What Is Plant Disease Detection and Which Technology Works Best?

• Plant health monitoring can improve agricultural practices, profits, crop yield, safety, and quality.
• Non-destructive technologies for early disease detection on-site include spectroscopy and image processing.
• Among these techniques, vis-NIR spectroscopy-based devices are the most widely used for non-destructive, early detection of plant diseases.

Plant diseases result in the annual global loss of over 30% of crop yield and “hundreds of billions of dollars”. Hence, early detection of diseases is a priority as food production for a growing population is threatened on many fronts by uncontrollable factors like climate change-driven droughts and the scarcity of natural resources. Nowadays, the emphasis has been on non-destructive, real-time, easy methods of disease detection for efficient management of crop health. This article discusses the reasons behind the recent shift in technology and the most feasible options available.

Need for Early Crop Disease Detection

Plant diseases are caused by a wide range of above- and below-ground pathogens, including bacteria, fungi, viruses, and nematodes. Diseases not only reduce yield, but also the quality, nutritional value, and safety of food, affecting marketability and consumer confidence.

Pathogens reduce photosynthesis, nutrient uptake, plant growth, and vigor. Diseases also cause flower and fruit abnormalities, defoliation, and premature plant senescence. Growers also have to spend more time and money on disease control, which can reduce their ROI.

Early detection of diseases and severity allows growers to take quick, appropriate action to limit negative effects on crop health and their spread to neighboring plants. It also reduces chemical use, which improves the sustainability of food production by reducing the harmful effects of chemicals on the environment and costs.

Conventional Methods and Their Disadvantages

Early disease detection has always been a priority for growers and other stakeholders, and several conventional methods are widely used. However, these methods have many disadvantages, as listed below.

Visual symptoms: Growers use symptoms to detect and identify diseases. However, by the time symptoms are visible to the human eye, plant health is often compromised, requiring intensive use of chemicals or alternative methods to control the disease, and the possibility that it has spread to neighboring plants is high. Early detection is not possible by this method. Moreover, if growers do not have the skills and expertise to identify the disease, they may need outside expert advice.

Laboratory assessment: Laboratory tests are complex and require expensive equipment, expertise in the relevant pathology, and skilled personnel to conduct the analyses. Some methods used are as follows:

• Molecular methods include PCR-based detection, RNA interference, and microsclerotia morphological characteristics.
• Pathological methods involve taking plant tissue sections for microscopic examination of plant-pathogen interaction or response to diseases.
• Serological analyses test for antibodies produced to specific viruses; these include enzyme-linked immunosorbent assay (ELISA), dot blot immunobinding assay (DBIA), and tissue blot immuno assay (TBIA).
• Chromatography-based analyses are used to identify volatile organic compounds (VOCs) produced by plants in defense against pathogens. The VOCs are first extracted from the plant, then isolated by various chromatographic methods, after which the compounds’ makeup and properties are identified through mass spectrometry (MS), infrared spectroscopy (IR), or nuclear magnetic resonance (NMR).

A complete diagnosis by any of the above methods requires time. Also, laboratories are often far away, increasing waiting time for results due to transportation. Growers cannot use these methods often in a single season due to transport costs and waiting time, and they are not feasible for optimizing daily agricultural practice. Moreover, these methods require destructive sampling that can damage a plant.

Growers need an alternative to these laborious, time-consuming methods for early plant disease detection.

Non-destructive Disease Detection

Non-destructive methods that require no sampling and can repeatedly test plants and provide accurate, reliable results in real time are needed. Hence, many such technologies have recently been developed to enable early detection and increase the efficiency of daily crop management.

These technologies trace changes in plant tissue due to pathogen attacks, a form of biotic stress. They convert signals from plant physiological processes and biochemistry in response to pathogens into digital signals in real time. Since these technologies tap into internal processes, they can detect plant responses to diseases before they produce external symptoms visible to people. So, growers can take corrective measures much earlier than with conventional technologies and reduce disease impact on individual plants and overall yield. These detection methods are combined with advanced analytics algorithms that convert complex plant data into easy-to-understand insights growers can use immediately. These technologies help optimize agricultural practices, while simultaneously minimizing resource use and environmental damage. These methods are used in crop research and for commercial purposes, such as precision agriculture and precision forestry.

The main non-destructive technologies for on-site use in early disease detection are spectroscopy and image processing.

