Early and accurate knowledge of crop acreage and yields is important for decision-making in food security policy and the grain-trading business. Crop acreage estimations achieved by applying machine learning to satellite data require the appropriate training data. Herein, we developed a new training data set for rice cultivation in California (USA) by combining the historical Cropland Data Layer provided by the U.S. Department of Agriculture (USDA) with the latest Sentinel-2 satellite data, which can be applied even when cropping patterns differ significantly from the past cropping patterns due to extreme events such as large-scale droughts. We then applied these training data to random forests with the Advanced Land Observing Satellite-2 (ALOS-2) data for classifying rice cultivation over a wide area with high accuracy, without the effects of clouds. An assessment of the proposed method's accuracy was then conducted. The results demonstrated that our proposed method can estimate paddy rice acreage with ＜1 % error compared to the USDA statistics even in 2021, when a large-scale drought occurred. We also evaluated the relationship between the lead time to harvest and the accuracy of the proposed method's area estimation; the results confirmed that the method estimated the area with an approx. 1 % estimation error even in late July, which was >1 month before the harvest in 2021. Our proposed method can thus be used for early and accurate estimations of paddy rice acreage even in years of drought extremes, and the method can be expected to be applied to food security policy, the grain-trading business, and more.

A seed potato can produce approximately 10 potato tubers, and diseased seed potatoes can also multiply 10 times in each propagation step and through the seed-potato production cycle. To promote stable potato production, quality seed potatoes that are healthy and disease-free are necessary. However, experienced laborers are required in the fields to visually inspect and rogue abnormal plants during seed potato production. Our previous study developed an automated abnormal potato plant detection system using deep learning models and portable video cameras. The developed system detects abnormal plants or leaves considering the growth stage. Since the proposed system met the required accuracy for the roguing task, we are researching its application for practical use. Portable cameras can support the diagnosis of abnormal plants, but they cannot reduce the damage to plants caused by farmers entering the fields for the roguing task. Therefore, a practical application method for detecting abnormal plants without damaging plants is needed. In this study, we examined the effectiveness of the method developed for portable cameras by applying it to drone imagery. In terms of abnormal and healthy potato plant classification, the accuracy was 86%, and the average precision (AP) for detection was 80.7%. Furthermore, we investigated the spatial resolution required for detecting abnormal plants in the early-growth and middle-growth stages. We found that the spatial resolution required for extracting abnormal plants (7.5 mm in this study) was sufficient; however, when judging using not only the size of the plant but also the state of the leaves, the highest possible resolution image should be used for the early-growth stage, and 2.5-mm resolution is required for the middle stage of growth. We demonstrated that it is possible to put into practical use automatic detection of abnormal potato plants without leaf symptoms using drones by applying the proposed methods.