Fusion of Deep Supervision and Improved YOLOv8 for Marine Targets Detection
CSTR:
Author:
Affiliation:

Tianjin Communication Center, Northern Navigation Service Center, Maritime Safety Administration

  • Article
  • | |
  • Metrics
  • |
  • Reference [21]
  • | |
  • Cited by
  • | |
  • Comments
    Abstract:

    Aiming at the problem that a large number of existing artificial intelligence algorithms are difficult to detect stably due to the complex attitude and variable scale of marine targets, a detection algorithm based on depth supervision and improved YOLOv8 is proposed. Firstly, a multi-scale convolution module is designed to extract the feature information of the target's multi-receptive fields and reduce the missed detection rate. Then, a deep supervision network is added to improve the utilization ratio of deep class information and shallow location information, and optimize the performance of target feature extraction of the backbone network. Finally, a channel attention mechanism is introduced in the detection head to filter the irrelevant information and enhance the recognition rate of the key features. The experimental results in the marine target dataset show that the mAP value of the improved algorithm reaches 93.69% and the recall rate reaches 85.16%, which is 7.38 and 8.52 percentage points higher than the original model respectively, and is better than the classical algorithm and the novel algorithm, The detection time is about 14ms, which can meet the requirements of real-time target detection at marine and provide effective technical references for shipping management and marine accident prevention.

    Reference
    [1] 王亮,陈建华,李烨.一种基于深度学习的无人艇海上目标识别技术[J].兵工学报,2022,43(S2):13-19.ANG Liang, CHEN Jianhua, LI Ye. A Target Identification Technique for Unmanned Surface Vessel Based on Deep Learning[J].Acta Armamentarii,2022,43(S2):13-19.
    [2] 刘安邦,施赛楠,杨静等.基于虚警可控梯度提升树的海面小目标检测[J].南京信息工程大学学报(自然科学版),2022,14(3):341-347.IU Anbang, SHI Sainan, YANG Jing, et al. Sea-surface small target detection based on false-alarm-controllable gradient boosting decision tree[J]. Journal of Nanjing University of Information Science Technology(Natural Science Edition),2022,14(3):341-347.
    [3] 陈卓,王飞,陈奕宏等.基于激光雷达的无人艇海上目标检测与跟踪方法研究[J].中国造船,2022,63(6):264-272.HEN Zhuo, WANG Fei, CHEN Yihong, et al. Research on Marine Target Detection and Tracking Method of Unmanned Surface Vehicle Based on Lidar[J].Shipbuilding of China,2022,63(6):264-272.
    [4] Meiyan P A N, Jun S U N, Yuhao Y, et al. Improved TQWT for marine moving target detection[J]. Journal of Systems Engineering and Electronics, 2020, 31(3): 470-481.
    [5] 薛安克,毛克成,张乐.多分类器联合虚警可控的海上小目标检测方法[J].电子与信息学报,2023,45(7):2528-2536.UE Anke, MAO Kecheng, ZHANG Le. Multi-feature Marine Small Target Detection Based on Multi-class Classifier[J].Journal of Electronics Information Technology,2023,45(7):2528-2536.
    [6] Redmon J, Divvala S, Girshick R, et al. You only look once: Unified, real-time object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 779-788.
    [7] 周薇娜,丁豪文,周颖.一种海上弱小运动船舶实时检测方法[J].合肥工业大学学报(自然科学版),2021,44(9):1187-1192.HOU Weina, DING Haowen, ZHOU Ying. A real-time detection method for dim and small moving ships at sea[J].Journal of Hefei University of Technology,2021.44(9):1187-1192.
    [8] Hong Z, Yang T, Tong X, et al. Multi-scale ship detection from SAR and optical imagery via a more accurate YOLOv3[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 6083-6101.
    [9] Fu H, Song G, Wang Y. Improved YOLOv4 marine target detection combined with CBAM[J]. Symmetry, 2021, 13(4): 623.
    [10] Zheng J C, Sun S D, Zhao S J. Fast ship detection based on lightweight YOLOv5 network[J]. IET Image Processing, 2022, 16(6): 1585-1593.
    [11] Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2023: 7464-7475.
    [12] Feng C, Zhong Y, Gao Y, et al. Tood: Task-aligned one-stage object detection[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV). IEEE Computer Society, 2021: 3490-3499.
    [13] Li X, Wang W, Wu L, et al. Generalized focal loss: Learning qualified and distributed bounding boxes for dense object detection[J]. Advances in Neural Information Processing Systems, 2020, 33: 21002-21012.
    [14] Lee C Y, Xie S, Gallagher P, et al. Deeply-supervised nets[C]//Artificial intelligence and statistics. Pmlr, 2015: 562-570.
    [15] 李校林,王复港,张鹏飞等.基于多尺度特征提取的YOLOv5s算法优化[J].计算机工程与科学,2023,45(6):1054-1062.I Xiaolin, WANG Fugang, ZHANG Pengfei, et al. YOLOv5s algorithm optimization based on multi-scale feature extraction[J].Computer Engineering and Science,2023,45(6):1054-1062.
    [16] Yu W, Yang K, Yao H, et al. Exploiting the complementary strengths of multi-layer CNN features for image retrieval[J]. Neurocomputing, 2017, 237: 235-241.
    [17] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.
    [18] 赵春江,梁雪文,于合龙等.基于改进YOLO v7的笼养鸡/蛋自动识别与计数方法[J].农业机械学报,2023,54(7):300-312.HAO Chunjiang, LIANG Xuewen, YU Helong, et al. Automatic Identification and Counting Method of Caged Hens and Eggs Based on Improved YOLO v7[J].Transactions of the Chinese Society for Agricultural Machinery,2023,54(7):300-312.
    [19] Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox detector[C]//Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14. Springer International Publishing, 2016: 21-37.
    [20] Cai Z, Vasconcelos N. Cascade r-cnn: Delving into high quality object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 6154-6162.
    [21] Tian Z, Shen C, Chen H, et al. Fcos: Fully convolutional one-stage object detection[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2019: 9627-9636.
    Related
    Cited by
    Comments
    Comments
    分享到微博
    Submit
Get Citation
Share
Article Metrics
  • Abstract:99
  • PDF: 0
  • HTML: 0
  • Cited by: 0
History
  • Received:August 23,2023
  • Revised:October 08,2023
  • Adopted:October 16,2023
Article QR Code

Address:No. 219, Ningliu Road, Nanjing, Jiangsu Province

Postcode:210044

Phone:025-58731025