蕎麦の実自動選別装置の開発 – YOLOv7 と空気圧制御を用いたリアルタイム色選別手法の検討 –

Read full text Summarise with VIRAT
Authors: 大貴 鈴木, 知宏 小高, 丈介 黒岩, Daiki Suzuki, Tomohiro Odaka, Jousuke Kuroiwa
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study develops an automatic buckwheat grain sorting system using real-time image recognition and pneumatic control. A YOLOv7 model was optimized with ONNX and executed on a Raspberry Pi 5 to classify defective grains at high speed. The system removes detected grains through an air-jet mechanism synchronized with inference results. Experiments with small batches achieved nearly 100 % sorting accuracy, demonstrating the feasibility of low-cost, real-time automation for reducing manual labor in buckwheat processing.

Fields of Study & Subjects

Smart Agriculture and AI Industrial Vision Systems and Defect Detection Agricultural Engineering and Mechanization

Publication details

Journal / source
Institutional Repositories DataBase (IRDB)
Volume
74
Publication date
1 February 2026
Language
JA