Lightweight Fry Detection Algorithm Based on Improved YOLOv8: FD-YOLO

doi:10.19678/j.issn.1000-3428.0069278

Abstract

Abstract:

Based on deep learning, fish fry detection in aquaculture presents a potential for automated and precision management. To address the challenges of low device performance and high real-time requirements in fish fry detection, this paper proposes an improved lightweight fish fry detection algorithm called FD-YOLO. This algorithm replaces the original CSPDarkNet feature extraction network in YOLOv8 with a FasterNet variant and introduces Partial Convolutions (PConv) to reduce redundant computations and memory access. In the feature fusion stage, Depthwise Separable Convolutions (DWConv) are adopted, and the standard convolution process is decomposed into two relatively simple depthwise and pointwise convolutions executed in parallel, thereby further reducing the model complexity and computational resource demands. The model employs the Focal-EIoU loss function, enhancing detection accuracy and robustness. The experimental results demonstrate that the improved detection model significantly reduces the number of parameters and computational load by 91% and 85%, respectively. Moreover, the inference speed on the CPU is tripled compared with the baseline. The optimized fish fry detection algorithm effectively balances high precision with real-time performance, making it suitable for deployment on hardware platforms with limited resources. The enhanced algorithm demonstrates superior adaptability and practicality for addressing the critical needs of real-world aquaculture applications.

Key words: object detection, fry detection, lightweight, Partial Convolutions(PConv), Depthwise Separable Convolutions(DWConv)

摘要：

基于深度学习的鱼苗检测在水产养殖中的应用为自动化和精确化管理提供了可能。针对鱼苗检测中设备性能低、实时性要求高等问题, 提出一种改进YOLOv8的轻量化鱼苗检测算法FD-YOLO。将快速网络(FasterNet)替换YOLOv8原CSPDarkNet特征提取网络, 采用局部卷积(PConv)减少冗余计算和内存访问。在特征融合中引入深度可分离卷积(DWConv), 将标准卷积过程分解为相对简单的深度卷积和逐点卷积两个步骤并行处理, 进一步减少模型的复杂性和计算资源消耗。使用Focal-EIoU作为模型损失函数, 提高检测精度, 使得模型更具鲁棒性。实验结果表明, 改进后的检测模型参数量和计算量大幅降低, 模型参数量下降了91%, 计算量下降了85%, 在CPU上的推理速度加快了3倍。改进后的鱼苗检测算法能更好地兼顾高精度和实时性之间的平衡, 便于部署在资源有限的硬件平台上。

关键词: 目标检测, 鱼苗检测, 轻量化, 局部卷积, 深度可分离卷积

WANG Zeyu, XU Huiying, ZHU Xinzhong, HUANG Xiao, LIANG Jiajie, LI Chen. Lightweight Fry Detection Algorithm Based on Improved YOLOv8: FD-YOLO[J]. Computer Engineering, 2025, 51(4): 327-338.

王泽宇, 徐慧英, 朱信忠, 黄晓, 梁佳杰, 李琛. 基于改进YOLOv8的轻量化鱼苗检测算法: FD-YOLO[J]. 计算机工程, 2025, 51(4): 327-338.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0069278

https://www.ecice06.com/EN/Y2025/V51/I4/327

Figures/Tables 15

Fig.1 YOLOv8 network structure

Fig.2 FD-YOLO network structure

Fig.3 Overall architecture of the fast feature extraction network

Fig.4 Fast network module and local convolution architecture

Fig.5 Structure of depthwise separable convolution

Fig.6 Schematic diagram of EIoU each parameters

Fig.7 Schematic diagram of Frydatasets dataset collection

Fig.8 Comparison of FD-YOLO with mainstream models and bounding box loss curves at present

