Abstract
To achieve efficient and real-time classification and recognition of freshwater fish on limited equipment resources, this study proposes a method based on the YOLOv5s network to learn freshwater fish dataset images based on machine vision technology. Firstly, replace the backbone network of YOLOv5s with ShuffleNetV2 of the lightweight network to improve the efficiency of architecture design and ensure the detection speed. Secondly, the lightweight convolution GSConv is introduced to reduce the complexity of the network, maintain the accuracy of the network, and achieve the lightweight effect. Finally, the combination of ShuffleNetV2 and lightweight convolution GSConv can significantly improve the lightweight effect while maintaining accuracy. The algorithm was validated with four datasets of Rhodeinae, goldfish, grass carp, and stone moroko, which are all typical carp. The results showed that the improved network model FLOPs is 4.9 G, which was 69% less than YOLOv5s, the model size was 6.1 MB, which was 58% less than YOLOv5s, and the improved model mAP0.5 could reach 95.5%.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
The dataset generated and/or analyzed during this study has not yet been publicly available but can be obtained from the author upon reasonable request.
References
Guan W (2016) Research on China’s marine resource development and comprehensive marine management. Science and Technology Innovation Herald 22:120–121
Huang H, Chen X, Han Z et al (2022) Tea buds grading method based on multiscale attention mechanism and knowledge distillation. Trans Chinese Soc Agric Mach 53(9):399–407
Huang J (2020) Research on underwater fish target detection algorithm based on bionic dolphins [D]. Shenzhen University. https://doi.org/10.27321/d.cnki.gszdu.2020.000860
Huang Jing, Zhang Jin (2023) Object recognition camera based on YOLOv5. Computer Era 367(1):91–94. https://doi.org/10.16644/j.cnki.cn33-1094/tp.2023.01.023
Jilong LI, Cao Kun, Ding Fang et al (2017) Changes in trophic-level structure of the main fish species caught by China and their relationship with fishing method. J Fishery Sci China 24(1):109–119
Liang X, Xiao H (2023) Lightweight strip defect real-time detection algorithm based on SDD-YOLO [J/OL]. China Measur Test 1–8. http://kns.cnki.net/kcms/detail/51.1714.TB.20230109.1648.002.html
Liu Li, Kuang G (2009) Overview of image textural feature extraction methods. J Image Graphics 14(04):622–635
Liu Y, Zhang B, Wang X, etc. 2023 Research on real-time detection algorithm of ship fire based on lightweight CNN. Fire Sci Technol 42(1):42-46
Lyu X, Yang S, Zhang M, etc. 2023 lmproved fisheye lmage target detection algorithm based on YOLOv5 network [J/OL]. Comput Eng Appl: 1-13 [2023-02-28
Ma N, Zhang X, Zheng HT et al (2018) ShuffleNet V2: practical guidelines for efficient CNN architecture design. In: Proceedings of the European Conference on Computer Vision. ECCV. https://doi.org/10.48550/arXiv.1807.11164
Meng F, Cheng H, Li Ke et al (2020) Pruning filter in filter. Adv Neural Inf Process Syst 33:17629–17640
Ming Y, Qian W, Songling T et al (2022) Equipment identification and localization method based on improved YOLOv5s model for production line. Sensors 22(24). https://doi.org/10.3390/s222410011
Peng D, Wu T, Wang S, etc. 2023 YOLOv3-CBAM model of Yangtze River fishing ban scene recognition based on transfer learning. Yangtze Rive 54(2): 220–226+233. DOI: https://doi.org/10.16232/j.cnki. 1001–4179.2023.02.033
Prasetyo E, Suciati N, Fatichah C (2021) Multi-level residual network VGGNet for fish species classification. J King Saud Univ - Comput Inf Sci 34(8):5286–5295. https://doi.org/10.1016/j.jksuci.2021.05.015
Qian Chengwu, Zhang Xinxin (2022) A survey of lightweight object detection networks based on deep learning. Data Commun 211(6):39–44
Redmon J, Divvala S, Girshick R, et al. 2016 You only look once: unified, real-time object detection[C]. 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, USA: 779–788. doi: https://doi.org/10.1109/CVPR.2016.91
Rum MNS, Nawawi ZAF (2021) FishDeTec: a fish identification application using image recognition approach. Int J Adv Comput Sci Appl (IJACSA) 12(3). https://doi.org/10.14569/IJACSA.2021.0120312
