Optical Screening of Citrus Leaf Diseases Using Label-Free Spectroscopic Tools: A Review

TL;DR

This review discusses optical spectroscopic techniques as promising, rapid, and non-destructive tools for early detection of citrus leaf diseases, highlighting their principles, applications, limitations, and future potential with AI integration.

Contribution

It provides a comprehensive overview of optical spectroscopy methods for citrus disease detection, emphasizing their technical specifications, applications, and future prospects including AI-based analysis.

Findings

Optical spectroscopic tools are sensitive, rapid, and suitable for field measurements.

AI can enhance spectral analysis for early disease detection.

Current methods are mostly in trial phases and need further development.

Abstract

Citrus diseases pose threats to citrus farming and result in economic losses worldwide. Nucleic acid and serology-based methods of detection and, immunochromatographic assays are commonly used but these laboratory tests are laborious, expensive and might be subjected to cross-reaction and contamination. Modern optical spectroscopic techniques offer a promising alternative as they are label-free, sensitive, rapid, non-destructive, and demonstrate the potential for incorporation into an autonomous system for disease detection in citrus orchards. Nevertheless, the majority of optical spectroscopic methods for citrus disease detection are still in the trial phases and, require additional efforts to be established as efficient and commercially viable methods. The review presents an overview of fundamental working principles, the state of the art, and explains the applications and limitations of the optical spectroscopy technique including the spectroscopic imaging approach (hyperspectral imaging) in the identification of diseases in citrus plants. The review highlights (1) the technical specifications of optical spectroscopic tools that can potentially be utilized in field measurements, (2) their applications in screening citrus diseases through leaf spectroscopy, and (3) discusses their benefits and limitations, including future insights into label-free identification of citrus diseases. Moreover, the role of artificial intelligence is reviewed as potential effective tools for spectral analysis, enabling more accurate detection of infected citrus leaves even before the appearance of visual symptoms by leveraging compositional, morphological, and chemometric characteristics of the plant leaves. The review aims to encourage stakeholders to enhance the development and commercialization of field-based, label-free optical tools for the rapid and early-stage screening of citrus diseases in plants.