EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Disease Identification on Fig Leaf Images Using Deep Learning Method

Waeisul Bismi, Dwiza Riana, Alya Shafira Hewiz

International Journal of Advanced Science Computing and Engineering · 07.07.2024 · 6 (2) · 57-63

DOI: 10.62527/ijasce.6.2.203 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Advanced Science Computing and Engineering, online erschienen am 07.07.2024. Verfasst von Waeisul Bismi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The fig plant, known as Ficus carica, has been cultivated worldwide, including in Indonesia. It has nutritional benefits and medicinal properties. However, there are still difficulties in growing it, making the plant scarce. The scarcity of fig plants in Indonesia is mainly due to the threat of diseases and viruses that affect them. There are various diseases that affect fig plants, including leaf rust (Cerotelium fici), mosaic disease, and Bemisia tabaci (whitefly) disease. Infected fig plants become unhealthy, experiencing stunted growth and deformed fruits thus it is necessary to identify the diseases accurately using technological assistance. This research aimed to automatically identify diseases in fig leaves. The method started by digitizing fig leaf images and confirming with botanical experts specializing in fig plants to label the types of diseases present. The research produced a dataset of fig leaf images consisting of four classes of fig leaves: Cerotelium fici, mosaic disease, whitefly, and healthy fig leaves. The dataset resulted in confirmed 300 fig leaf images. The augmentation techniques were applied to increase images to 3,300 fig leaf images. This dataset was then divided into subsets for training, validation, and testing. For the classification and identification, a Deep Learning approach was used with three models: VGG16, VGG19, and MobileNet. Among these models, MobileNet achieved the highest accuracy of 98.79%. Subsequently, the identification system was implemented by converting the generated model into TensorFlow Lite and integrating it into the Android Studio software, enabling it to function as a mobile application on Android devices.

Zitieren

APA

Bismi, W., Riana, D., & Hewiz, A. S. (2024). Disease Identification on Fig Leaf Images Using Deep Learning Method. International Journal of Advanced Science Computing and Engineering, 6(2), 57-63. https://doi.org/10.62527/ijasce.6.2.203

Vancouver

Bismi W, Riana D, Hewiz AS. Disease Identification on Fig Leaf Images Using Deep Learning Method. International Journal of Advanced Science Computing and Engineering. 2024;6(2):57-63. doi:10.62527/ijasce.6.2.203

BibTeX
@article{Bismi2024,
  author = {Bismi, Waeisul and Riana, Dwiza and Hewiz, Alya Shafira},
  title = {Disease Identification on Fig Leaf Images Using Deep Learning Method},
  journal = {International Journal of Advanced Science Computing and Engineering},
  year = {2024},
  volume = {6},
  number = {2},
  pages = {57--63},
  doi = {10.62527/ijasce.6.2.203}
}
Kostenlos starten →