http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
머신러닝 기반 돼지의 미래 활동 지수에 영향을 미치는 요인과 예측에 관한 연구
조예성(Yesung Cho),Nabilah Muallifah,Fachrina Dwei Puspitasari,김유승(Yusung Kim),Mhd Anas Al Saidi,이문용(Munyong Yi) 한국정보기술학회 2024 한국정보기술학회논문지 Vol.22 No.4
The term activity index refers to a measure of the activity level of pig herds that has been applied as an assessment tool in pig farming practice. The index has become an indicator of pig’s health, physical growth, behavior control, and quality of living conditions. In large-scale pig farms, however, manual observation of pig activity is impractical due to huge resource and potentially inconsistent recording results. Thus, our study aims to explore the automatic monitoring of pig activity level by predicting the near feature activity index. We calculated the activity indexes by utilizing background subtraction techniques from video frames. Further, we also harnessed other farming attributes including biological characteristics and the occurrence of exogenous events. Our analysis uncovers the relationship between activity index and these farming attributes, and pinpoints several attributes that influence the change in activity index.
이미지 기반 축산물 불량 탐지에서의 희소 클래스 처리 전략
이범호,조예성,이문용,Lee, Bumho,Cho, Yesung,Yi, Mun Yong 한국정보통신학회 2022 한국정보통신학회논문지 Vol.26 No.11
The industrial 4.0 era has been opened with the development of artificial intelligence technology, and the realization of smart farms incorporating ICT technology is receiving great attention in the livestock industry. Among them, the quality management technology of livestock products and livestock operations incorporating computer vision-based artificial intelligence technology represent key technologies. However, the insufficient number of livestock image data for artificial intelligence model training and the severely unbalanced ratio of labels for recognizing a specific defective state are major obstacles to the related research and technology development. To overcome these problems, in this study, combining oversampling and adversarial case generation techniques is proposed as a method necessary to effectively utilizing small data labels for successful defect detection. In addition, experiments comparing performance and time cost of the applicable techniques were conducted. Through experiments, we confirm the validity of the proposed methods and draw utilization strategies from the study results.