APPLICATION OF YOLO MODELS IN THE DETECTION OF FISH BEHAVIORAL CHANGE UNDER ACUTE EXPOSURE TO SYNTHETIC ESTROGEN IN THE ENVIRONMENT
DOI:
https://doi.org/10.11113/aej.v15.22771Keywords:
17alpha-ethinylestradiol, YOLO, computer vision, fish behavior, object detectionAbstract
Changes in the behavior of small fish have recently been commonly used in assessing the impact of water pollutants, especially those of the group of endocrine disruptors. Behavioral studies mostly use visual observations, which can introduce bias and inconsistency in observational results. Recent studies have developed computer vision tools for tracking fish movements that allow automatic detection of small fish movements with high accuracy and consistency. In addition, computer vision combined with machine learning can help analyze, identify, and predict changes in fish behavior, easily integrated into environmental and ecological monitoring systems. This study uses YOLO (You Only Look One) algorithm models to detect fish in video data. Comparing the effectiveness of YOLO versions with the training data set shows that the YOLOv8s model has the highest efficiency and is selected for detecting and analyzing fish behavior in the environmental impact assessment model. The amount of image data for training the YOLOv8s model is also determined to be approximately 800 images. The training results show that YOLOv8s has high detection efficiency with a high frequency of detecting 11 fish in video frames. Results from detecting and analyzing fish positions in video data using the trained YOLOv8s model showed that the males of both mosquitofish (Gambusia affinis) and medaka (Oryzias latipes) species were affected following a two-day acute exposure to the estrogenic stressor, 17a-ethinylestradiol, in the aquatic environment at a concentration of 5 ng/L. While male mosquitofish when not exposed to estrogen tended to pay more attention towards the tank compartment containing female medaka, when exposed to estrogen, they increased their tendency towards the compartment containing male medaka. Additional research is needed to increase the accuracy and effectiveness of the YOLO algorithms in fish detection for behavior evaluation in an environmental impact assessment model.
References
Thacharodi, A., Hassan, S., Hegde, T. A., Thacharodi, D. D., Brindhadevi, K. and Pugazhendhi, A. 2023. Water a major source of endocrine-disrupting chemicals: An overview on the occurrence, implications on human health and bioremediation strategies. Environmental Research. 231(1): 116097. DOI: https://doi.org/10.1016/j.envres.2023.116097
Petrie, B., Barden, R. and Kasprzyk-Hordern, B. 2015. A review on emerging contaminants in wastewaters and the environment: Current knowledge, understudied areas and recommendations for future monitoring. Water Research. 72: 3-27. DOI: https://doi.org/10.1016/j.watres.2014.08.053
Léon-Olea, M., Martyniuk, C. J., Orlando, E. F., Ottinger, M. A., Rosenfeld, C., Wolstenholme, J. and Trudeau, V L. 2014. Current concepts in neuroendocrine disruption. General and Comparative Endocrinology. 203: 158-173. DOI: https://doi.org/10.1016/j.ygcen.2014.02.005
Zhou, X., Yang, Z., Lou, Z., Li, H. and Chen, G. 2019. Endocrine disrupting chemicals in wild freshwater fishes: Species, tissues, sizes and human health risks. Environmental Pollution. 244: 462-468. DOI: https://doi.org/10.1016/j.envpol.2018.10.026
Mill, L. J. and Chichester, C. 2005. Review of evidence: Are endocrine-disrupting chemicals in the aquatic environment impacting fish populations? Science of the Total Environment. 343: 1-34. DOI: https://doi.org/10.1016/j.scitotenv.2004.12.070
Söffker, M. and Tyler, C. R. 2012. Endocrine disrupting chemicals and sexual behaviors in fish – a critical review on effects and possible consequences. Critical Reviews in Toxicology. 42(8): 653-668. DOI: https://doi.org/10.3109/10408444.2012.692114