Shafait F, Mian A, Shortis M et al (2016) Fish identification from videos captured in uncontrolled underwater environments. ICES J Mar Sci 73(10):2737–2746. https://doi.org/10.1093/icesjms/fsw106
Shao Y, Zhang D, Chu H, etc. 2022 A review of YOLO object detection based on deep learning. J Electron Inf Technol 44(10): 3697-3708
Sun J, Zhu W, Luo Y, etc. 2021 Recognizing the diseases of crop leaves in fields using improved Mobilenet-V2. Trans Chinese Soc Agric Eng 37 (22): 161-169
Vilas C, Antelo LT, Martin-Rodriguez F et al (2019) Use of computer vision onboard fishing vessels to quantify catches: the iObserver. Mar Policy 116:103714
Wang DF, Wang YT, Yang ZJ et al (2012) Research and thinking on the multifunctionality of my country’s fishery (serial part 4). China Fisheries (4):9–14. https://kns.cnki.net/kcms2/article/abstract?v=fCqJ37DMrBke48T639xwpH2s0bCq-WGPR0IQtYD9Osq2EH8X1lScaYfFPYeOljvMO_j9stsSVwXuzTj7KM1CwiTPYHKpXEHPT-ft6ddz37sBzzxIWtYFaoCar4iaXh3b&uniplatform=NZKPT&language=CHS
Wang K (2015) Study of freshwater fish species identification and online weight detection and equipment design [D]. Huazhong Agricultural University
Wan P, Pan H, Zong Li, Chen H (2012a) Study on crucian and cyprinoid species identification method based on machine vision. Guangdong Agric Sci 39(17):184–187. https://doi.org/10.16768/j.issn.1004-874x.2012.17.025
Wan P, Pan H, Long C, Chen H (2012b) Design of online identification device for freshwater fish species based on machine vision technology. Food Mach 28(6):164–167
Wang Y (2012) The development status and trend of my country’s protected aquaculture industry. China Fisheries 10:7–10
Xxa C, Wl D, Qda B (2020) Transfer learning and SE-ResNet152 networks-based for small-scale unbalanced fish species identification. Comput Electron Agric 180. https://kns.cnki.net/kcms2/article/abstract?v=fCqJ37DMrBlPl_yn7prKtFDpXfVHsmKOtSyFHgcAimwWD61h9bqqlWFPvVOEPQkFWFwE3jYJI0Wr2p6N6563fJ_OZpt_Zn6Ak26EiOmiJx4Shf5MckvtqwerRmgWYbS59zlIO0f5kxiowiTnGG9fKidR9N5rYO3KoCeA8KtsmVk=&uniplatform=NZKPT&language=CHS
Yang D, Zhang S, Tang X (2019) Research and development of fish species identification based on machine vision technology. Fisheries Inf Strategy 34(2):112–120. https://doi.org/10.13233/j.cnki.fishis.2019.02.006
Wu Y, Dai Y, Yuan Y (2014) Identification method of freshwater fish species using multi-kernel support vector machine with bee colony optimization. Trans Chinese Soc Agric Eng 30(16):312–319
Zhang H, Li Y, Zhou L et al (2022) Multi-target skeleton extraction method of beef cattlebased on improved YOLO 3. Trans Chinese Soc Agric Mach 53(3):285–293
Zhang M, Gao F, Yang W, Zhang H (2023) Wild-life object detection method applying segmentation gradient flow and feature dimensionality reduction. Electronics 12(8):377. https://doi.org/10.3390/electronics12081923
Zhang R, Ou J, Li X et al (2023) Lightweight algorithm for pineapple plant center detection based on improved an YoloV4 model. Trans Chinese Soc Agric Eng 39(4):135–143
Zhang S, Zhao S, Zhang H, Yang D (2017) Study on fish moving object extraction based on video analysis. Fisheries Inf Strategy 32(01):44–50. https://doi.org/10.13233/j.cnki.fishis.2017.01.008
Zhang Z, Niu Z, Zhao S (2011) Identification of freshwater fish species based on computer vision. Trans Chinese Soc Agric Eng 27(11):388–392
Funding
This work was supported by the Fishery Administration Project (D8021210076). The funders had no role in study design, data collection and analysis, the decision to publish, or the preparation of the manuscript.
Author information
Authors and Affiliations
Contributions
All authors have read and agreed to the published version of the manuscript.
Corresponding author
Ethics declarations
Ethical approval
This declaration is not applicable at this research.
Competing interests
The authors declare no competing interests.
Additional information
Handling Editor: Michael Hartnett
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Lyu, C., Zhong, WC. & Liu, S. Improved YOLOv5s for typical carp target detection. Aquacult Int (2024). https://doi.org/10.1007/s10499-024-01405-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10499-024-01405-7