Rice (Oryza sativa L.) is one of the most important cereals, which provides 20% of the world's food energy. However, its productivity is poorly assessed especially in the global South. Here, we provide a first study to perform a deep-learning-based approach for instantaneously estimating rice yield using red-green-blue images. During ripening stage and at harvest, over 22,000 digital images were captured vertically downward over the rice canopy from a distance of 0.8 to 0.9 m at 4,820 harvesting plots having the yield of 0.1 to 16.1 t·ha^-1 across 6 countries in Africa and Japan. A convolutional neural network applied to these data at harvest predicted 68% variation in yield with a relative root mean square error of 0.22. The developed model successfully detected genotypic difference and impact of agronomic interventions on yield in the independent dataset. The model also demonstrated robustness against the images acquired at different shooting angles up to 30° from right angle, diverse light environments, and shooting date during late ripening stage. Even when the resolution of images was reduced (from 0.2 to 3.2 cm·pixel^-1 of ground sampling distance), the model could predict 57% variation in yield, implying that this approach can be scaled by the use of unmanned aerial vehicles. Our work offers low-cost, hands-on, and rapid approach for high-throughput phenotyping and can lead to impact assessment of productivity-enhancing interventions, detection of fields where these are needed to sustainably increase crop production, and yield forecast at several weeks before harvesting.

Rice is not merely a staple food but an important source of income in Cambodia. Rapid socioeconomic development in the country affects farmers’ management practices, and rice production has increased almost three-fold over two decades. However, detailed information about the recent changes in rice production is quite limited and mainly obtained from interviews and statistical data. Here, we analyzed MODIS LAI data (MCD152H) from 2003 to 2019 to quantify rice production changes in Pursat Province, one of the great rice-producing areas in Cambodia. Although the LAI showed large variations, the data clearly indicate that a major shift occurred in approximately 2010 after applying smoothing methods (i.e., hierarchical clustering and the moving average). This finding is consistent with the results of the interviews with the farmers, which indicate that earlier-maturing cultivars had been adopted. Geographical variations in the LAI pattern were illustrated at points analyzed along a transverse line from the mountainside to the lakeside. Furthermore, areas of dry season cropping were detected by the difference in monthly averaged MODIS LAI data between January and April, which was defined as the dry season rice index (DSRI) in this study. Consequently, three different types of dry season cropping areas were recognized by nonhierarchical clustering of the annual LAI transition. One of the cropping types involved an irrigation-water-receiving area supported by canal construction. The analysis of the peak LAI in the wet and dry seasons suggested that the increase in rice production was different among cropping types and that the stagnation of the improvements and the limitation of water resources are anticipated. This study provides valuable information about differences and changes in rice cropping to construct sustainable and further-improved rice production strategies.

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.

Crop productivity is poorly assessed globally. Here, we provide a deep learning-based approach for estimating rice yield using RGB images. During ripening stage and at harvest, over 22,000 digital images were captured vertically downwards over the rice canopy from a distance of 0.8 to 0.9 m, and rice yields were obtained in the corresponding area ranging from 0.1 and 16.1 t ha⁻¹. A convolutional neural network (CNN) applied to these data at harvest predicted 70% variation in rice yield with a relative root mean square error (rRMSE) of 0.22. Images obtained during the ripening stage can also be used to forecast the final rice yield. Our work suggests that this low-cost, hands-on, and rapid approach can provide a breakthrough solution to assess the impact of productivity-enhancing interventions and identify fields where these are needed to sustainably increase crop production.

Accurate estimation of the leaf area index (LAI) is essential for evaluating crop growth, biomass accumulation, and yield potential in rice (Oryza sativa L.). Vegetation indices (VIs) from remote sensing, estimate LAI nondestructively, but canopy architecture differences among cultivars can limit accuracy. This study aimed to elucidate how canopy architecture influences the relationship between the LAI and VIs in rice. Field experiments were conducted over three years (2022–2024) using two japonica cultivars: ‘Koshihikari’ with normal panicles and leaves and ‘Liaojing 5’ with erect panicles and leaves. Rice plants were grown under three fertility conditions, and canopy reflectance was measured using a proximal spectroradiometer and UAV-mounted multispectral cameras to estimate LAI. Nine VIs were calculated and compared with measured LAI values. UAV-based indices generally showed stronger correlations with the LAI than those obtained from proximal measurements. Most VIs exhibited saturation at high LAI values; however, Green Atmospherically Resistant Index (GARI) and MERIS Terrestrial Chlorophyll Index (MTCI) maintained their sensitivities. Importantly, the LAI–VI relationship differed significantly between cultivars at higher LAI values (≥2), reflecting differences in canopy architecture. When cultivar-specific models were applied, the accuracy of LAI estimation improved markedly. These results demonstrate that the canopy architecture must be considered when developing models for LAI estimation in rice. Because cultivar-specific LAI–VI relationships capture differences in canopy architecture, these models may also be used to evaluate cultivar-specific traits, thereby offering additional value for breeding programs.