Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images
Tree species identification is critical to support their conservation, sustainable management and, particularly, the fight against illegal logging. Therefore, it is very important to develop fast and accurate identification systems even for non-experts. In this research we have achieved three main r...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-04-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2022.789227/full |
| _version_ | 1811328273939955712 |
|---|---|
| author | Geovanni Figueroa-Mata Erick Mata-Montero Juan Carlos Valverde-Otárola Juan Carlos Valverde-Otárola Dagoberto Arias-Aguilar Nelson Zamora-Villalobos |
| author_facet | Geovanni Figueroa-Mata Erick Mata-Montero Juan Carlos Valverde-Otárola Juan Carlos Valverde-Otárola Dagoberto Arias-Aguilar Nelson Zamora-Villalobos |
| author_sort | Geovanni Figueroa-Mata |
| collection | DOAJ |
| description | Tree species identification is critical to support their conservation, sustainable management and, particularly, the fight against illegal logging. Therefore, it is very important to develop fast and accurate identification systems even for non-experts. In this research we have achieved three main results. First, we developed—from scratch and using new sample collecting and processing protocols—an dataset called CRTreeCuts that comprises macroscopic cross-section images of 147 native tree species from Costa Rica. Secondly, we implemented a CNN for automated tree species identification based on macroscopic images of cross-sections of wood. For this CNN we apply the fine-tuning technique with VGG16 as a base model, pre-trained with the ImageNet data set. This model is trained and tested with a subset of 75 species from CRTreeCuts. The top-1 and top-3 accuracies achieved in the testing phase are 70.5% and 80.3%, respectively. The Same-Specimen-Picture Bias (SSPB), which is known to erroneously increase accuracy, is absent in all experiments. Finally, the third result is Cocobolo, an Android mobile application that uses the developed CNN as back-end to identify Costa Rican tree species from images of cross-sections of wood. |
| first_indexed | 2024-04-13T15:23:24Z |
| format | Article |
| id | doaj.art-7b3f40e72c4b47c9bf82712df240be68 |
| institution | Directory Open Access Journal |
| issn | 1664-462X |
| language | English |
| last_indexed | 2024-04-13T15:23:24Z |
| publishDate | 2022-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj.art-7b3f40e72c4b47c9bf82712df240be682022-12-22T02:41:35ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2022-04-011310.3389/fpls.2022.789227789227Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut ImagesGeovanni Figueroa-Mata0Erick Mata-Montero1Juan Carlos Valverde-Otárola2Juan Carlos Valverde-Otárola3Dagoberto Arias-Aguilar4Nelson Zamora-Villalobos5School of Mathematics, Costa Rica Institute of Technology, Cartago, Costa RicaSchool of Computing, Costa Rica Institute of Technology, Cartago, Costa RicaSchool of Forestry Engineering, Costa Rica Institute of Technology, Cartago, Costa RicaCooperativa de Productividad Forestal, Facultad de Ciencias Forestales, Universidad de Concepción, Concepción, ChileSchool of Forestry Engineering, Costa Rica Institute of Technology, Cartago, Costa RicaSchool of Forestry Engineering, Costa Rica Institute of Technology, Cartago, Costa RicaTree species identification is critical to support their conservation, sustainable management and, particularly, the fight against illegal logging. Therefore, it is very important to develop fast and accurate identification systems even for non-experts. In this research we have achieved three main results. First, we developed—from scratch and using new sample collecting and processing protocols—an dataset called CRTreeCuts that comprises macroscopic cross-section images of 147 native tree species from Costa Rica. Secondly, we implemented a CNN for automated tree species identification based on macroscopic images of cross-sections of wood. For this CNN we apply the fine-tuning technique with VGG16 as a base model, pre-trained with the ImageNet data set. This model is trained and tested with a subset of 75 species from CRTreeCuts. The top-1 and top-3 accuracies achieved in the testing phase are 70.5% and 80.3%, respectively. The Same-Specimen-Picture Bias (SSPB), which is known to erroneously increase accuracy, is absent in all experiments. Finally, the third result is Cocobolo, an Android mobile application that uses the developed CNN as back-end to identify Costa Rican tree species from images of cross-sections of wood.https://www.frontiersin.org/articles/10.3389/fpls.2022.789227/fulldeep learningconvolutional neural networkplant classificationautomated image-based tree species identificationcosta rican tree speciesxylotheques |
| spellingShingle | Geovanni Figueroa-Mata Erick Mata-Montero Juan Carlos Valverde-Otárola Juan Carlos Valverde-Otárola Dagoberto Arias-Aguilar Nelson Zamora-Villalobos Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images Frontiers in Plant Science deep learning convolutional neural network plant classification automated image-based tree species identification costa rican tree species xylotheques |
| title | Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images |
| title_full | Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images |
| title_fullStr | Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images |
| title_full_unstemmed | Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images |
| title_short | Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images |
| title_sort | using deep learning to identify costa rican native tree species from wood cut images |
| topic | deep learning convolutional neural network plant classification automated image-based tree species identification costa rican tree species xylotheques |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2022.789227/full |
| work_keys_str_mv | AT geovannifigueroamata usingdeeplearningtoidentifycostaricannativetreespeciesfromwoodcutimages AT erickmatamontero usingdeeplearningtoidentifycostaricannativetreespeciesfromwoodcutimages AT juancarlosvalverdeotarola usingdeeplearningtoidentifycostaricannativetreespeciesfromwoodcutimages AT juancarlosvalverdeotarola usingdeeplearningtoidentifycostaricannativetreespeciesfromwoodcutimages AT dagobertoariasaguilar usingdeeplearningtoidentifycostaricannativetreespeciesfromwoodcutimages AT nelsonzamoravillalobos usingdeeplearningtoidentifycostaricannativetreespeciesfromwoodcutimages |