Implementation of CNN for Plant Leaf Classification

  • Mohammad Diqi Universitas Respati Yogyakarta
  • Sri Hasta Mulyani

Abstract

Many deep learning-based approaches for plant leaf stress identification have been proposed in the literature, but there are only a few partial efforts to summarize various contributions. This study aims to build a classification model to enable people or traditional medicine experts to detect medicinal plants by using a scanning camera. This Android-based application implements the Java programming language and labels using the Python programming language to build deep learning applications. The study aims to construct a deep learning model for image classification for plant leaves that can help people determine the types of medicinal plants based on android. This research can help the public recognize five types of medicinal plants, including spinach Duri, Javanese ginseng, Dadap Serep, and Moringa. In this study, the accuracy is 0.86, precision 0.22, f-1 score 0.23, while recall is 0.2375.

Downloads

Download data is not yet available.
Published
2021-03-10
How to Cite
DIQI, Mohammad; MULYANI, Sri Hasta. Implementation of CNN for Plant Leaf Classification. International Journal of Informatics and Computation, [S.l.], v. 2, n. 2, p. 1-10, mar. 2021. ISSN 2714-5263. Available at: <http://ijicom.respati.ac.id/index.php/ijicom/article/view/28>. Date accessed: 16 oct. 2021. doi: https://doi.org/10.35842/ijicom.v2i2.28.