本郷 千春

An efficient assessment of rice varieties in tropical regions is crucial for selecting cultivars suited to unique environmental conditions. This study explores machine learning algorithms that leverage multispectral sensor data from UAVs to evaluate rice varieties. It focuses on three paddy rice types at different ages (six, nine, and twelve weeks after planting), analyzing data from four spectral bands and vegetation indices using various algorithms for classification. The results show that the neural network (NN) algorithm is superior, achieving an area under the curve value of 0.804. The twelfth week post-planting yielded the most accurate results, with green reflectance the dominant predictor, surpassing the traditional vegetation indices. This study demonstrates the rapid and effective classification of rice varieties using UAV-based multispectral sensors and NN algorithms to enhance agricultural practices and global food security.

Red crown rot (RCR) is a soil-borne disease that damages soybean growth and decreases yield. Infected plants show earlier defoliation and pencil-like roots, sometimes resulting in mortality. This disease became common relatively recently, and information about its field-scale appearance is insufficient. Insufficient data is a major constraint when planning countermeasures. In this study, unmanned aerial vehicle (UAV)-acquired images were used to visualize the spatial and time series variation in the area damaged by RCR in the same farmer fields in 2018 and 2020. Field investigation showed that RCR severely damaged soybean production. The reductions of yield were estimated at 17.5% and 12.7% in 2018 and 2020, respectively. The visualized damage clarified the difference in the increasing rate and patterns of RCR between the 2 years. In 2018, the damaged area expanded along the planting row to the whole field, but in 2020, the expansion along the planting row was not great, and half of the fields remained sparsely damage. This difference implies that various factors are associated with damage occurrence and pathogen distribution. The method applied in this study is effective in visualizing RCR damage, but further improvement is required in the evaluation of intermediate damage and the generalization of the evaluation procedure.

Drought is increasingly threatening smallholder farmers in Southeast Asia. The crop insurance system is one of the promising countermeasures that was implemented in Indonesia in 2015. Because the damage assessment in the present system is conducted through direct investigations based on appearance, it is not objective and needs a long time to cover large areas. In this study, we investigated a rapid assessment method for paddy fields using a vegetation index (VI) taken by an unmanned aerial vehicle (UAV) with a multispectral camera in 2019 and 2021. Then, two ways of assessment for drought damage were tested: linear regression (LR) based on a visually assessed drought level (DL), and k-means clustering without an assessed DL. As a result, EVI2 could represent the damage level, showing the tendency of the decrease in the value along with the increasing DL. The estimated DL by both methods mostly coincided with the assessed DL, but the concordance rates varied depending on the locations and the number of assessed fields. Differences in the growth stage and rice cultivars also affected the results. This study revealed the feasibility of the UAV-based rapid and objective assessment method. Further data collection and analysis would be required for implementation in the future.