Machine Learning for Asymptomatic Ratoon Stunting Disease Detection With Freely Available Satellite Based Multispectral Imaging

2024-10-04

arXiv.org

AI Summary

Early detection of asymptomatic ratoon stunting disease (RSD) in sugarcane remains challenging due to the absence of visible symptoms. Method: This study proposes a low-cost, large-scale remote sensing approach leveraging freely available multispectral satellite imagery and machine learning. A feature set comprising multiple vegetation indices—including NDVI and EVI—was extracted and used to systematically evaluate SVM (with RBF kernel), random forest, and gradient boosting models. Contribution/Results: SVM-RBF demonstrated superior performance for asymptomatic RSD identification—achieving classification accuracies of 85.64%–96.55% across diverse sugarcane cultivars—significantly outperforming conventional laboratory diagnostics. Crucially, the study reveals that integrating cultivar-specific information with vegetation indices markedly enhances model generalizability. This work establishes a scalable, transferable technical paradigm for remote sensing–based monitoring of latent crop diseases.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Remote Sensing / Geospatial AI
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Machine learning and data science for the Web
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Disease detection in sugarcane, particularly the identification of asymptomatic infectious diseases such as Ratoon Stunting Disease (RSD), is critical for effective crop management. This study employed various machine learning techniques to detect the presence of RSD in different sugarcane varieties, using vegetation indices derived from freely available satellite-based spectral data. Our results show that the Support Vector Machine with a Radial Basis Function Kernel (SVM-RBF) was the most effective algorithm, achieving classification accuracy between 85.64% and 96.55%, depending on the variety. Gradient Boosting and Random Forest also demonstrated high performance achieving accuracy between 83.33% to 96.55%, while Logistic Regression and Quadratic Discriminant Analysis showed variable results across different varieties. The inclusion of sugarcane variety and vegetation indices was important in the detection of RSD. This agreed with what was identified in the current literature. Our study highlights the potential of satellite-based remote sensing as a cost-effective and efficient method for large-scale sugarcane disease detection alternative to traditional manual laboratory testing methods.

Problem

• Detecting asymptomatic Ratoon Stunting Disease in sugarcane crops
• Using satellite-based multispectral imaging for disease identification
• Evaluating machine learning methods for accurate RSD detection

Innovation

• Machine learning for disease detection
• Satellite-based multispectral imaging data
• Support Vector Machine with RBF kernel