Figure 1: Electromagnetic spectrum of radiation reaching the Earth, NASA. (Image credits: https://mynasadata.larc.nasa.gov/basic-page/electromagnetic-spectrum-diagram)

Spectroscopy-based Approaches

Spectroscopy is a broad approach with several applications. It involves the interaction of the electromagnetic spectrum, which includes visible light, infrared, ultraviolet, and X- and gamma-ray wavelengths, with compounds in plant tissue; see Figure 1. When electromagnetic radiation hits the tissue, certain wavelengths are absorbed, reflected, or transmitted, and these spectra are measured to identify and quantify internal compounds.  Common spectral techniques in agriculture include near-infrared, Raman, and terahertz spectroscopy.

Near-infrared spectroscopy

The technique involves estimating the amount of near-infrared (NIR) light that is absorbed by the plant tissue.  NIR detects changes in the chemical composition of tissues to identify disease stress, since specific wavelengths are absorbed by different compounds, helping differentiate between healthy and diseased plants. Usually, leaves are the organs that are examined by NIR spectrometers. Leaf NIR spectroscopy also identifies changes in color, because many diseases affect pigmentation and chlorophyll content. Devices irradiate the plant with a chosen band and collect the response spectra. Portable devices have integrated software with algorithms for preprocessing and feature extraction, and a model to analyze complex spectral data and provide simple digital results.

NIR combined with visible (Vis) light spectroscopy has several advantages, as it is rapid, simple, highly precise, accurate, efficient, and low-cost. However, it can be affected by humidity, temperature, and differences in crop variety and growth stages. It has already been used for early warning and diagnosis of several diseases in various crops, such as aflatoxins in corn, root rot and yellow rust in wheat, late blight disease, powdery mildew in sugar beets, rice blast, Fusarium in barley, apple water heart disease, and oil palm base rot, to name a few.

Raman spectroscopy

Raman spectroscopy technology is highly sensitive and can identify pathogenic bacteria. The light scattered by a sample is measured to analyze molecular structure. Hence, it differentiates between healthy and diseased plants and diagnoses the causal pathogen and virus. It has been used for early warning of Leaf Curl Sardinia virus and spotted wilt virus in tomatoes, and Fusarium wilt and canker in oranges, to name a few.

The technique is more expensive than vis-NIR spectroscopy, and the wavelength penetration is shallower. The technique has yet to be popularized to fully exploit its potential.

Terahertz spectroscopy

Terahertz spectroscopy measures the absorption and emission by an object of terahertz waves, which lie between microwaves and infrared waves. It can reach deep tissue to analyze chemical components and provide information on internal structure. This method also leverages terahertz’s high sensitivity to water content.

Terahertz spectroscopy can give information on the severity of plant diseases. It is a method that has emerged in recent years and needs to be combined with other technologies like laser-induced breakdown spectroscopy (LIBS) to improve accuracy and efficiency.

A detailed comparison of the various spectroscopy techniques can be seen in Table 1.

Table 1: Comparison of different technologies, Wang et al. (2025). (Credits: https://www.mdpi.com/2077-0472/15/15/1670)

Techniques	Cost	Portability	Depth of Penetration	Humidity Sensitivity
Near-infrared spectroscopy	Lower, relatively inexpensive equipment and low operation cost, suitable for large-scale application	Higher, portable devices (such as handheld spectrometers) can be used in field sites	Medium, can obtain information on internal structure and composition of plant tissues, but with limited penetration depth	Higher, spectral information is susceptible to humidity, which may cause data fluctuations and affect detection accuracy
Raman spectroscopy	Medium, moderate cost for ordinary equipment, enhanced technologies (such as SERS) may be more expensive	Higher, handheld devices can be used for on-site detection	Shallow, mainly detects molecular vibration information on the surface or shallow layers of samples	Medium, humidity may have some impact on detection, but the degree of influence is relatively small
Terahertz Spectroscopy	Higher, expensive equipment limits its wide application	Lower, equipment is large in size and has poor portability, currently mainly used in laboratories	Deeper, can penetrate many non-conductive materials and obtain deeper information on plant tissues	Higher, terahertz waves are susceptible to humidity during propagation and detection performance may decline in high-humidity environments
Hyperspectral imaging	Higher, expensive equipment and high data processing and storage costs	Lower, equipment is usually large, although it can be combined with drones, overall portability is still limited, more used for laboratory fine analysis	Medium, can simultaneously obtain image and spectral information, with moderate penetration depth for plant tissues	Higher, data collection is affected by environmental factors such as light and weather, humidity may also have some impact
Digital Imaging	Lower, equipment (such as ordinary cameras, cameras on drones) has low cost, easy to obtain	Higher, equipment has good portability, can be collected on-site using smartphones, cameras, or drones	Shallow, mainly obtains color, texture, etc., features of plant surfaces	Lower, humidity has a relatively small impact on digital imaging
Thermal Imaging	Medium, moderate equipment cost, handheld devices and devices that can be mounted on drones	Higher, can be monitored on-site using handheld cameras or thermal imaging sensors mounted on drones	Shallow, mainly detects temperature changes on the surface of plants to identify diseases	Lower, humidity has a relatively small impact on thermal imaging