Fig.9 Comparison of fry detection results

References 36

1	张涵钰, 李振波, 李蔚然, 等. 基于机器视觉的水产养殖计数研究综述. 计算机应用, 2023, 43 (9): 2970- 2982.
	ZHANG H Y , LI Z B , LI W R , et al. Review of research on aquaculture counting based on machine vision. Journal of Computer Applications, 2023, 43 (9): 2970- 2982.
2	LI J , SUN J N , CUI X R , et al. Automatic counting method of fry based on computer vision. IEEJ Transactions on Electrical and Electronic Engineering, 2023, 18 (7): 1151- 1159. doi: 10.1002/tee.23821
3	LI W R , ZHU Q , ZHANG H Y , et al. A lightweight network for portable fry counting devices. Applied Soft Computing, 2023, 136, 110140. doi: 10.1016/j.asoc.2023.110140
4	REN S Q , HE K M , GIRSHICK R , et al. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39 (6): 1137- 1149. doi: 10.1109/TPAMI.2016.2577031
5	CAI Z W, VASCONCELOS N. Cascade R-CNN: delving into high quality object detection[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 6154-6162.
6	LIU W , ANGUELOV D , ERHAN D , et al. SSD: single shot MultiBox detector. Berlin, Germany: Springer, 2016.
7	LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[EB/OL]. [2023-12-10]. https://arxiv.org/abs/1708.02002v2.
8	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2016: 779-788.
9	CAI K W , MIAO X Y , WANG W , et al. A modified YOLOv3 model for fish detection based on MobileNetv1 as backbone. Aquacultural Engineering, 2020, 91, 102117. doi: 10.1016/j.aquaeng.2020.102117
10	REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. [2023-12-10]. https://arxiv.org/abs/1804.02767.
11	LI X, TANG Y H, GAO T W. Deep but lightweight neural networks for fish detection[C]//Proceedings of OCEANS'17. Washington D. C., USA: IEEE Press, 2017: 1-5.
12	SZEGEDY C, IOFFE S, VANHOUCKE V, et al. Inception-v4, inception-ResNet and the impact of residual connections on learning[C]//Proceedings of the 31st AAAI Conference on Artificial Intelligence. New York, USA: ACM Press, 2017: 4278-4284.
13	RAZA K , HONG S . Fast and accurate fish detection design with improved YOLO-v3 model and transfer learning. International Journal of Advanced Computer Science and Applications, 2020, 11 (2): 516- 525.
14	FENG D C , XIE J F , LIU T L , et al. Fry counting models based on attention mechanism and YOLOv4-tiny. IEEE Access, 2022, 10, 217854- 217866.
15	BOCHKOVSKIY A, WANG C Y, LIAO H M. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2004.10934v1.
16	黎袁富, 杜家豪, 莫家浩, 等. 基于YOLOX的鱼苗检测与计数. 电子元器件与信息技术, 2022, 6 (5): 192- 194.
	LI Y F , DU J H , MO J H , et al. Fry detection and counting based on YOLOX. Electronic Components and Information Technology, 2022, 6 (5): 192- 194.
17	GE Z, LIU S, WANG F, et al. YOLOx: exceeding YOLO series in 2021[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2107.08430.
18	ZHANG H Y , LI W R , QI Y Y , et al. Dynamic fry counting based on multi-object tracking and one-stage detection. Computers and Electronics in Agriculture, 2023, 209, 107871. doi: 10.1016/j.compag.2023.107871
19	刘康. 基于深度学习的鱼苗自动计数方法的研究与实现[D]. 马鞍山: 安徽工业大学, 2021.
	LIU K. Research and implementation of automatic counting method of fry based on deep learning[D]. Maanshan: Anhui Universit of Technology, 2021. (in Chinese)
20	CHEN J R, KAO S H, HE H, et al. Run, don't walk: chasing higher FLOPS for faster neural networks[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2023: 12021-12031.
21	HOWARD A G, ZHU M L, CHEN B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[EB/OL]. [2023-12-10]. https://arxiv.org/abs/1704.04861.
22	ZHENG Z H , WANG P , LIU W , et al. Distance-IoU loss: faster and better learning for bounding box regression. Artificial Intelligence, 2020, 34 (7): 12993- 13000.
23	ZHANG Y F , REN W Q , ZHANG Z , et al. Focal and efficient IOU loss for accurate bounding box regression. Neurocomputing, 2022, 506, 146- 157. doi: 10.1016/j.neucom.2022.07.042
24	CHOLLET F. Xception: deep learning with depthwise separable convolutions[EB/OL]. [2023-12-10]. https://arxiv.org/abs/1610.02357.
25	LU W, XU S, ZHAO Y, et al. Detrs beat YOLOs on real-time object detection[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2304.08069.
26	WANG C Y, BOCHKOVSKIY A, LIAO H M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2023: 7464-7475.
27	WANG C Y, YEH I H, LIAO H M. YOLOv9: learning what you want to learn using programmable gradient information[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2402.13616v2.
28	LIU Z, MAO H Z, WU C Y, et al. A ConvNet for the 2020s[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2022: 11976-11986.
29	HAN K, WANG Y H, TIAN Q, et al. GhostNet: more features from cheap operations[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, USA: IEEE Press, 2020: 1580-1589.
30	CUI C, GAO T, WEI S, et al. PP-LCNet: a lightweight CPU convolutional neural network[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2109.15099.
31	MA N N, ZHANG X Y, ZHENG H T, et al. ShuffleNet V2: practical guidelines for efficient CNN architecture design[C]//Proceedings of ECCV'18. Berlin, Germany: Springer, 2018: 122-138.
32	LI C, ZHOU A, YAO A. Omni-dimensional dynamic convolution[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2209.07947.
33	LI H L, LI J, WEI H B, et al. Slim-neck by GSConv: a lightweight-design for real-time detector architectures[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2206.02424v3.
34	LI J F, WEN Y, HE L H. SCConv: spatial and channel reconstruction convolution for feature redundancy[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2023: 6153-6162.
35	ZHANG X, SONG Y, SONG T, et al. AKConv: convolutional kernel with arbitrary sampled shapes and arbitrary number of parameters[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2311.11587.
36	ZHANG X, LIU C, YANG D G, et al. RFAConv: innovating spatial attention and standard convolutional operation[EB/OL]. [2023-12-10]. https://arxiv.org/abs/2304.03198.

