Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Robust GOA-Driven CNN Parameter Optimization for Effective Rice Leaf Disease Classification

Rektorat UDINUS 2026 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2026
Goals
Record
af81e0ca-47c8-48c3-8e8a-4f9cb56e3b93
Published
2026-10-30

Universitas Dian Nuswantoro conducted research to improve rice leaf disease classification by optimizing the parameters of a Convolutional Neural Network. The activity addressed precision agriculture in relation to global challenges including population growth, declining fertile land, and climate change. It also considered the importance of accurate crop nutrient management and the difficulty of this task because plant populations have genetic diversity and environmental conditions fluctuate. Researchers participating in the activity were Pramunendar, Ricardus Anggi; Alzami, Farrikh; Wahyudi, Firman; Mambang; Sari, Yuslena; Alomoush, Ashraf; and Andono, Pulung Nurtantio.

The research used publicly available images of rice leaf diseases as its dataset. A CNN was employed because of its strength in visual pattern recognition. Several metaheuristic optimization algorithms, including the Grasshopper Optimization Algorithm and the Salp Swarm Algorithm, were applied together with other comparative methods. The algorithms were used to fine-tune the number of epochs, dropout rate, number of hidden neurons, learning rate, activation function, and optimizer.

The results showed that GOA achieved the highest accuracy of 72.88 percent using eight epochs, a 0.1 dropout rate, 512 neurons, a learning rate of 0.001, the ReLU activation function, and the Adam optimizer. SSA demonstrated strong generalization capability with faster training, while GTO offered a balanced trade-off between accuracy and computational cost. The findings suggest that optimization algorithms have potential to enhance plant disease classification systems. The source identifies hybrid optimization strategies and larger datasets as directions for improving robustness and practical applicability in field settings. The activity provides research evidence connected with the agricultural and climate-related objectives identified for this submission.

Reference https://www.scopus.com/pages/publications/105030684215?origin=resultslist