DOI QR코드

DOI QR Code

적외선 카메라를 활용한 급이 유무에 따른 어류 활동성 분석

Analysis of Fish Activity in Relation to Feeding Events Using Infrared Cameras

  • 노태경 (동의대학교 부산IT융합부품연구소) ;
  • 하상현 (동의대학교 인공지능 그랜드 ICT 연구센터) ;
  • 김기환 (동의대학교 인공지능 그랜드 ICT 연구센터) ;
  • 강영진 (동의대학교 인공지능 그랜드 ICT 연구센터) ;
  • 정석찬 (동의대학교 e비즈니스학과, 인공지능그랜드ICT연구센터, 부산IT융합부품연구소)
  • 투고 : 2023.11.21
  • 심사 : 2023.12.08
  • 발행 : 2023.12.31

초록

Purpose The domestic aquaculture industry in South Korea utilizes both formulated feeds and live feeds for the cultivation of fish. While nutrient-rich live feeds, particularly using fry, have been preferred since the past, formulated feeds are gaining attention due to issues related to overfishing and environmental concerns. Formulated feeds are advantageous for storage and supply but require a sustained feeding regimen due to the comparatively slower growth rate compared to live feeds. As the aging population in rural areas leads to a shortage of labor, automated feeding systems are increasingly being adopted in aquaculture facilities. To enhance the efficiency of such systems, it is crucial to quantitatively analyze the behavioral changes in fish based on the presence or absence of feed. Design/methodology/approach In the study, RGB cameras and infrared cameras were used to analyze fish activity according to feeding, and an outline extraction algorithm was applied to analyze the differences resulting from this. Findings Unlike RGB cameras, infrared cameras are more suitable for analyzing underwater fish activity as they convert objects' thermal energy into images. It was observed that Canny, Sobel, and Prewitt filters showed the most distinct identification of fish activity.

키워드

과제정보

이 논문은 2023학년도 동의대학교 교내연구비에 의해 연구되었음(202301170001).

참고문헌

  1. Adegboye, M. A., Aibinu, A. M., Kolo, J. G., Aliyu, I., Folorunso, T. A., and Lee, S. H., "Incorporating intelligence in fish feeding system for dispensing feed based on fish feeding intensity," IEEE, Vol. 8, No. 19659429, May 2020, pp. 91948-91960.
  2. Ballester-Molto, M., Sanchez-Jerez, P., Cerezo-Valverde, J., and Aguado-Gimenez, F., "Particulate waste outflow from fish-farming cages. How much is uneaten feed?," Marine Pollution Bulletin, Vol. 119, June 2017, pp. 23-30. https://doi.org/10.1016/j.marpolbul.2017.03.004
  3. Canny, J., "A Computational Approach to Edge Detection," IEEE, No. 6, November 1986, pp. 679-698.
  4. Li, D., Wang, Z., Wu, S., Miao, Z., Du, L., and Duan, Y., "Automatic recognition methods of fish feeding behavior in aquaculture: A review," Aquaculture, Vol. 528, November 2020, pp. 735508.
  5. Maloy, H., Aamodt, A., and Misimi, E., "A spatio-temporal recurrent network for salmon feeding action recognition from underwater videos in aquaculture," Computers and Electronics in Agriculture, Vol. 167, December 2019, pp. 105087.
  6. Yang, L., Yu, H., Cheng, Y., Mei, S., Duan, Y., Li, D., and Chen, Y., " A dual attention network based on efficientNet-B2 for short-term fish school feeding behavior analysis in aquaculture," Computers and Electronics in Agriculture, Vol. 187, August 2021, pp. 106316.
  7. Zhou, C., Xu, D., Chen, L., Zhang, S., Sun, C., Yang, X., and Wang, Y., "Evaluation of fish feeding intensity in aquaculture using a convolutional neural network and machine vision," Aquaculture, Vol. 507, May 2019, pp. 457-465. https://doi.org/10.1016/j.aquaculture.2019.04.056
  8. 농수축산신문, [분석]양어용 배합사료 확대, 무엇이 필요한가, Retrieved October 08, 2019, Available:https://www.aflnews.co.kr/news/articleView.html?idxno=162268
  9. 한국수산경제, 생사료, 미래 양식의 재앙이 될 수 있다, Retrieved February 23, 2017, Available:http://www.fisheco.com/news/articleView.html?idxno=61953
  10. 해양수산부, 수산혁신 2030 계획 발표, Retrieved February 13, 2019, Available: https://www.mof.go.kr/doc/ko/selectDoc.do?menuSeq=971&bbsSeq=10&docSeq=26599
  11. 해양수산부, 우리 수산업을 보호하는 마법, 배합사료, Retrieved September 29, 2021, Available:https://blog.naver.com/koreamof/222519795287
  12. 현대해양, 생사료 먹인 고기가 더 잘팔리는데 배합사료 의무화 가능할까?, Retrieved June 05, 2020, Available: http://www.hdhy.co.kr/news/articleView.html?idxno=12315