Application of Remote Sensing Technology in Pest and Disease Observation Rongling Ye (*), Miho Koike, Osamu Watanabe, Ryusei Seki
Graduate School of Agriculture Science, Shinshu University, Japan
*ye[at]shinshu-u.ac.jp
Abstract
Pest and disease infestations are major factors affecting crop yield and quality. With the rising awareness of food safety among consumers and the advancement of sustainable agriculture, it is essential to re-evaluate traditional methods of pest and disease control, such as the use of chemical pesticides and varietal improvement by transgenesis. Early detection of pests and diseases allows for prompt intervention, reducing agrochemical use, protecting the environment, and meeting modern health requirements.
Remote sensing technologies have been widely applied in agriculture. Remote sensing technologies allows for the rapid acquisition of large amounts of information, labor saving, and providing comprehensive farm monitoring with greater precision than traditional methods. Additionally, different sensors can capture data that is invisible to the human eye, enhancing the potential for early detection of anomalies.
This presentation presents case studies on pest and disease monitoring which aimed to establish methods for using drones to detect pests and diseases. When plants are subjected to stresses such as pests, their stomata close, leading to a decrease in transpiration and subsequently causing an increase in leaf temperature, which can be recognized by thermal cameras (Nakamura et al., 2021). This experiment used drones equipped with thermal imaging cameras (Mavic 2 Enterprise Advanced, DJI) to capture images of soybean fields in Japan and examined the effectiveness of pest and disease detection. The results demonstrated that thermal imaging cameras can identify pest- and disease-infested plants at an early stage. Different objects reflect sunlight in varying areas and intensities, which allows for the differentiation of plants and the background in reflection maps. Additionally, plants in different health situations also show different reflections, which can be used to assess their growth conditions. To provide a more accurate evaluation, vegetation indices derived from mathematical calculations using two or more reflections have been developed. In this study, the Green Normalized Difference Vegetation Index (GNDVI), commonly used for assessing photosynthesis, was employed to evaluate the pest infestation of cabbage in Japan. Data was taken by a drone with multispectral camera (Mavic 3 Multispectral, DJI). The results indicate that GNDVI effectively distinguishes between healthy plants and those infested by the cotton leaf aphid.The advantages of thermal imaging lie in its ability to rapidly and promptly identify abnormal plants. However, it only provides instantaneous temperature, which is significantly influenced by environmental factors such as wind or high ground temperatures during midday. In contrast, multispectral cameras offer more stable information, although they require relatively longer processing times. Currently, both methods are only capable of determining the presence or absence of pest and disease issues, future research is needed to refine techniques for accurately assessing the severity of damage.
Keywords: crop, pest, disease, drone
Topic: Smart technology for sustainable agro-industry
