RISS 학술연구정보서비스

검색

인기 검색어

    다국어 입력

    http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

    변환된 중국어를 복사하여 사용하시면 됩니다.

    예시)
    • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
    • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
    닫기

    YOLO-DenseTrack 밀도 맵 융합 기반 밀집 환경 육계 추적 시스템 = YOLO-densetrack: a density map fusion-based broiler tracking system for high-density environments

    한글로보기

    https://www.riss.kr/link?id=T17408394

    • 0

      상세조회
    • 0

      다운로드
    서지정보 열기
    • 내보내기
    • 내책장담기
    • 공유하기
    • 오류접수
    인용문이 복사되었습니다.

    부가정보

    다국어 초록 (Multilingual Abstract) kakao i 다국어 번역

    Behavior analysis of broilers using computer vision technology is emerging as a core technology for managing health, welfare, and productivity in smart farms. However, in actual broiler house environments, existing object detection models face limitations in securing stable tracking data due to extreme occlusion caused by high rearing density and facilities, as well as appearance ambiguity among individuals.
    To address these issues, this study proposes YOLO-DenseTrack, a novel fusion framework combining the YOLO (You Only Look Once) object detection model and a density map estimation model. The proposed system mutually complements YOLO's precise localization capability with the density map's ability to estimate the global count and identify dense regions.
    Specifically, this study designed a bidirectional correction process using YOLO detection results and density map information. Areas identified as hotspots with high object probability in the density map but missed by YOLO are classified as "candidate areas," and re-detection with a lower threshold is performed to correct false negatives. Simultaneously, false positives from YOLO are removed based on the density values of the density map, thereby enhancing detection reliability.
    Experimental results demonstrated that the proposed fusion model achieved significant improvements in Recall and Precision compared to the baseline YOLO model, proving superior detection performance especially in high-difficulty scenarios where objects are heavily overlapped or occluded. Furthermore, applying the enhanced detection results to a multi-object tracking algorithm (ByteTrack) improved tracking performance metrics such as MOTA and IDF1 Score. This study is expected to contribute to improving the accuracy of individual broiler behavior analysis and welfare monitoring systems by securing robustness in detection and tracking within high-density environments.
    번역하기

    Behavior analysis of broilers using computer vision technology is emerging as a core technology for managing health, welfare, and productivity in smart farms. However, in actual broiler house environments, existing object detection models face limitat...

    Behavior analysis of broilers using computer vision technology is emerging as a core technology for managing health, welfare, and productivity in smart farms. However, in actual broiler house environments, existing object detection models face limitations in securing stable tracking data due to extreme occlusion caused by high rearing density and facilities, as well as appearance ambiguity among individuals.
    To address these issues, this study proposes YOLO-DenseTrack, a novel fusion framework combining the YOLO (You Only Look Once) object detection model and a density map estimation model. The proposed system mutually complements YOLO's precise localization capability with the density map's ability to estimate the global count and identify dense regions.
    Specifically, this study designed a bidirectional correction process using YOLO detection results and density map information. Areas identified as hotspots with high object probability in the density map but missed by YOLO are classified as "candidate areas," and re-detection with a lower threshold is performed to correct false negatives. Simultaneously, false positives from YOLO are removed based on the density values of the density map, thereby enhancing detection reliability.
    Experimental results demonstrated that the proposed fusion model achieved significant improvements in Recall and Precision compared to the baseline YOLO model, proving superior detection performance especially in high-difficulty scenarios where objects are heavily overlapped or occluded. Furthermore, applying the enhanced detection results to a multi-object tracking algorithm (ByteTrack) improved tracking performance metrics such as MOTA and IDF1 Score. This study is expected to contribute to improving the accuracy of individual broiler behavior analysis and welfare monitoring systems by securing robustness in detection and tracking within high-density environments.

    더보기

    목차 (Table of Contents)

    • 목차 ⅰ
    • 그림 목차 ⅲ
    • 표 목차 ⅴ
    • Abstract ⅵ
    • 목차 ⅰ
    • 그림 목차 ⅲ
    • 표 목차 ⅴ
    • Abstract ⅵ
    • 제 1 장 서 론 01
    • 제 2 장 관련 연구 04
    • 2.1 딥러닝 기반 객체 탐지 기술과 스마트 양계 적용의 한계 04
    • 2.1.1 탐지 모델의 진화와 농업 분야 적용 05
    • 2.1.2 밀집 환경에서의 탐지 실패 원인: NMS의 딜레마 06
    • 2.2 밀도맵 추정을 이용한 육계 계수(Counting)에 관한 연구 08
    • 2.2.1 밀도맵 생성 방법 08
    • 2.2.2 밀도맵 학습 전략 및 손실 함수 09
    • 2.2.3 밀도맵 모델 성능 비교 및 선정 기준 11
    • 2.2.4 UNet 기반 모델 (UNet-based Models) 11
    • 2.2.5 CSRNet (Congested Scene Recognition Network) 12
    • 2.2.6 SANet (Scale Aggregation Network) 13
    • 2.2.7 경량화 모델 (Lightweight Density Models) 14
    • 2.3 객체 탐지와 밀도맵 융합 기술에 관한 연구 15
    • 2.4 컴퓨터 비전 기반 육계 행동 분석에 관한 연구 16
    • 2.4.1 육계 행동의 시각적 특징 및 분류 16
    • 제 3 장 YOLO-DenseTrack 밀도 맵 융합 기반 밀집 환경 육계 추적 시스템 19
    • 3.1 제안 시스템 개요 19
    • 3.2 데이터 수집 20
    • 3.3 알고리즘 구성 과 설계 24
    • 3.3.1 초기 접근: 영상 분할 기반 탐지 개선 및 한계 25
    • 3.3.2 제안하는 융합 프레임워크: YOLO-밀도맵 양방향 보정 25
    • 3.3.3 밀도맵 백본 네트워크 선정 및 분석 26
    • 3.3.4 알고리즘 구성 요소 29
    • 3.3.5 융합 및 보정 프로세스 30
    • 제 4 장 연구 결과 32
    • 4.1 정성적 평가: 탐지 결과 시각화 및 분석 32
    • 4.2 분할 탐지 전략에 따른 성능 분석 34
    • 4.3 융합 모델(YOLO-DenseTrack) 성능 평가 35
    • 제 5 장 결론 및 향후 연구 40
    • 제 6 장 참고 문헌 43
    더보기

    분석정보

    View

    상세정보조회

    0

    Usage

    원문다운로드

    0

    대출신청

    0

    복사신청

    0

    EDDS신청

    0

    동일 주제 내 활용도 TOP

    더보기

    주제

    연도별 연구동향

    연도별 활용동향

    연관논문

    연구자 네트워크맵

    공동연구자 (7)

    유사연구자 (20) 활용도상위20명

    이 자료와 함께 이용한 RISS 자료

    나만을 위한 추천자료

    해외이동버튼