인문학
사회과학
자연과학
공학
의약학
농수해양학
예술체육학
복합학
지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 저자정보
초록·키워드
The increasing accumulation of plastic waste over recent decades has become a pressing environmental concern. Although public awareness of recycling has grown, studies show that up to 75% of recyclable municipal solid waste is still discarded in landfills. To address this issue, we propose a deep learningbased pipeline for plastic waste sorting that combines the YOLOv8 object detection framework with contrast-enhanced image preprocessing using CLAHE (Contrast-Limited Adaptive Histogram Equalization). CLAHE is applied independently to each RGB channel of the input image and blended with the original to enhance visual features while preserving color fidelity. The proposed system is designed to detect and classify multiple plastic types (PE, PET, PP, PS) from real-world images. Experimental results demonstrate that the model achieves a mean Average Precision (mAP@0.5) of 0.878 and an F1-score of 0.83 at the optimal confidence threshold, indicating robust detection performance. Notably, the balanced blending strategy outperformed both unprocessed and fully enhanced inputs. These results suggest that the integration of adaptive contrast enhancement can significantly improve detection robustness in practical waste sorting systems. Our approach offers a scalable and accurate solution for real-time deployment in automated recycling facilities, contributing to improved plastic recovery and sustainable waste management.
본문·목차
인공지능 문자 인식 모델을 통해 추출된 텍스트로, 일부 오타나 오류가 포함될 수 있으나 지속적으로 개선 중입니다.
오류를 발견하셨다면 해당 부분을 드래그한 후 ' 를 통해 신고해주세요.
오류를 발견하셨다면 해당 부분을 드래그한 후 ' 를 통해 신고해주세요.
최근 본 자료 전체보기
UCI(KEPA) : I410-151-25-02-093929071