle.net/10722/309208 | - |
dc.description.abstract | This research applied the Change Vector Analysis (CVA) in areas with high occurrence of forest fires in the Brazilian Amazon, aiming to recognize changes that could: 1) Identify areas with high risk of being burnt and 2) Improve current fire scars mapping methods by allowing the discrimination of fires in primary forests and fires in previously burnt areas. A Decision trees (DT) was designed and evaluated through the C 4.5 algorithm to classify sample pixels extracted from four selected classes inside the change vector images: A) Forest; B) Agricultural areas; C) Forests to be burnt and D) Already degraded areas to be re-burnt. The DT achieved a global accuracy of 90.21%. Samples from classes B and D were the main responsible for the DT confusion, with omission errors of 9.5% and 24.5%, respectively. | - |
dc.language | eng | - |
dc.relation.ispartof | Proceedings, 33rd International Symposium on Remote Sensing of Environment, ISRSE 2009 | - |
dc.subject | Brazilian Amazon | - |
dc.subject | Change vector analysis | - |
dc.subject | Decision tree | - |
dc.subject | Forest fire | - |
dc.title | Fire risk assessment in the Brazilian Amazon using MODIS imagery and change vector analysis | - |
dc.type | Conference_Paper | - |
dc.identifier.scopus | eid_2-s2.0-84879981283 | - |
dc.identifier.spage | 575 | - |
dc.identifier.epage | 578 | - |