Answers are grounded in this paper and cite their sources.
Rice is a crucial food commodity worldwide, particularly in Asian countries. However, various factors, such as drought, floods, and pest attacks, can lead to the emergence of diseases in rice plants. Accurately identifying these diseases poses a significant challenge for farmers, often leading to significant yield losses. Conventionally, farmers rely on manual methods based on their experience and visual inspections to identify rice diseases. However, this approach is highly ineffective, time-consuming, and prone to error. This study aimed to address this issue by proposing advanced deep learning techniques, an ensemble learning method, to automate and enhance the identification of rice plant diseases. The ensemble learning method was proposed by leveraging two state-of-the-art pre-trained models: EfficientNetV2B0 and MobileNetV3-Large. The proposed Average Ensemble method demonstrates superior performance compared with single models. The proposed Average Ensemble achieved superior performance with an average precision of 0.9339, a recall of 0.9330, an F1-score of 0.9328, and a test accuracy of 0.9330. The results of this study can be used to aid farmers and researchers in accurately identifying rice diseases, ultimately supporting better disease management practices, and enhancing the agricultural productivity.
No takes yet. Share an insight, caveat, or question.
Pratama et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: