Lightweight ship target detection algorithm based on improved YOLOv5

doi:10.11772/j.issn.1001-9081.2022071096

Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (3): 923-929.DOI: 10.11772/j.issn.1001-9081.2022071096

• Multimedia computing and computer simulation • Previous Articles

Lightweight ship target detection algorithm based on improved YOLOv5

Jiadong LI¹^,²(), Danpu ZHANG², Yaqiong FAN², Jianfeng YANG²

^1.The 2nd Institute of China Aerospace Science and Industry Corporation，Beijing 100039，China
^2.Changfeng Science Technology Industry Group Company Limited，Beijing Aerospace Changfeng Company Limited，Beijing 100039，China

Received:2022-07-28 Revised:2022-09-21 Accepted:2022-09-21 Online:2022-11-16 Published:2023-03-10
Contact: Jiadong LI
About author:ZHANG Danpu， born in 1986， Ph. D.， senior engineer. Her research interests include intelligent video analysis， big data analysis.
FAN Yaqiong， born in 1984， M. S.， research fellow. Her research interests include image processing， data analysis and mining.
YANG Jianfeng， born in 1991， M. S. His research interests include image processing， machine learning.
Supported by:
National Key Research and Development Program of China(2020YFC0833406)

基于改进YOLOv5的轻量级船舶目标检测算法

李佳东¹^,²(), 张丹普², 范亚琼², 杨剑锋²

^1.中国航天科工集团第二研究院，北京 100039
^2.北京航天长峰股份有限公司北京航天长峰科技工业集团有限公司，北京 100039

通讯作者: 李佳东
作者简介:李佳东（1998—），男，河北邯郸人，硕士研究生，主要研究方向：深度学习、图像处理
张丹普（1986—），女，河南平顶山人，高级工程师，博士，主要研究方向：视频智能分析、大数据分析
范亚琼（1984—），女，山西祁县人，研究员，硕士，主要研究方向：图像处理、数据分析挖掘
杨剑锋（1991—），男，四川成都人，硕士，主要研究方向：图像处理、机器学习。
基金资助:
国家重点研发计划项目(2020YFC0833406)

Abstract

Abstract:

Aiming at the problem of low accuracy of ship target detection at sea， a lightweight ship target detection algorithm YOLOShip was proposed on the basis of the improved YOLOv5. Firstly， dilated convolution and channel attention were introduced into Spatial Pyramid Pooling-Fast （SPPF） module， which integrated spatial feature details of different scales， strengthened semantic information， and improved the model’s ability to distinguish foreground and background. Secondly， coordinate attention and lightweight mixed depthwise convolution were introduced into Feature Pyramid Network （FPN） and Path Aggregation Network （PAN） structures to strengthen important features in the network， obtain features with more detailed information， and improve model detection ability and positioning precision. Thirdly， considering the uneven distribution and relatively small scale changes of targets in the dataset， the model performance was further improved while the model was simplified by modifying the anchors and decreasing the number of detection heads. Finally， a more flexible Polynomial Loss （PolyLoss） was introduced to optimize Binary Cross Entropy Loss （BCE Loss） to improve the model convergence speed and model precision. Experimental results show that on dataset SeaShips， in comparison with YOLOv5s，YOLOShip has the Precision， Recall， mAP@0.5 and mAP@0.5：0.95 increased by 4.2， 5.7， 4.6 and 8.5 percentage points. Thus， by using the proposed algorithm， better detection precision can be obtained while meeting the requirements of detection speed， effectively achieving high-speed and high-precision ship detection.

Key words: ship detection, YOLOv5 (You Only Look Once version 5), attention mechanism, dilated convolution, mixed depthwise convolution

