양봉을 위한 합성 이미지 훈련 기반 비전 인식 및 로봇 자동화 시스템의 타당성 연구
A Feasibility Study of Vision Recognition based on Synthetic Image Training and Robotic Automation System for Beekeeping
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Beekeeping is an essential industry that supports honey production and insect-mediated pollination. A critical aspect of beekeeping is the regular inspection of beehives, particularly to confirm the presence of the queen bee, whose role is vital to the health and productivity of the colony. However, these inspections must be frequent and repetitive, placing a significant labor burden on beekeepers. Despite technological advancements, this reliance on manual processes often limits productivity and operational stability. To address these challenges, this study proposes an automated crane system with a computer vision-based approach for queen bee detection. Additionally, by utilizing synthetic image-based deep learning, the study aims to enhance the accuracy and efficiency of queen bee identification. The research presents the development of a cost-effective automated inspection system featuring a robotic crane for hive monitoring. This system enables regular inspection of multiple beehives without direct human intervention, reducing labor demands and improving efficiency. By combining advanced vision recognition technology with automated hardware solutions, this approach has the potential to revolutionize hive management, making beekeeping operations more sustainable and productive.
Abstract
Beekeeping is an essential industry that supports honey production and insect-mediated pollination. A critical aspect of beekeeping is the regular inspection of beehives, particularly to confirm the presence of the queen bee, whose role is vital to the health and productivity of the colony. However, these inspections must be frequent and repetitive, placing a significant labor burden on beekeepers. Despite technological advancements, this reliance on manual processes often limits productivity and operational stability. To address these challenges, this study proposes an automated crane system with a computer vision-based approach for queen bee detection. Additionally, by utilizing synthetic image-based deep learning, the study aims to enhance the accuracy and efficiency of queen bee identification. The research presents the development of a cost-effective automated inspection system featuring a robotic crane for hive monitoring. This system enables regular inspection of multiple beehives without direct human intervention, reducing labor demands and improving efficiency. By combining advanced vision recognition technology with automated hardware solutions, this approach has the potential to revolutionize hive management, making beekeeping operations more sustainable and productive.