| Full text | |
| Author(s): |
Miyoshi, Gabriela Takahashi
[1]
;
Arruda, Mauro dos Santos
[2]
;
Osco, Lucas Prado
[3, 4]
;
Marcato Junior, Jose
[3]
;
Goncalves, Diogo Nunes
[2]
;
Imai, Nilton Nobuhiro
[1, 5]
;
Garcia Tommaselli, Antonio Maria
[1, 5]
;
Honkavaara, Eija
[6]
;
Goncalves, Wesley Nunes
[2, 3]
Total Authors: 9
|
| Affiliation: | [1] Sao Paulo State Univ UNESP, Grad Program Cartog Sci, BR-19060900 Presidente Prudente, SP - Brazil
[2] Fed Univ Mato Grosso do Sul UFMS, Fac Comp Sci, Grad Program Comp Sci, Av Costa & Silva, BR-79070900 Campo Grande, MS - Brazil
[3] Fed Univ Mato Grosso do Sul UFMS, Fac Engn Architecture & Urbanism & Geog, Av Costa & Silva, BR-79070900 Campo Grande, MS - Brazil
[4] Univ Western Sao Paulo UNOESTE, Fac Engn & Architecture & Urbanism, R Jose Bongiovani, Cidade Univ, BR-19050920 Presidente Prudente, SP - Brazil
[5] Sao Paulo State Univ UNESP, Dept Cartog, BR-19060900 Presidente Prudente, SP - Brazil
[6] Natl Land Survey Finland, Finnish Geospatial Res Inst, Geodeetinrinne 2, Masala 02430 - Finland
Total Affiliations: 6
|
| Document type: | Journal article |
| Source: | REMOTE SENSING; v. 12, n. 8 APR 2 2020. |
| Web of Science Citations: | 12 |
| Abstract | |
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. (AU) | |
| FAPESP's process: | 13/50426-4 - Unmanned airborne vehicle-based 4D remote sensing for mapping Rain Forest biodiversity and its change in Brazil (UAV_4D_Bio) |
| Grantee: | Antonio Maria Garcia Tommaselli |
| Support Opportunities: | Regular Research Grants |