摘要：

针对海上船舶目标检测准确率不高的问题，提出一种基于改进YOLOv5的轻量级船舶目标检测算法YOLOShip。首先将空洞卷积与通道注意力（CA）引入空间金字塔快速池化（SPPF）模块，以融合不同尺度的空间特征细节信息，强化语义信息，提升区分前景与背景的能力；其次将协同注意力与轻量化的混合深度卷积引入特征金字塔网络（FPN）和路径聚合网络（PAN）结构中，以强化网络中的重要特征，获取含有更多细节信息的特征，并提升模型检测能力及定位精度；然后考虑到数据集中目标分布不均匀及尺度变化相对较小的特点，在修改锚框，减少检测头数量以精简模型的同时进一步提升模型性能；最后，引入更加灵活的多项式损失（PolyLoss）以优化二元交叉熵损失（BCE Loss），提升模型收敛速度及模型精度。在SeaShips数据集上的实验结果表明，相较于YOLOv5s，YOLOShip的精确率、召回率、mAP@0.5与mAP@0.5：0.95分别提升4.2、5.7、4.6和8.5个百分点，能在满足检测速度要求的同时得到更优的检测精度，有效地实现了高速、高精度的船舶检测。

关键词: 船舶检测, YOLOv5, 注意力机制, 空洞卷积, 混合深度卷积

CLC Number:

TP183

Jiadong LI, Danpu ZHANG, Yaqiong FAN, Jianfeng YANG. Lightweight ship target detection algorithm based on improved YOLOv5[J]. Journal of Computer Applications, 2023, 43(3): 923-929.

李佳东, 张丹普, 范亚琼, 杨剑锋. 基于改进YOLOv5的轻量级船舶目标检测算法[J]. 《计算机应用》唯一官方网站, 2023, 43(3): 923-929.

Figures/Tables 11

References 22

1	齐亮，李邦昱，陈连凯. 基于改进的Faster R-CNN船舶目标检测算法［J］. 中国造船， 2020， 61（S1）： 40-51. 10.3969/j.issn.1000-4882.2020.z1.006
	QI L， LI B Y， CHEN L K. Ship target detection algorithm based on improved Faster R-CNN［J］. Shipbuilding of China， 2020， 61（S1）：40-51. 10.3969/j.issn.1000-4882.2020.z1.006
2	SUN J W， XU Z J， LIANG S S. NSD-SSD： a novel real-time ship detector based on convolutional neural network in surveillance video［J］. Computational Intelligence and Neuroscience， 2021， 2021： No.7018035. 10.1155/2021/7018035
3	段敬雅，李彬，董超，等. 基于YOLOv2的船舶目标检测分类算法［J］.计算机工程与设计， 2020， 41（6）：1701-1707.
	DUAN J Y， LI B， DONG C， et al. Detection and classification of ship target based on YOLOv2［J］. Computer Engineering and Design， 2020， 41（6）：1701-1707.
4	盛明伟，李俊，秦洪德，等. 基于改进YOLOv3的船舶目标检测算法［J］. 导航与控制， 2021， 20（2）：95-109.
	SHENG M W， LI J， QIN H D， et al. Ship target detection algorithm based on the improved YOLOv3［J］. Navigation and Control， 2021， 20（2）：95-109.
5	CHEN D H， SUN S R， LEI Z J， et al. Ship target detection algorithm based on improved YOLOv3 for maritime image［J］. Journal of Advanced Transportation， 2021， 2021： No.9440212. 10.1155/2021/9440212
6	LI H， DENG L B， YANG C， et al. Enhanced YOLO v3 tiny network for real-time ship detection from visual image［J］. IEEE Access， 2021， 9： 16692-16706. 10.1109/access.2021.3053956
7	孔刘玲，刘秀文. 基于改进YOLOv4算法的船舶目标检测方法［J］.船舶工程， 2022， 44（1）： 96-103， 147.
	KONG L L， LIU X W. Ship target detection algorithm based on improved YOLOv4［J］. Ship Engineering， 2022， 44（1）：96-103， 147.
8	HAN X， ZHAO L， NING Y， et al. ShipYOLO： an enhanced model for ship detection［J］. Journal of Advanced Transportation， 2021， 2021： No.1060182. 10.1155/2021/1060182
9	HE K M， ZHANG X Y， REN S Q， et al. Spatial pyramid pooling in deep convolutional networks for visual recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2015， 37（9）： 1904-1916. 10.1109/tpami.2015.2389824
10	ZHOU S Y， YIN J. YOLO-Ship： a visible light ship detection method［C］// Proceedings of the 2nd International Conference on Consumer Electronics and Computer Engineering. Piscataway： IEEE， 2022： 113-118. 10.1109/iccece54139.2022.9712768
11	JOCHER G. YOLOv5 releases v 6.1 - TensorRT， TensorFlow Edge TPU and OpenVINO export and inference［CP/OL］. ［2022-03-10］..
12	LIN T Y， GOYAL P， GIRSHICK R， et al. Focal Loss for dense object detection［C］// Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway： IEEE， 2017： 2999-3007. 10.1109/iccv.2017.324
13	LIN T Y， DOLLÁR P， GIRSHICK R， et al. Feature pyramid networks for object detection［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2017： 936-944. 10.1109/cvpr.2017.106
14	LIU S， QI L， QIN H F， et al. Path aggregation network for instance segmentation［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 8759-8768. 10.1109/cvpr.2018.00913
15	LENG Z Q， TAN M X， LIU C X， et al. PolyLoss： a polynomial expansion perspective of classification Loss functions［EB/OL］. ［2022-06-21］..
16	CHEN L C， PAPANDREOU G， KOKKINOS I， et al. DeepLab： semantic image segmentation with deep convolutional nets， atrous convolution， and fully connected CRFs［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2018， 40（4）： 834-848. 10.1109/tpami.2017.2699184
17	WOO S， PARK J， LEE J Y， et al. CBAM： convolutional block attention module［C］// Proceedings of the 2018 European Conference on Computer Vision， LNCS 11211. Cham： Springer， 2018： 3-19.
18	TAN M， LE Q V. MixConv： mixed depthwise convolutional kernels［C］// Proceedings of the 2019 British Machine Vision Conference. Durham： BMVA Press， 2019： No.116. 10.1109/iccvw.2019.00249
19	CHOLLET F. Xception： deep learning with depthwise separable convolutions［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2017：1800-1807. 10.1109/cvpr.2017.195
20	HOU Q B， ZHOU D Q， FENG J S. Coordinate attention for efficient mobile network design［C］// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2021： 13708-13717. 10.1109/cvpr46437.2021.01350
21	HURTIK P， MOLEK V， HULA J， et al. Poly-YOLO： higher speed， more precise detection and instance segmentation for YOLOv3［J］. Neural Computing and Applications， 2022， 34（10）： 8275-8290. 10.1007/s00521-021-05978-9
22	SHAO Z F， WU W J， WANG Z Y， et al. SeaShips： a large-scale precisely annotated dataset for ship detection［J］. IEEE Transactions on Multimedia， 2018， 20（10）： 2593-2604. 10.1109/tmm.2018.2865686

