Leaf Based Plant Species Classification Using Deep Convolutional Neural Network

• Published in: 2022 10th International Conference on Information and Communication Technology, ICoICT 2022
• Main Author: Minarno A.E.; Ibrahim Z.; Nur A.; Hasanuddin M.Y.; Diah N.M.; Munarko Y.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85141595162&doi=10.1109%2fICoICT55009.2022.9914851&partnerID=40&md5=d856d9cc71571fc6b654911e6b95d8b4
• DOI: 10.1109/ICoICT55009.2022.9914851

Kingdom Plantae consists of hundreds of thousands of species that play a critical role in ensuring the survival of living things on earth. However, the International Union for Conservation of Nature (IUCN) pointed out that around 34,000 of the recorded species were threatened with extinction until 1997. Therefore, it is deemed essential to conserve biodiversity with the initial attempt by crafting a system that identifies plants on earth, conducted through utilizing a computer vision to determine plant identity based on leaf images. This paper proposes a model to identify plants based on leaf images by implementing Convolutional Neural Network (CNN. We tested the model to the Flavia dataset and got 95.3% accuracy. Additionally, this paper presents a leaf dataset recognized as the Indonesian Herb Leaf Dataset (IHLD), which aims to compare the accuracy and to measure the robustness of the proposed model on a different dataset. © 2022 IEEE.