The Rural Development Administration is accelerating on-site adoption by advancing a "phenotype image analysis technology" that quantifies the shape, color, and growth of seeds and crops through imaging. By replacing the process that people used to observe directly with digital equipment, the period for variety development and the manpower required are expected to decrease.
According to the Rural Development Administration on the 1st, this technology can automatically analyze 11 characteristics per seed in one second. With the conventional method, a person had to spend about five minutes on the task. Converted to a baseline of 40,000 seeds, the conventional approach required four people to analyze them over 40 days, but with the technology, one person can finish in a day.
Applied to 62 commercial seed types, the analysis accuracy reached 97%. The Rural Development Administration assessed it as the world's best level, comparable to advanced countries such as the United States and Europe. The Rural Development Administration is spreading the technology by transfer to the National Seed Management Office's variety registration program, the Korea Atomic Energy Research Institute (KAERI) for distinguishing general seeds from mutant seeds, and private bioinformatics companies for trait auto-analysis services. So far, 36 cases have been carried out across institutions, universities, and industry.
Its scope of use is expanding beyond seed analysis to farms and processing sites overall. It enables selecting high-sugar strawberries, detecting blemishes on apples, counting enoki mushroom caps, checking the number of corn kernels, and predicting harvest times for flowers and fruits. In the baking process, it automatically sorts out defective products at the dough and baking stages, and in the forestry sector, it analyzes the germination rate of bracken fern spores to produce healthy spores.
The Rural Development Administration is also applying this technology to a rice variety development project in 15 African countries. It said it is contributing to international agricultural cooperation by using image analysis technology to select varieties suitable for local climates in Ghana, Senegal, and elsewhere.
In 2017, the Rural Development Administration completed the country's largest "crop phenotype research building" and has continued related research. In 2022, it was designated by the Ministry of Trade, Industry and Energy as Korea's first "crop phenotype reference standard data center," establishing standard data for nine rice varieties. It is currently building a supercomputer-based digital breeding platform for 65 major crops, including wheat, soybeans, corn, and chili peppers. The goal is to develop new varieties resilient to abnormal weather.
Kim Nam-jeong, head of the Agricultural Genetic Resources Bureau at the Rural Development Administration, said, "In line with the new government's national agenda to advance smart agriculture, phenotype image analysis technology will accelerate agriculture's digital innovation," adding, "It can solve challenges in the field, spread into various areas, and become a driving force to pioneer future food sources."