实验编号	模型	精确率	召回率	mAP@0.5	mAP@0.5：0.95
1	YOLOv5s	91.5	81.1	88.1	52.9
2	YOLOv5s+PolyLoss	90.7	83.6	89.0	53.0
3	YOLOv5s+SPPDC+PolyLoss	90.9	85.7	90.2	53.2
4	YOLOv5s+Improved FPN+PAN+PolyLoss	93.1	86.0	90.7	56.7
5	YOLOv5s+4Anchors+PolyLoss	89.3	85.6	90.4	54.8
6	YOLOv5s+9Anchors+PolyLoss	91.5	84.4	90.6	53.2

实验编号	模型	精确率	召回率	mAP@0.5	mAP@0.5：0.95
1	YOLOv5s	91.5	81.1	88.1	52.9
2	YOLOv5s+PolyLoss	90.7	83.6	89.0	53.0
3	YOLOv5s+SPPDC+PolyLoss	90.9	85.7	90.2	53.2
4	YOLOv5s+Improved FPN+PAN+PolyLoss	93.1	86.0	90.7	56.7
5	YOLOv5s+4Anchors+PolyLoss	89.3	85.6	90.4	54.8
6	YOLOv5s+9Anchors+PolyLoss	91.5	84.4	90.6	53.2

数据集	模型	精确率	召回率	mAP@0.5	mAP@0.5：0.95
验证集	YOLOv5s	91.5	81.1	88.1	52.9
验证集	YOLOShip	95.7	86.8	92.7	61.4
测试集	YOLOv5s	85.2	80.8	86.9	49.8
测试集	YOLOShip	90.8	88.3	93.3	57.1

