Land Use/Land Cover Classification In Amazon Using Alos Palsar Polarimetric Data
Luciano Vieira Dutra(1), Corina Costa Freitas(1), Sidnei Sant'Anna(1), Graziela Scofield(1), Sumaia Aboud(1), Rogerio Negri(1), Joao Roberto Santos(1), Jose Claudio Mura(1) and Marcos Timbo Elmiro(1)
(1) National Institute for Space Research of Brazil, Av. dos Astronautas, 1758, 12227-010 Sao Jose dos Campos, SP, Brazil
The importance of Microwave Remote Sensing for tropical forest monitoring is well known. The new generation of orbital SAR plataforms is capable of delivering several types of polarimetric products, but the full polarimetric mode is of restricted use because of operational limitations. This investigation is focused on determining how much the use of full polarized (HH, VV and HV) PALSAR data information can improve, or not, the overall classification accuracy in comparison with the standard products, which for Palsar instrument is the HH (like JERS-1) or the dual polarization product HH-HV. Particular attention is given to the discrimination capability of perceiving land cover alterations of great importance on tropical environment monitoring, like recent deforestation, degraded forest and regeneration, as a function of the channels set used. Several types of supervised classifiers are tested for having, as much as possible, an assessment rather independent of the classifier type. The classification results improvements are statistically tested for significance. Results showed that HH-HV combination has discriminatory power significantly better than using any only one channel, but classification accuracy, using this particular combination, is not significantly better than using all 3 channels, HH, VV and HV. Phase influence is now under evaluation.
Two study areas are presently under assessment. The first one, Tapajós National Forest at the south of Santarém City, in the Brazilian Amazon, Pará State, has being object of intensive scientific observation for more than 15 years. The second area is Sao Félix do Xingu, also in Pará State, where the deforestation pressure is important.
Initial results indicate that, no phase considered the dual polarizarition product HH-HV is the better combination for general mapping of the rain forest problem composed by the primary forest, secondary forest, bare soil, agriculture and degraded forest. Region based classification, particularly the one developed to take in account as much as possible the radar statistical behavior, performed better in this experiment. The results for Sao Felix area and those including phase information is now under evaluation and will be timely reported.