Discriminating Vegetation Cover Classes of na Amazon/Cerrado Transition Region in the Mato Grosso State Using ALOS-2/PALSAR-2 Satellite Images
Main Article Content
Abstract
The transition region between Amazonia and Cerrado, especially in the Mato Grosso State, is environmentally sensitive because of the high levels of biodiversity and high grain and cattle beef productions. The objective of this study was to discriminate representative land use and land cover (LULC) classes found in the Sinop/Mato Grosso region based on the ALOS-2/PALSAR-2 satellite images. The following thematic classes were considered: primary forest, secondary forest, croplands, and cultivated pasturelands. The scenes were obtained in February (rainy season) and September (dry season) of 2017, with spatial resolution of 6.25 m, HH and HV polarizations, and L-band (wavelength of 23 cm) and processed by the Random Forest (RF) and Support Vector Machine (SVM) classifiers. Training and validation samples were obtained in the field (65 samples) and complemented based on the LULC maps produced by the MapBiomas and TerraClass projects (135 samples). It was possible to discriminate two groups of thematic classes: primary forest and secondary forest; and croplands and pasturelands. Although the overall accuracy of RF was higher than that from SVM, both classifiers presented statistically similar performances.
Downloads
Metrics
Article Details
Authors who publish in this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors can enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) before and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (see "The Effect of Open Access").