数据集	模型	精确率	召回率	mAP@0.5	mAP@0.5：0.95
验证集	YOLOv5s	91.5	81.1	88.1	52.9
验证集	YOLOShip	95.7	86.8	92.7	61.4
测试集	YOLOv5s	85.2	80.8	86.9	49.8
测试集	YOLOShip	90.8	88.3	93.3	57.1

硬件配置	Batch Size	帧率/（frame·s^-1）
i5-4200H+940M	1	14
	8	17
	16	17
i9-9900+RTX2060	1	52
	8	156
	16	161

Lightweight ship target detection algorithm based on improved YOLOv5

基于改进YOLOv5的轻量级船舶目标检测算法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 11

References 22

Related Articles 15

Recommended Articles

Metrics

[1]	Cong YIN, Hanping HU. Parameter identification model for time-delay chaotic systems based on temporal attention mechanism [J]. Journal of Computer Applications, 2023, 43(3): 842-847.
[2]	Jie SUN, Shaoxin WU, Xuejun WANG, Jing HUA. Efficient person search algorithm and optimization with Sophon SC5+ chip architecture [J]. Journal of Computer Applications, 2023, 43(3): 744-751.
[3]	Guihui CHEN, Jinyu LIN, Yuehua LI, Zhongbing LI, Yuli WEI, Kai LU. Multi-stage low-illuminance image enhancement network based on attention mechanism [J]. Journal of Computer Applications, 2023, 43(2): 552-559.
[4]	Ping WANG, Nan CHEN, Lei LU. Fall detection algorithm based on scene prior and attention guidance [J]. Journal of Computer Applications, 2023, 43(2): 529-535.
[5]	Ming XU, Linhao LI, Qiaoling QI, Liqin WANG. Abductive reasoning model based on attention balance list [J]. Journal of Computer Applications, 2023, 43(2): 349-355.
[6]	Xiaomeng SHAO, Meng ZHANG. Temporal convolutional knowledge tracing model with attention mechanism [J]. Journal of Computer Applications, 2023, 43(2): 343-348.
[7]	Cong LIU, Genshun WAN, Jianqing GAO, Zhonghua FU. End-to-end speech recognition method based on prosodic features [J]. Journal of Computer Applications, 2023, 43(2): 380-384.
[8]	Wenju LI, Gan ZHANG, Liu CUI, Wanghui CHU. Lightweight traffic sign recognition model based on coordinate attention [J]. Journal of Computer Applications, 2023, 43(2): 608-614.
[9]	Hui LIU, Xiang MA, Linyu ZHANG, Rujin HE. Aspect-based sentiment analysis model integrating match-LSTM network and grammatical distance [J]. Journal of Computer Applications, 2023, 43(1): 45-50.
[10]	Zhijun SHEN, Lina MU, Jing GAO, Yuanhang SHI, Zhiqiang LIU. Review of fine-grained image categorization [J]. Journal of Computer Applications, 2023, 43(1): 51-60.
[11]	Zihao GUO, Lele DONG, Zhijian QU. Arthropod object detection method based on improved Faster RCNN [J]. Journal of Computer Applications, 2023, 43(1): 88-97.
[12]	Zeqiang SUN, Bingcai CHEN, Xiaobo CUI, Lei WANG, Yanuo LU. Strip steel surface defect detection by YOLOv5 algorithm fusing frequency domain attention mechanism and decoupled head [J]. Journal of Computer Applications, 2023, 43(1): 242-249.
[13]	Honggang YANG, Jiejie CHEN, Mengfei XU. Bilinear involution neural network for image classification of fundus diseases [J]. Journal of Computer Applications, 2023, 43(1): 259-264.
[14]	Jun ZHANG, Pengli WU, Lukui SHI, Jin SHI, Bin PAN. Deep learning model for multi-station temperature prediction combined with MOD11A1 and surface meteorological station data [J]. Journal of Computer Applications, 2023, 43(1): 321-328.
[15]	Bin ZOU, Cong ZHANG. Dense crowd detection algorithm based on Faster R-CNN [J]. Journal of Computer Applications, 2023, 43(1): 61-66.