Imaging Technology

Imaging involves capturing images of an object or phenomenon using different principles and methods. These images are then analyzed by relevant algorithms for observation and diagnosis. Imaging is used to differentiate and quantify the diseased areas of a plant from the healthy parts. The various technologies used for imaging are photo, hyperspectral, digital, and thermal imaging.

Photo-based imaging

Photos can provide information on the morphology and pathology of plants. However, mathematical quantification requires complex and special techniques, so this method is not often used.

Digital images

This method overcomes the problems posed by image capture through photos. It uses optical imaging to convert analog image signals to digital signals. Enhancement, compression, and recognition can be performed to identify changes in color, texture, and shape of leaves to identify plant diseases.

Hyperspectral imaging

The technique is very sensitive and captures physiological changes within the plants. The method can detect diseases and their extent in the early stages by using reflection or emission of a continuous band of light from 400 to 2500 nm. The spectral data is analyzed by models that combine complex and precise processing techniques and algorithms to produce efficient deep learning models. The technique can be miniaturized into portable devices useful for commercial applications and for research in plant resistance breeding programs.

Thermal imaging

This technology relies on infrared imaging. It measures the infrared waves emitted by a body due to its temperature and converts them to electrical signals, which are then processed and reconstructed to show the thermal distribution of an object. When plants are attacked by pathogens, their temperature changes, which thermal imaging captures to detect diseases before visible signs appear. It is a non-contact technology and avoids cross-infection during monitoring. It allows quick scanning of large areas. The technique has great potential and is expected to play a greater role in early plant detection.

Besides these on-site techniques, remote sensing uses multispectral and hyperspectral data acquired through spectroscopy and imaging to cover large areas and provide early disease detection, disease intensity, and the spatial distribution of diseased plants in the field.

CI-710s SpectraVue Leaf Spectrometer

The CI-710s SpectraVue Leaf Spectrometer, offered by CID BioScience Inc., uses vis-NIR spectroscopy for rapid, nondestructive early detection of diseases. It is a portable device with a touchscreen display and integrated analysis software that includes many vegetative indices for stress detection. The device is accurate and precise enough for research and robust enough for field use by growers.

Contact us to find out more about the CI-710s SpectraVue Leaf Spectrometer for early nondestructive plant disease detection.

Sources

Abd El-Aziz, M. H. (2019). Three modern serological methods to detect plant viruses. Journal of Plant Science and Phytopathology, 3(3), 101–106. https://doi.org/10.29328/journal.jpsp.1001039

 

Ali, M. M., Bachik, N. A., Muhadi, N. A., Yusof, T. N. T., & Gomes, C. (2019). Non-destructive techniques of detecting plant diseases: A review. Physiological and Molecular Plant Pathology, 108, 101426.

 

Ang MC-Y and Lew TTS (2022) Non-destructive Technologies for Plant Health Diagnosis. Front. Plant Sci. 13:884454. doi: 10.3389/fpls.2022.884454

 

Chase, J., and Hroncich, C. (2025, June 17). Advances in Chromatographic Techniques for the Extraction and Characterization of Plant-Based Bioactive Compounds. Retrieved from https://www.chromatographyonline.com/view/advances-in-chromatographic-techniques-for-the-extraction-and-characterization-of-plant-based-bioactive-compounds

 

Gai, Y., & Wang, H. (2024). Plant Disease: A Growing Threat to Global Food Security. Agronomy, 14(8), 1615. https://doi.org/10.3390/agronomy14081615

 

Kutama, A. S., and Zafar, S. (2026). Socio-Economic Impacts Of Plant Diseases On Human Civilization: A Recap Of The Past, Present And The Future. MSI Journal of Multidisciplinary Research, ISSN 3049-0669 (Online). Retrieved from https://msipublishers.com/wp-content/uploads/2026/01/MSIJMR3032025-GS.pdf

 

Wang, Y., Sun, J., Wu, Z., Jia, Y., & Dai, C. (2025). Application of Non-Destructive Technology in Plant Disease Detection: Review. Agriculture, 15(15), 1670. https://doi.org/10.3390/agriculture15151670