This study demonstrates improved disease severity estimation in wheat leaves using CNN ensemble methods, suggesting deep learning's potential in agricultural monitoring.
The estimation of plant disease intensity is essential for various purposes, including monitoring epidemics, understanding yield loss, and evaluating treatment effects. Despite the availability of sensor technology to measure disease severity using the visible spectrum or other spectral range imaging, deep learning has emerged as a recent and advanced technique for image processing and data analysis. To enhance the severity estimation in diseased leaves, a CNN ensemble is proposed by fusing deep featuresextracted from outputs of fully connected layers of various CNN models. A SVM (Support Vector Machine) is utilized to achieve the classification stage. Experiments are carried out on wheat leaf images infected by the Yellow rust disease. Experiments conducted using three CNN models that are VGG16, MobileNetV2 and a customized CNN reveal that the CNN ensemble outperforms individual models.
No takes yet. Share an insight, caveat, or question.
Lakehal et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: