AUTHOR=Tariq Maria , Ali Usman , Abbas Sagheer , Hassan Shahzad , Naqvi Rizwan Ali , Khan Muhammad Adnan , Jeong Daesik 

TITLE=Corn leaf disease: insightful diagnosis using VGG16 empowered by explainable AI

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 15 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2024.1402835

DOI=10.3389/fpls.2024.1402835

ISSN=1664-462X

ABSTRACT=An intelligent agricultural system is vital to food and the economy. Diseases harm crops and reduce yields. Early detection and classification can minimize further damage. The model's ability to explain its decisions can ensure trust and credibility in machine learning models. However, Deep Learning(DL) models pose a significant challenge regarding explainability. Layer-wise Relevance Propagation (LRP) is a well-established explainability technique aimed at deep models in computer vision. It generates heat maps of input images that are intuitive and human-readable. This paper uses the Visual Geometry Group 16(VGG16) model to classify corn leaf diseases.The dataset contains four classes: blight, gray spot, common rust, and healthy. LRP is used to interpret the decision-making process of the VGG16 model and enhance its interpretability. The simulation results demonstrate that the proposed approach is more accurate as compared to previously published approaches.