메뉴 건너뛰기

ENG

추천
검색
질문

논문 기본 정보

자료유형
학술저널
저자정보
(Incheon National University) (Incheon National University) (Incheon National University) (Incheon National University) (Incheon National University) (Incheon National University)
저널정보
한국양봉학회 Journal of Apiculture Journal of Apiculture Vol.36 No.4
발행연도
수록면
273 - 280 (8page)
DOI
10.17519/apiculture.2021.11.36.4.273

이용수

표지
📌
연구주제
📖
연구배경
🔬
연구방법
이 논문의 연구방법이 궁금하신가요?
🏆
연구결과
이 논문의 연구결과가 궁금하신가요?
AI에게 요청하기
추천
검색
질문

초록· 키워드

A new honeybee in-out monitoring system is proposed using real-time deep-learning based image recognition and tracking. The specific design of beehive gate is turned out to be an important factor for accurate bee movement monitoring. We check a series of beehive gate designs for the monitoring system. A novel gate design employing heart valve structure is proposed for ensuring one-way traffic for the bees as well as one-at-a-time gate passing, resulting in an improved bee detection accuracy. As for the deep-learning based image recognition framework, YOLOv4 is used in the proposed system for a better honeybee-detection accuracy as well as a faster detection in comparison to YOLOv3 which was employed for our previous study. In addition, DeepSORT algorithm is employed for a reliable tracking of the detected honeybees. In our experiments the proposed honeybee monitoring system exhibited 99.5% detection accuracy, while our previous system resulted in 97.5% in the same settings.
#Honeybee #Monitoring #Beehive gate #YOLO #DeepSORT
상세정보 수정요청해당 페이지 내 제목·저자·목차·페이지
정보가 잘못된 경우 알려주세요!

목차

  1. Abstract
  2. INTRODUCTION
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. LITERATURE CITED

참고문헌

참고문헌 신청

최근 본 자료

전체보기
260319_딥리서치
260319_AI리더
260304_새학기에이전트
260122_AIagent
260304_아그논문롤
260127_아그논문
260127_AI뷰어
UCI(KEPA) : I410-ECN-0101-2022-527-000067736