색상에 따른 후숙과일 숙성도 분류 방법
Color-Based Ripening Classification Method
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
본 논문에서는 후숙 과일의 후숙도 예측을 목표로 한다. 이는 최적 수확 시기를 결정하는 중요한 요인을 예측하는 문제에 대해 다룬다. 고령화로 인해 농가인구 감소로 노동력이 줄어들고 있는 농부들에게 도움을 주고 농산물의 품질 및 경제적 가치를 향상하게 시키는데 기여한다. 과일 소비의 변화하는 경향, 해외 과일에 대한 선호 증가, 그리고 기후 변화가 한국 지역에서 열대 과일 재배를 가능하게 하는 영향을 배경으로, 우리 연구는 새로운 해결책을 제시한다. 과일 인식을 위해 YOLOv8을, 객체 추출을 위해 OpenCV를, 색상 및 분포 분석을 위해 K-Means Clustering을 사용하여 후숙 정도를 데이터화하여 Soft Voting을 이용해 과일의 후숙도를 분류한다. 또한, 본 연구에서 사용된 과일 말고 비슷한 방식의 과일 또는 농산물이 있다면 적용이 가능할 것으로 예상한다.
Abstract
In this paper, the goal is to predict the degree of ripeness of late-ripening fruits. It addresses the problem of predicting the important factors that determine the optimal harvest time. It provides help to farmers whose labor force is decreasing due to the aging population and contributes to improving the quality and economic value of agricultural products. Against the backdrop of changing trends in fruit consumption, increasing preference for foreign fruits, and the impact of climate change on enabling tropical fruit cultivation in the Korean region, our study suggests a new solution. YOLOv8 for fruit recognition, OpenCV for object extraction, and K-Means Clustering for color and distribution analysis are used to collect data on the degree of ripeness, and then classify the degree of ripeness of the fruit using soft voting. In addition, it is expected that it can be applied if there are fruits or agricultural products with similar methods other than the fruits used in this study.
참고문헌 (10)
- Chapter 14 Fruit supply and demand trends and prospects, "Won-tae Kim", Korea Rural Economic Institute, 533~618 (2023)
- Let’s solve crises and opportunities in agriculture and rural areas through innovation and inclusion, "Lee Soo-haeng", Gyeonggi Research Institute, 1~24 (2023)
- Gyeonguk Park, "Fruit Classifiers Using Deep Learning-Based Object Detection Technology", Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp.1197~1198 (2023)
- Myung-Jin Kim, "Development of Automatic Classification Deep Learning System for Shine-muscat and Greengrape", Journal of Digital Contents Society, 22(10), pp.1637~1644 (2021)[DOI]
- Yeong-Min Kim, "Deep Learningbased Mango Classification and Prediction System of Fruit Ripening using YOLO", Proceedings of the Korean Society of Computer Information Conference, pp.187~188 (2021)
- "Object Detection with YOLOv8"[URL]
- Si-Hyeon Koo, "Cluster Tracking Algorithm using K-means", Proceedings of the Domestic Conference of the Society of Control and Robotics Systems, pp.76~77 (2023)
- Sung-Hwa Hong, "Video Detection using a Multi Camera Module with OpenCV", Proceedings of the Korea Information and Communication Society General Conference, pp.406~408 (2023), Hankyung National University Anseong Campus
- Pil-Do Yoon, "Malicious traffic detection using ensemble learning based on UNSW-NB15dataset", Proceedings of Symposium of the Korean Institute of communications and Information Sciences, pp.952~953 (2021)
- SeokYoon Hong, "How to choose clustering algorithm for image color extraction", Korean Society of Information Scientists and Engineers Academic Presentation Papers, pp.1948~1950 (2021)