Parker M. O. 2016. Adult vertebrate behavioural aquatic toxicology: Reliability andvalidity. Aquatic Toxicology. 170 (2016) 323–329. DOI: http://dx.doi.org/10.1016/j.aquatox.2015.09.001
Melvin S. D., et al. 2017. Towards improved behavioural testing in aquatic toxicology: Acclimation and observation times are important factors when designing behavioural tests with fish. Chemosphere. 180 (2017) 430e436. DOI: http://dx.doi.org/10.1016/j.chemosphere.2017.04.058
Hong X. and Zha J. 2019. Fish behavior: A promising model for aquatic toxicology research. Science of the Total Environment. 686 (2019) 311–321. DOI: https://doi.org/10.1016/j.scitotenv.2019.06.028
Orger, M. B. and de Polavieja, G. G. 2017. Zebrafish behavior: Opportunities and Challenges. Annual Review of Neuroscience. 2017. 40: 125–147. DOI: https://doi.org/10.1146/annurev-neuro-071714-033857
Dang, H. M., Inagaki, Y., Yamauchi, Y., Kurihara, T., Vo, C. H. and Sakakibara, Y. 2017. Acute exposure to 17alpha-ethinylestradiol alters aggressive behavior of mosquitofish (Gambusia affinis) toward Japanese medaka (Oryzias latipes). Bulletin of Environmental Contamination and Toxicology. 98(5): 643-648. DOI: https://doi.org/10.1007/s00128-016-2016-z
Voulodimos, A., Doulamis, N., Doulamis, A., and Protopapadakis, E. 2018. Deep learning for computer vision: A brief review. Computational Intelligence and Neuroscience. 2018: 7068349. DOI: https://doi.org/10.1155/2018/7068349
Majaj, N. J. and Pelli, D. G. 2028. Deep learning—Using machine learning to study biological vision. Journal of Vision. 18(13): 2. DOI: https://doi.org/10.1167/18.13.2
Shreesha, S., Manohara, P. M. M., Verma, U. and Pai, R. M. 2020. Computer Vision Based Fish Tracking and Behaviour Detection System. Proceedings of the IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER). 252-257. DOI: https://doi.org/10.1109/DISCOVER50404.2020.9278101
Wang, Q., Long, W., Wang, Y., Jiang, L. and Hu, L. 2021. Fish Behavior Detection Using Computer Vision: A Review. Proceedings of the 3rd International Conference on Artificial Intelligence and Advanced Manufacture (AIAM). 375 381. DOI: https://doi.org/10.1109/AIAM54119.2021.00082
Yang, L., Liu, Y., Yu, H., Fang, X., Song, L., Li, D. and Chen, Y. 2021. Computer Vision Models in Intelligent Aquaculture with Emphasis on Fish Detection and Behavior Analysis: A Review. Archives of Computational Methods in Engineering. 28: 2785–2816. DOI: https://doi.org/10.1007/s11831-020-09486-2
Nay, Z., Huggins, A. and Deane, F. 2021. Automated Decision-Making and Environmental Impact Assessments: Decisions, Data Analysis and Predictions. Law, Technology and Humans. 3(2): 76-90. DOI: https://doi.org/10.5204/lthj.1846
Yoneyama, A., Yeh, C. H. and JayKuo, C. C. 2005: Robust Vehicle and Traffic Information Extraction for Highway Surveillance. Eurasip Journal on Applied Signal Processing 2005. 2305-21. DOI: http://dx.doi.org/10.11113/jt.v79.9987
Redmon, J., Divvala, S., Girshick, R. and Farhadi, A. 2016. You Only Look Once: Unified, Real-Time Object Detection. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 779–788. DOI: https://doi.org/10.48550/arXiv.1506.02640
Terven, J., Córdova-Esparza, D. -M. and Romero-González, J. -A. 2023. A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS. Machine Learning and Knowledge Extraction. 5(4):1680 1716. DOI: https://doi.org/10.3390/make5040083
Redmon, J. and Farhadi, A. 2017. YOLO9000: Better, Faster, Stronger. Proceedings of the IEEE conference on computer vision and pattern recognition. 7263-7271. DOI: https://doi.org/10.1109/CVPR.2017.690
Redmon, J. and Farhadi, A. 2018. YOLOv3: An Incremental Improvement. ArXiv. arXiv: 1804.02767. DOI: https://doi.org/10.48550/arXiv.1804.02767
