画像認識技術を活用したきゅうり農園の付加価値生産

抄録

Against the backdrop of the challenges facing Japanese agriculture, such as the declining birthrate, aging population, and lack of successors, the objective of this research is to introduce digital transformation (DX) using a small- to medium-scale cucumber farm as a case study to improve the efficiency and sophistication of sorting operations. In particular, the aim of the project was to build a sorting system that can be handled even by newcomers to the farm by transforming manual cucumber sorting work, which relies on the experience and tacit knowledge of skilled workers, into formal knowledge using AI's image recognition technology. By using image recognition technology, we verified the possibility of automating the sorting work that has traditionally relied on the manual labor of skilled workers, and of improving work efficiency and sorting accuracy even for new farmers.

As a result of a demonstration experiment using the prototype sorting machine, it was possible to reduce sorting time by up to 57.6% and improve sorting accuracy by up to 80.4% compared to conventional manual work. The results of this research provide a new approach that contributes to improving the management efficiency and reducing the workload of small and medium-sized farmers.