Using Deep Convolutional Networks for Species Identification of Xylotheque Samples

Geovanni Figueroa-Mata; Erick Mata-Montero; Juan Carlos Valverde-Otarola; Dagoberto Arias-Aguilar

doi:10.1109/IWOBI.2018.8464216

Using Deep Convolutional Networks for Species Identification of Xylotheque Samples

School of Mathematics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

15 Scopus citations

Abstract

Forest species identification is critical to scientifically support many environmental, commercial, forensic, archaeological, and paleontological actividades. Therefore, it is very important to develop fast and accurate identification systems. We present a deep CNN for automated forest species identification based on macroscopic images of wood cuts. We first implement and study a modified version of the LeNet convolutional network, which is trained from scratch with a database of macroscopic images of 41 forest species of the Brazilian flora. With this network we achieve a top-1 accuracy of 93.6%. Additionally, we fine-tune the Resnet50 model with pre-trained weights on Imagenet to reach a top-1 accuracy of 98.03%, which improves previous published results of research on the same image database.

Original language	English
Title of host publication	2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Print)	9781538675069
DOIs	https://doi.org/10.1109/IWOBI.2018.8464216
State	Published - 12 Sep 2018
Event	2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - San Carlos, Costa Rica Duration: 18 Jul 2018 → 20 Jul 2018

Publication series

Name	2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings

Conference

Conference	2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018
Country/Territory	Costa Rica
City	San Carlos
Period	18/07/18 → 20/07/18

Keywords

Automated plant identification
Biodiversity informatics
Convolutional neural networks
Deep learning
Forest species identification
Image processing

Access to Document

10.1109/IWOBI.2018.8464216

Cite this

Figueroa-Mata, G., Mata-Montero, E., Valverde-Otarola, J. C., & Arias-Aguilar, D. (2018). Using Deep Convolutional Networks for Species Identification of Xylotheque Samples. In 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings Article 8464216 (2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IWOBI.2018.8464216

Figueroa-Mata, Geovanni ; Mata-Montero, Erick ; Valverde-Otarola, Juan Carlos et al. / Using Deep Convolutional Networks for Species Identification of Xylotheque Samples. 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. (2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings).

@inproceedings{17828c473f344be3bee456a54709e2e5,

title = "Using Deep Convolutional Networks for Species Identification of Xylotheque Samples",

abstract = "Forest species identification is critical to scientifically support many environmental, commercial, forensic, archaeological, and paleontological actividades. Therefore, it is very important to develop fast and accurate identification systems. We present a deep CNN for automated forest species identification based on macroscopic images of wood cuts. We first implement and study a modified version of the LeNet convolutional network, which is trained from scratch with a database of macroscopic images of 41 forest species of the Brazilian flora. With this network we achieve a top-1 accuracy of 93.6\%. Additionally, we fine-tune the Resnet50 model with pre-trained weights on Imagenet to reach a top-1 accuracy of 98.03\%, which improves previous published results of research on the same image database.",

keywords = "Automated plant identification, Biodiversity informatics, Convolutional neural networks, Deep learning, Forest species identification, Image processing",

author = "Geovanni Figueroa-Mata and Erick Mata-Montero and Valverde-Otarola, \{Juan Carlos\} and Dagoberto Arias-Aguilar",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 ; Conference date: 18-07-2018 Through 20-07-2018",

year = "2018",

month = sep,

day = "12",

doi = "10.1109/IWOBI.2018.8464216",

language = "Ingl{\'e}s",

isbn = "9781538675069",

series = "2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings",

}

Figueroa-Mata, G, Mata-Montero, E, Valverde-Otarola, JC & Arias-Aguilar, D 2018, Using Deep Convolutional Networks for Species Identification of Xylotheque Samples. in 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings., 8464216, 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018, San Carlos, Costa Rica, 18/07/18. https://doi.org/10.1109/IWOBI.2018.8464216

Using Deep Convolutional Networks for Species Identification of Xylotheque Samples. / Figueroa-Mata, Geovanni; Mata-Montero, Erick; Valverde-Otarola, Juan Carlos et al.
2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. 8464216 (2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Using Deep Convolutional Networks for Species Identification of Xylotheque Samples

AU - Figueroa-Mata, Geovanni

AU - Mata-Montero, Erick

AU - Valverde-Otarola, Juan Carlos

AU - Arias-Aguilar, Dagoberto

PY - 2018/9/12

Y1 - 2018/9/12

N2 - Forest species identification is critical to scientifically support many environmental, commercial, forensic, archaeological, and paleontological actividades. Therefore, it is very important to develop fast and accurate identification systems. We present a deep CNN for automated forest species identification based on macroscopic images of wood cuts. We first implement and study a modified version of the LeNet convolutional network, which is trained from scratch with a database of macroscopic images of 41 forest species of the Brazilian flora. With this network we achieve a top-1 accuracy of 93.6%. Additionally, we fine-tune the Resnet50 model with pre-trained weights on Imagenet to reach a top-1 accuracy of 98.03%, which improves previous published results of research on the same image database.

AB - Forest species identification is critical to scientifically support many environmental, commercial, forensic, archaeological, and paleontological actividades. Therefore, it is very important to develop fast and accurate identification systems. We present a deep CNN for automated forest species identification based on macroscopic images of wood cuts. We first implement and study a modified version of the LeNet convolutional network, which is trained from scratch with a database of macroscopic images of 41 forest species of the Brazilian flora. With this network we achieve a top-1 accuracy of 93.6%. Additionally, we fine-tune the Resnet50 model with pre-trained weights on Imagenet to reach a top-1 accuracy of 98.03%, which improves previous published results of research on the same image database.

KW - Automated plant identification

KW - Biodiversity informatics

KW - Convolutional neural networks

KW - Deep learning

KW - Forest species identification

KW - Image processing

UR - https://www.scopus.com/pages/publications/85054521882

U2 - 10.1109/IWOBI.2018.8464216

DO - 10.1109/IWOBI.2018.8464216

M3 - Contribución a la conferencia

AN - SCOPUS:85054521882

SN - 9781538675069

T3 - 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings

BT - 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018

Y2 - 18 July 2018 through 20 July 2018

ER -

Figueroa-Mata G, Mata-Montero E, Valverde-Otarola JC , Arias-Aguilar D. Using Deep Convolutional Networks for Species Identification of Xylotheque Samples. In 2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. 8464216. (2018 IEEE International Work Conference on Bioinspired Intelligence, IWOBI 2018 - Proceedings). doi: 10.1109/IWOBI.2018.8464216

Using Deep Convolutional Networks for Species Identification of Xylotheque Samples

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this