Bochkovskiy, A., Wang, C. -Y. and Liao, H. -Y. M. 2020. YOLOv4: Optimal Speed and Accuracy of Object Detection. ArXiv. arXiv:2004.10934. DOI: https://doi.org/10.48550/arXiv.2004.10934
Jocher, G., Chaurasia, A., Stoken, A. et al. 2022. ultralytics/yolov5: v7.0 - YOLOv5 SOTA Realtime Instance Segmentation. DOI: 10.5281/zenodo.7347926
Abdulsalam, M. and Aouf, N. 2020. Deep weed detector/classifier network for precision agriculture. Proceedings of the 2020 28th Mediterranean Conference on Control and Automation (MED). 1087-1092. DOI: https://doi.org/10.1109/MED48518.2020.9183325
Sanchez, P. R., Zhang, H., Ho, S. S. and De Padua, E. 2021. Comparison of one-stage object detection models for weed detection in mulched onions. Proceedings of the 2021 IEEE International Conference on Imaging Systems and Techniques (IST). 1-6. DOI: https://doi.org/10.1109/IST50367.2021.9651352
Rahman, A., Lu, Y. and Wang, H. 2022. Deep Neural Networks for Weed Detections Towards Precision Weeding. Proceedings of the 2022 ASABE Annual International Meeting. DOI: https://doi.org/10.13031/aim.202200845
Fatima, H. S., ul Hassan, I., Hasan, S., Khurram, M., Stricker, D. and Afzal, M. Z. 2023. Formation of a Lightweight, Deep Learning-Based Weed Detection System for a Commercial Autonomous Laser Weeding Robot. Applied Sciences. 13(6): 3997. DOI: https://doi.org/10.3390/app13063997
Terven, J., Córdova-Esparza, D.-M., Romero-González, J.-A.. 2023. A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS. Machine Learning and Knowledge Extraction. 5: 1680-1716. DOI: https://doi.org/10.3390/make5040083
Reyhanian, N., Volkova, K., Hallgren, S., Bollner, T., Olsson, P. E., Olsén, H. and Hällström, I. P. 2011. 17α-Ethinyl estradiol affects anxiety and shoaling behavior in adult male zebra fish (Danio rerio). Aquatic Toxicololgy. 105(1-2): 41-48. DOI: https://doi.org/10.1016/j.aquatox.2011.05.009
Dzieweczynski, T. L, Campbell, B. A, Marks, J. M. and Logan, B. 2014. Acute exposure to 17α-ethinylestradiol alters boldness behavioral syndrome in female Siamese fighting fish. Hormones and Behavior. 66(4): 577-584. DOI: https://doi.org/10.1016/j.yhbeh.2014.08.005
Heintz, M. M., Brander, S. M. and White, J. W. 2015. Endocrine Disrupting Compounds Alter Risk-Taking Behavior in Guppies (Poecilia reticulata). Ethology. 121(5): 480-491. DOI: https://doi.org/10.1111/eth.12362
León-Olea, M., Martyniuk, C. J., Orlando, E. F., Ottinger, M. A., Rosenfeld, C., Wolstenholme, J. and Trudeau, V. L. 2014. Current concepts in neuroendocrine disruption. General and Comparative Endocrinology. 203: 158-173. DOI: https://doi.org/10.1016/j.ygcen.2014.02.005
Petrie, B., Barden, R. and Kasprzyk-Hordern, B. 2015. A review on emerging contaminants in wastewaters and the environment: current knowledge, understudied areas and recommendations for future monitoring. Water Research. 72: 3-27. DOI: https://doi.org/10.1016/j.watres.2014.08.053
Parker, M. O. 2016. Adult vertebrate behavioural aquatic toxicology: Reliability and validity. Aquatic Toxicology. 170: 323–329. DOI: http://dx.doi.org/10.1016/j.aquatox.2015.09.001
Melvin, S. D., Petit, M. A., Duvignacq, M. C. and Sumpter, J. P. 2017. Towards improved behavioural testing in aquatic toxicology: Acclimation and observation times are important factors when designing behavioural tests with fish. Chemosphere. 180: 430-436. DOI: http://dx.doi.org/10.1016/j.chemosphere.2017.04.058
Hong, X. and Zha, J. 2019. Fish behavior: A promising model for aquatic toxicology research. Science of the Total Environment. 686: 311–321. DOI: https://doi.org/10.1016/j.scitotenv.2019.06.028
Orger, M. B. and de Polavieja, G. G. 2017. Zebrafish behavior: Opportunities and Challenges. Annual Review of Neuroscience. 40: 125–47. DOI: https://doi.org/10.1146/annurev-neuro-071714-033
