Analysing the performance and interpretability of CNN-based architectures for plant nutrient deficiency identification

Loading...
Thumbnail Image

Authors

Mkhatshwa, Junior
Kavu, Tatenda
Daramola, Olawande

Journal Title

Journal ISSN

Volume Title

Publisher

MDPI

Abstract

Early detection of plant nutrient deficiency is crucial for agricultural productivity. This study investigated the performance and interpretability of Convolutional Neural Networks (CNNs) for this task. Using the rice and banana datasets, we compared three CNN architectures (CNN, VGG16, Inception-V3). Inception-V3 achieved the highest accuracy (93% for rice and banana), but simpler models such as VGG-16 might be easier to understand. To address this trade-off, we employed Explainable AI (XAI) techniques (SHAP and Grad-CAM) to gain insights into model decision-making. This study emphasises the importance of both accuracy and interpretability in agricultural AI and demonstrates the value of XAI for building trust in these models.

Description

DATA AVAILABITY STATEMENT: The data are publicly available.

Keywords

Machine learning, Deep learning, Convolutional neural network, Plant nutrient deficiency, Explainable artificial intelligence, SDG-02: Zero hunger, SDG-09: Industry, innovation and infrastructure, Artificial intelligence (AI)

Sustainable Development Goals

SDG-02:Zero Hunger
SDG-09: Industry, innovation and infrastructure

Citation

Mkhatshwa, J.; Kavu, T.; Daramola, O. Analysing the Performance and Interpretability of CNN-Based Architectures for Plant Nutrient Deficiency Identification. Computation 2024, 12, 113. https://doi.org/10.3390/computation12060113.