Autor(es):
Miyoshi, Gabriela Takahashi [UNESP] ; Arruda, Mauro dos Santos ; Osco, Lucas Prado ; Junior, José Marcato ; Gonçalves, Diogo Nunes ; Imai, Nilton Nobuhiro [UNESP] ; Tommaselli, Antonio Maria Garcia [UNESP] ; Honkavaara, Eija ; Gonçalves, Wesley Nunes
Data: 2020
Identificador Persistente: http://hdl.handle.net/11449/200401
Origem: Oasisbr
Assunto(s): Band selection; Convolutional neural network; Data-reduction; High-density object; Tree species identification; Band selection; Band selection; Convolutional neural network; Convolutional neural network; Data-reduction; Data-reduction; High-density object; High-density object; Tree species identification; Tree species identification
Descrição
Made available in DSpace on 2020-12-12T02:05:38Z (GMT). No. of bitstreams: 0 Previous issue date: 2020-04-01
Deep neural networks are currently the focus of many remote sensing approaches related to forest management. Although they return satisfactory results in most tasks, some challenges related to hyperspectral data remain, like the curse of data dimensionality. In forested areas, another common problem is the highly-dense distribution of trees. In this paper, we propose a novel deep learning approach for hyperspectral imagery to identify single-tree species in highly-dense areas. We evaluated images with 25 spectral bands ranging from 506 to 820 nm taken over a semideciduous forest of the Brazilian Atlantic biome. We included in our network's architecture a band combination selection phase. This phase learns from multiple combinations between bands which contributed the most for the tree identification task. This is followed by a feature map extraction and a multi-stage model refinement of the confidence map to produce accurate results of a highly-dense target. Our method returned an f-measure, precision and recall values of 0.959, 0.973, and 0.945, respectively. The results were superior when compared with a principal component analysis (PCA) approach. Compared to other learning methods, ours estimate a combination of hyperspectral bands that most contribute to the mentioned task within the network's architecture. With this, the proposed method achieved state-of-the-art performance for detecting and geolocating individual tree-species in UAV-based hyperspectral images in a complex forest.
Graduate Program in Cartographic Sciences São Paulo State University (UNESP)
Graduate Program in Computer Sciences Faculty of Computer Science Federal University of Mato Grosso do Sul (UFMS), Av. Costa e Silva
Faculty of Engineering and Architecture and Urbanism University ofWestern São Paulo (UNOESTE) Cidade Universitária, R. José Bongiovani
Faculty of Engineering Architecture and Urbanism and Geography Federal University of Mato Grosso do Sul (UFMS), Av. Costa e Silva
Department of Cartography São Paulo State University (UNESP)
Finnish Geospatial Research Institute National Land Survey of Finland, Geodeetinrinne 2
Graduate Program in Cartographic Sciences São Paulo State University (UNESP)
Department of Cartography São Paulo State University (UNESP)
