1. Identity statement | |
Reference Type | Journal Article |
Site | plutao.sid.inpe.br |
Holder Code | isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S |
Identifier | 8JMKD3MGP3W/43NH7AB |
Repository | sid.inpe.br/plutao/2020/12.07.14.57 (restricted access) |
Last Update | 2020:12.08.12.38.39 (UTC) lattes |
Metadata Repository | sid.inpe.br/plutao/2020/12.07.14.57.17 |
Metadata Last Update | 2022:01.04.01.31.24 (UTC) administrator |
DOI | 10.5194/isprs-annals-v-3-2020-193-2020 |
ISSN | 0924-2716 |
Label | lattes: 1861914973833506 3 SötheLAGSCFDLLMT:2020:EVCONE |
Citation Key | SötheLAGSCFDLLMT:2020:EvCoNe |
Title | Evaluating a convolutional neural network for feature extraction and tree species classification using uav-hyperspectral images |
Year | 2020 |
Access Date | 2024, May 18 |
Type of Work | journal article |
Secondary Type | PRE PI |
Number of Files | 1 |
Size | 1064 KiB |
|
2. Context | |
Author | 1 Söthe, Camile 2 La Rosa, L. E. C. 3 Almeida, Cláudia Maria de 4 Gonsamo, A. 5 Schimalski, Marcos Benedito 6 Castro, J. D. B. 7 Feitosa, Raul Queiroz 8 Dalponte, Michele 9 Lima, Carla Luciane 10 Liesenberg, Veraldo 11 Miyoshi, Gabriela Takahashi 12 Tommaselli, Antonio Maria Garcia |
Resume Identifier | 1 2 3 8JMKD3MGP5W/3C9JGS3 |
Group | 1 2 3 DIDSR-CGOBT-INPE-MCTIC-GOV-BR |
Affiliation | 1 McMaster University 2 Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio) 3 Instituto Nacional de Pesquisas Espaciais (INPE) 4 McMaster University 5 Universidade do Estado de Santa Catarina (UDESC) 6 Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio) 7 Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio) 8 Fondazione Edmund Mach 9 Universidade do Estado de Santa Catarina (UDESC) 10 Universidade do Estado de Santa Catarina (UDESC) 11 Universidade Estadual Paulista (UNESP) 12 Universidade Estadual Paulista (UNESP) |
Author e-Mail Address | 1 sothec@mcmaster.ca 2 lauracuerosa@gmail.com 3 claudia.almeida@inpe.br 4 gonsamoa@mcmaster.ca 5 marcos.schimalski@udesc.br 6 bermudezjosedavid@gmail.com 7 raul@ele.puc-rio.br 8 michele.dalponte@fmach.it 9 carla_engflorestal@yahoo.com.br 10 veraldo@gmail.com 11 takahashi.gabi@gmail.com 12 a.tommaselli@gmail.com |
Journal | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Volume | 3 |
Pages | 193-199 |
History (UTC) | 2020-12-08 12:38:39 :: lattes -> administrator :: 2020 2022-01-04 01:31:24 :: administrator -> simone :: 2020 |
|
3. Content and structure | |
Is the master or a copy? | is the master |
Content Stage | completed |
Transferable | 1 |
Content Type | External Contribution |
Version Type | publisher |
Keywords | Tropical diversity unmanned aerial vehicle deep learning convolutional neural networks support vector machine data augmentation |
Abstract | The classification of tree species can significantly benefit from high spatial and spectral information acquired by unmanned aerial vehicles (UAVs) associated with advanced feature extraction and classification methods. Different from the traditional feature extraction methods, that highly depend on users knowledge, the convolutional neural network (CNN)-based method can automatically learn and extract the spatial-related features layer by layer. However, in order to capture significant features of the data, the CNN classifier requires a large number of training samples, which are hardly available when dealing with tree species in tropical forests. This study investigated the following topics concerning the classification of 14 tree species in a subtropical forest area of Southern Brazil: i) the performance of the CNN method associated with a previous step to increase and balance the sample set (data augmentation) for tree species classification as compared to the conventional machine learning methods support vector machine (SVM) and random forest (RF) using the original training data; ii) the performance of the SVM and RF classifiers when associated with a data augmentation step and spatial features extracted from a CNN. Results showed that the CNN classifier outperformed the conventional SVM and RF classifiers, reaching an overall accuracy (OA) of 84.37% and Kappa of 0.82. The SVM and RF had a poor accuracy with the original spectral bands (OA 62.67% and 59.24%) but presented an increase between 14% and 21% in OA when associated with a data augmentation and spatial features extracted from a CNN. |
Area | SRE |
Arrangement | urlib.net > BDMCI > Fonds > Produção anterior à 2021 > DIDSR > Evaluating a convolutional... |
doc Directory Content | access |
source Directory Content | there are no files |
agreement Directory Content | there are no files |
|
4. Conditions of access and use | |
Language | en |
Target File | sothe_evaluating.pdf |
User Group | lattes |
Reader Group | administrator lattes |
Visibility | shown |
Archiving Policy | denypublisher denyfinaldraft24 |
Read Permission | deny from all and allow from 150.163 |
Update Permission | not transferred |
|
5. Allied materials | |
Next Higher Units | 8JMKD3MGPCW/3ER446E |
Citing Item List | sid.inpe.br/bibdigital/2013/09.13.21.11 5 sid.inpe.br/mtc-m21/2012/07.13.14.43.49 1 |
URL (untrusted data) | http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/V-3-2020/193/2020/ |
Dissemination | WEBSCI; PORTALCAPES; COMPENDEX; SCOPUS. |
Host Collection | dpi.inpe.br/plutao@80/2008/08.19.15.01 |
|
6. Notes | |
Notes | Setores de Atividade: Atividades dos serviços de tecnologia da informação, Produção Florestal, Pesquisa e desenvolvimento científico. |
Empty Fields | alternatejournal archivist callnumber copyholder copyright creatorhistory descriptionlevel e-mailaddress format isbn lineage mark mirrorrepository month nextedition number orcid parameterlist parentrepositories previousedition previouslowerunit progress project rightsholder schedulinginformation secondarydate secondarykey secondarymark session shorttitle sponsor subject tertiarymark tertiarytype |
|
7. Description control | |
e-Mail (login) | simone |
update | |
|