[1]	LI Shuwei, HUANG Zhengxiang, HU Yun, LIU Xing, LU Xiao, GUO Chang, WU Chengzhong, WANG Yaonan. Low-Light Salient Object Detection Based on Source-Free Domain Adaptation [J]. Computer Engineering, 2025, 51(4): 75-84.
[2]	ZHANG Zhaoxin, HUANG Shize, ZHANG Bingjie, SHEN Tuo. Camouflaged Adversarial Example Generation Method for the Form of Motion Blur in Traffic Scenes [J]. Computer Engineering, 2025, 51(3): 45-53.
[3]	YUAN Yajian, MAO Li. A Lightweight Traffic Sign Detection Model with Enhanced Foregrounds [J]. Computer Engineering, 2025, 51(3): 54-63.
[4]	CUI Jinrong, YE Weihao, ZHENG Hong, LIU Tonglai, QI Long, XU Yong. Rice Seedling Counting in Complex Environments Based on Domain-Adaptive NWD-YOLOv5 [J]. Computer Engineering, 2025, 51(3): 320-333.
[5]	HUANG Shuyi, TAN Guang. Efficient Video Object Detection Based on Partitioning [J]. Computer Engineering, 2025, 51(2): 65-77.
[6]	SUN Haomiao, LI Zongmin, XIAO Qian, SUN Wenjie, ZHANG Wenxin. AI-Curling: An On-Site Curling Analysis and Decision-Making Method [J]. Computer Engineering, 2025, 51(2): 102-110.
[7]	LIU Shengjie, HE Ning, WANG Xin, YU Haigang, HAN Wenjing. Human Pose-Estimation Algorithm Based on Lightweight High-Resolution Network [J]. Computer Engineering, 2025, 51(2): 278-288.
[8]	ZHANG Yuan, LÜ Defang, MENG Jianjun, QI Wenzhe. Defect Detection of Rail Fasteners Based on Double Attention and GSSN Lightweight [J]. Computer Engineering, 2025, 51(2): 289-299.
[9]	LI Haifeng, LIU Sensen, WANG Huaichao, LI Nansha, ZHANG Yifan. Airport Pavement Underground Disease Detection Algorithm Integrating Association Reasoning [J]. Computer Engineering, 2025, 51(2): 397-406.
[10]	HU Shenglong, CHEN Bin, ZHANG Kaihua, SONG Huihui. Co-Saliency Object Detection Enhanced by Scene Structure Knowledge [J]. Computer Engineering, 2025, 51(1): 31-41.
[11]	ZHANG Changchang, LÜ Weidong, CAI Zijie, LIU Yankui. Lightweight Image Classification Algorithm Based on Domain Generalization [J]. Computer Engineering, 2025, 51(1): 182-189.
[12]	YANG Lisha, LI Maojun, HU Jianwen, WANG Dingxiang. Strip Steel Surface Defect Detection Algorithm Based on Improved YOLOv7-tiny [J]. Computer Engineering, 2025, 51(1): 208-215.
[13]	CAI Junmin, LIANG Zhengyou, SUN Yu, CHEN Ziao. Research on Lightweight Point Cloud Classification Based on Deformable 3D Graph Convolution [J]. Computer Engineering, 2024, 50(9): 255-265.
[14]	ZENG Yuqi, LIU Bo, ZHONG Baichang, ZHONG Jin. Student Classroom Behavior Detection Algorithm Based on Improved YOLOv8 in Smart Education [J]. Computer Engineering, 2024, 50(9): 344-355.
[15]	Rixin RAO, Yiwen WANG, Lizhi ZENG, Xintian TONG, Haitao ZHAO. Lightweight Network Model for Waste Cable Detection [J]. Computer Engineering, 2024, 50(8): 22-30.

Please choose a citation manager

Content to export