Analysing the Performance and Interpretability of CNN-Based Architectures for Plant Nutrient Deficiency Identification

被引：3

作者：

Mkhatshwa, Junior ^{[1
]}

Kavu, Tatenda ^{[1
]}

Daramola, Olawande ^{[2
]}

机构：

[1] Cape Peninsula Univ Technol, Dept Informat Technol, Dist Six Campus,POB 8000, Cape Town, South Africa

[2] Univ Pretoria, Dept Informat, POB 0028, Pretoria, South Africa

来源：

COMPUTATION | 2024年 / 12卷 / 06期

关键词：

machine learning; deep learning; convolutional neural network; plant nutrient deficiency; explainable artificial intelligence; CONVOLUTIONAL NEURAL-NETWORKS;

D O I：

10.3390/computation12060113

中图分类号：

O1 [数学];

学科分类号：

0701 ; 070101 ;

摘要：

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, VGG-16, 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.

引用

页数：17