The Rural Development Administration (Commissioner Lee Seung-don) announced that it has built an ‘AI-based rice growth monitoring system’ that combines video imagery with artificial intelligence (AI) for rice growth surveys, and that it is now verifying the system’s field applicability and effectiveness. The aim is to shift the conventional method, in which people entered paddies to inspect crops in person, to a data-driven approach that reduces survey labor by more than 60 percent.
 [코리안투데이] 사람 대신 영상과 인공지능 AI 으로 벼 생육 살핀다 © 장형임 기자 |
Periodically checking the condition of rice growth serves not only farmers’ cultivation management but also provides foundational data for agricultural research and policymaking. Yet as abnormal weather such as heat waves and torrential rain becomes more frequent, and as agricultural disasters including pests and diseases grow more common, the need to monitor growth conditions more often has been increasing.
At present, rice growth surveys are carried out by surveyors who enter paddies in person to examine growth conditions and yield-related traits. Conducted twice a month from transplanting through harvest, for a total of nine times, and requiring at least two people each session, the work demands considerable time and labor. In particular, when summer heat waves are as severe as this year, field surveys themselves became extremely difficult.
 [코리안투데이] 사람 대신 영상과 인공지능 AI 으로 벼 생육 살핀다 © 장형임 기자 |
In response, the National Institute of Crop Science developed a system that collects and analyzes imagery of the same location on a daily basis, making it possible to track growth changes every day. Because observation is carried out remotely, without personnel on site, and in a non-destructive manner, the number of field visits and the survey time can be reduced, cutting rice growth survey labor by roughly 60 percent or more. In effect, it becomes possible to assess the condition of rice every day without setting foot in the paddy during a heat wave.
The system consists of ‘imagery collection equipment,’ an ‘AI model for image-based growth surveys,’ and a ‘rice growth monitoring platform.’ Construction began in 2023 and the system remains in operation to date. To gather data and develop analytical models, the National Institute of Crop Science has installed and operated fixed imagery collection equipment at 19 provincial agricultural research and extension services nationwide. In 2025, it developed mobile imagery collection equipment that is easy to install and relocate, expanding coverage to 86 sites including city and county agricultural technology centers. The mobile equipment has been registered as a design patent (30-1350082) for a ‘monitoring camera mount.’
 [코리안투데이] 사람 대신 영상과 인공지능 AI 으로 벼 생육 살핀다 © 장형임 기자 |
Based on the collected imagery, an AI model that analyzes rice growth and yield-related traits has also been developed. Its current accuracy stands at 97 percent for the number of plants per unit area (planting density), 98 percent for the number of tillers, 99 percent for the heading date, and 88 percent for rice yield based on panicle area. The Rural Development Administration plans to raise the accuracy of growth trait analysis to 99 percent and the yield analysis model to 95 percent. In connection with rice yield prediction, it has filed a patent for a ‘method for predicting rice yield and a device for predicting rice yield using the same’ (10-2025-0205006).
The collected imagery and the AI analysis results are managed in an integrated manner on the ‘rice growth monitoring platform.’ It currently operates on the basis of data from fixed equipment at 19 provincial agricultural research and extension services, and functions will be improved so that data from mobile equipment can be linked as well.
 [코리안투데이] 사람 대신 영상과 인공지능 AI 으로 벼 생육 살핀다 © 장형임 기자 |
Research on applying video and AI technology to disaster response is also under way. A program has been developed to analyze the areas affected by rice lodging and by brown planthopper damage using footage captured by unmanned aerial vehicles (drones), laying a foundation for quickly grasping the scale of disasters over wide areas. The program links with map information from the Ministry of Land, Infrastructure and Transport’s V-WORLD service to extract the target parcels for analysis, calculates the damaged area and damage ratio for each parcel, and can save the results as video files and CSV files.
 [코리안투데이] 사람 대신 영상과 인공지능 AI 으로 벼 생육 살핀다 © 장형임 기자 |
The Rural Development Administration intends to expand the installation points of imagery collection equipment in stages, applying the system to 690 rice growth survey sites nationwide by 2030. Furthermore, it plans to extend the video and AI technology developed for rice to other crops, transforming the overall crop growth survey system into one based on video and AI.
Kim Ki-young, director-general of the Basic Food Crops Department at the Rural Development Administration, said, ‘The AI-based rice growth monitoring system is a key achievement of the agricultural AI transformation (AX), converting the existing labor-centered survey method to a data-based one.’ He added, ‘Going forward, we will broaden the scope of video and AI technology beyond rice growth and yield surveys to include disaster diagnosis, and we will do our utmost to ensure that our research results can be put to use on actual farms.’
지역 홍보대사 신청 기자양성과정 신청 코리안투데이 지부장 신청 코리안투데이 인천강화코리안브랜드대상 신청 최고경영인대상 신청 소비자만족도대상 신청
<저작권자 ⓒ 코리안투데이(The Korean Today) 무단전재 및 재배포 금지>
