2.4 Pests monitoring and forecasting

Crop Disease Monitoring


Monitoring plants and diseases in their early stages is of paramount importance, as it can prevent damage and allow for early action. In the past, detection and cure of plant diseases was done by the expert in the field. Disease monitoring requires a tremendous amount of work and time. Disease identification and diagnosis can be done directly on the plant. Manual monitoring of diseases does not give the desired result as naked eye observation is unreliable and increases the chances of misdiagnosis. It also requires the attention of an expert, which is time consuming and expensive. Therefore, manual methods are ineffective. Automatic and instantaneous plant disease detection is important to detect the symptoms of diseases at early stages when they appear on the growing leaf of the plant. It is used to segment the leaf, extract characteristics and classify it based on its appearance.

Some approaches focus on establishing a plant disease surveillance network based on the use of mobile phones, while others use satellite imagery. The disease detection aspect of the surveillance module uses computer vision and machine learning to detect plant diseases based on leaf images. However, leaf-based approaches rely on the use of imaging devices in low-resource approaches that use smartphones. This may be limited in areas with no or low smartphone usage.

In observing pathogen and host interactions, various plant diseases and pests would cause a variety of symptoms and plant damage, providing a physical basis for their remote monitoring. It should be noted that not all plant diseases are suitable for remote sensing, as some of them do not have identifiable characteristics. On the other hand, some soil-borne and root diseases that have systemic effects on plant physiology can be detected. Therefore, an essential requirement for the detection and monitoring of plant diseases and pests by remote sensing is the presence of a specific response that can be detected by a specific sensor or sensor system.


Figure 2.6.Disease detection of fungal plant diseases based on hyperspectral images. A: Cercospora leaf spot on sugar beet. B: Fusarium head blight on wheat (Mahlein, 2016)