Identification of Forest Species with YOLO: A Study Based in Leaf Images
Abstract
Managing and protecting natural resources is a primary concern for countries. The threat of climate change highlights the importance of forest conservation, and the first step towards preventing more damage is environmental education. The following work aims to create a deep learning model capable of identifying the species of a tree by its leaves. By utilizing an available dataset of 16 different species and a total of 698 images, with the samples taken from a controlled environment, to train the object detection model using YOLO, the goal is to provide a tool that can facilitate and provide extra assurance in identifying species on any location and making it possible to study forest areas that receive less attention from formal organizations and have fewer resources to put towards research. The model presents 93% accuracy with a confidence rate of 60%.
References
R. A. Kerr, “Global warming is changing the world,” Science, vol. 316, no. 5822, pp. 188–190, 2007.
P. A. L. MACHADO and M. A. d. S. ARAGÃO, Principios de Direito Ambiental. Editora Jus Podivm, 2022.
A. C. P. Mazzei, T. M. F. Floripes, and M. T. M. Feitosa, “A importância da educação ambiental no período operatório concreto para a sustentabilidade existencial,” in PSICOLOGIA: UM OLHAR DO MUNDO REAL VOLUME 1, vol. 1. Editora Científica Digital, 2020, pp. 18–25.
W. O. Pires, R. C. Fernandes, P. L. de Paula Filho, A. Candido Junior, and J. P. Teixeira, “Leaf-based species recognition using convolutional neural networks,” in Optimization, Learning Algorithms and Applications: First International Conference, OL2A 2021, Bragança, Portugal, July 19–21, 2021, Revised Selected Papers 1. Springer, 2021, pp. 367–380.
N. R. Da Silva, M. W. d. S. Oliveira, H. A. d. A. Filho, L. F. S. Pinheiro, D. R. Rossatto, R. M. Kolb, and O. M. Bruno, “Leaf epidermis images for robust identification of plants,” Scientific reports, vol. 6, no. 1, p. 25994, 2016.
R. C. V. Martins-da Silva, M. G. Hopkins, and I. S. Thompson, “Identificação botânica na Amazônia: situação atual e perspectivas.” Belém, PA: Embrapa Amazônia Oriental., 2003.
H. Jiang and E. Learned-Miller, “Face detection with the faster r-cnn,” in 2017 12th IEEE international conference on automatic face & gesture recognition (FG 2017). IEEE, 2017, pp. 650–657.
Y. Liu, P. Sun, N. Wergeles, and Y. Shang, “A survey and performance evaluation of deep learning methods for small object detection,” Expert Systems with Applications, vol. 172, p. 114602, 2021.
G. Jocher, A. Chaurasia, and J. Qiu, “Ultralytics yolov8,” 2023. [Online]. Available: [link]
B. Dwyer, J. Nelson, T. Hansen, and et al., “Roboflow (version 1.0),” 2024, computer vision software. [Online]. Available: [link]
C. Champagne and N. Sinha, “Compound leaves: equal to the sum of their parts?” Oxford University Press for The Company of Biologists Limited, 2004.
O. Mzoughi, I. Yahiaoui, N. Boujemaa, and E. Zagrouba, “Multiple leaflets-based identification approach for compound leaf species.” in EMR@ ICMR, 2014, pp. 53–60.
P. E. R. Carvalho, Espécies arbóreas brasileiras. Embrapa, 2008.
R. Rothe, M. Guillaumin, and L. Van Gool, “Non-maximum suppression for object detection by passing messages between windows,” in Computer Vision–ACCV 2014: 12th Asian Conference on Computer Vision, Singapore, Singapore, November 1-5, 2014, Revised Selected Papers, Part I 12. Springer, 2015, pp. 290–306.
D. M. Hawkins, “The problem of overfitting,” Journal of chemical information and computer sciences, vol. 44, no. 1, pp. 1–12, 2004
