基于改进YOLOv5s的红外舰船检测算法

doi:10.12305/j.issn.1001-506X.2023.08.15

摘要/Abstract

摘要：

针对反舰导弹红外成像导引头在舰船尺度角度变化剧烈以及检测弱小舰船目标时检测能力差的问题, 提出一种基于改进YOLOv5s的目标检测算法，该方法综合考虑舰船目标特点。首先, 采用深度可分离卷积模块降低网络模型参数; 其次, 在主干网络引入坐标注意力机制提升关注目标通道信息特征的能力; 然后, 使用自适应空间特征融合策略优化空间权重分配; 最后，对损失函数进行改进, 提高目标检测框的可靠度。对比实验证明所提算法可将检测精度由原来的86.47%提升至91.64%, 全类平均精度(mean average precision, mAP)指标由原来的85.56%提升到89.35%，且在相同条件下所提算法亦优于其他目标检测网络模型。

关键词: 舰船检测, 红外图像, YOLOv5s, 坐标注意力机制, 特征融合

Abstract:

Infrared imaging seeker of anti-ship missile varies sharply in ship scale angle and its detection ability is poor when detecting small and weak ship targets. To solve this problem, a target detection algorithm based on improved YOLOv5s is proposed in this paper. Firstly, the depth separable convolution module is used to reduce the network model parameters. Secondly, the coordinate attention mechanism is introduced into the backbone network to improve the ability to pay attention to the target channel information features. Then, the adaptive spatial feature fusion strategy is used to optimize the spatial weight allocation. Finally, the loss function is improved to improve the reliability of the target detection frame. Comparative experiments verify that the proposed method improve the detection precision from the original 86.47% to 91.64%, and the mean average precision (mAP) index is improved from the original 85.56% to 89.35%. The improved algorithm also outperforms other target detection network models under the same conditions.

Key words: ship detection, infrared imagery, YOLOv5s, coordinate attention mechanism, feature fusion

中图分类号:

TP391.4

李海军, 孔繁程, 林云. 基于改进YOLOv5s的红外舰船检测算法[J]. 系统工程与电子技术, 2023, 45(8): 2415-2422.

Haijun LI, Fancheng KONG, Yun LIN. Infrared ship detection algorithm based on improved YOLOv5s[J]. Systems Engineering and Electronics, 2023, 45(8): 2415-2422.

图/表 14

图1

图2

图3

图4

图5

图6

图7

表1

图8

图9

表2

图10

表3

图11

参考文献 30

14	BOCHKOYSKIY A, WANG C Y, LIAO H Y. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. [2022-03-31]. http://arxiv.org/abs/2004.10934.
15	FU Q, CHEN J, YANG W, et al. Nearshore ship detection on SAR image based on Yolov5[C]//Proc. of the 2nd China International SAR Symposium, 2021.
16	SONG T, SUNOK K, SUNGTAI K, et al. Context preserving instance level augmentation and deformable convolution networks for sar ship detection[EB/OL]. [2022-03-31]. http://arxiv.org/abs/2202.06513.
17	ZHU X K, LYU S C, WANG X, et al. TPH-YOLOv5: improved YOLOv5 based on transformer prediction head for object detection on drone capture scenarios[C]//Proc. of the IEEE International Conference on Computer Vision Workshops, 2021: 2778-2788.
18	CHOLLET F. Xception: deep learning with depthwise separable convolutions[EB/OL]. [2022-03-31]. http://arxiv.org/abs/1610.02357.
19	HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
20	WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proc. of the European Conference on Computer Vision, 2018: 3-19.
21	WANG Q L, WU B G, ZHU P F, et al. Efficient channel attention for deep convolutional neural networks[C]//Proc. of the Conference on Computer Vision and Pattern Recognition, 2019.
22	HOU Q B, ZHOU D Q, FENG J S. Coordinate attention for efficient mobile network design[EB/OL]. [2022-03-31]. http://arxiv.org/abs/2103.02907.
23	LIU S, QI L, QIN H, et al. Path aggregation network for instance segmentation[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 8759-8768.
24	ONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for bio-medical image segmentation[C]//Proc. of the International Conference on Medical Image Computing and Computer Assisted Intervention, 2015: 234-241.
25	LIN T Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proc. of the IEEE Confe-rence on Computer vision and Pattern Recognition, 2017: 2117-2125.
1	李海军, 魏嘉彧, 牟俊杰, 等. 雷达/红外复合制导空舰导弹搜捕问题综述[J]. 兵器装备工程学报, 2021, 42 (12): 1- 6.
	LI H J , WEI J Y , MU J J , et al. Overview of radar/infrared composite guided air-to-ship missile hunting[J]. Journal of Ordnance Equipment Engineering, 2021, 42 (12): 1- 6.
2	CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]//Proc. of the European Conference on Computer Vision, 2020: 213-229.
3	YAO Z Y, AI J Y, LI J B, et al. Efficient DETR: improving end-to-end object detector with dense prior[EB/OL]. [2022-03-31]. http://arxiv.org/abs/2104.01318.
4	HE K M, GKIOXARI G, P DOLLAR, et al. Mask R-CNN[C]//Proc. of the IEEE International Conference on Computer Vision, 2017: 2980-2988.
5	REN S Q , HE K M , GIRSHICK R , et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Trans.on Pattern Analysis & Machine Intelligence, 2017, 39 (6): 1137- 1149.
6	CHEN X L, GUPTA A. An implementation of faster RCNN with study for region sampling[EB/OL]. [2022-03-31]. http://arXiv.org/abs/02138.
7	DAI J F, LI Y, HE K M, et al. R-FCN: object detection via region-based fully convolutional networks[EB/OL]. [2022-03-31]. http://arxiv.org/abs/1605.06409.
8	SINGH B, LI H D, SHARMA A, et al. R-FCN-3000 at 30 FPS: decoupling detection and classification[EB/OL]. [2022-03-31]. http://arxiv.org/abs/1712.01802.
9	WANG R F, XU F Y, PEI J F, et al. An improved faster R-CNN based on MSER decision criterion for SAR image ship detection in harbor[C]//Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2019: 1322-1325.
10	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multi-Box detector[C]//Proc. of the European Conference on Computer Vision, 2016: 21-37.
11	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 779-788.
12	REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 7263-7271.
13	REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. [2022-03-31]. http://arxiv.org/abs/1804.02767.
26	LIU S T, HAUNG D, WANG Y H. Learning spatial fusion for single-shot object detection[EB/OL]. [2022-03-31]. http://arxiv.org.abs/1911.09516.
27	REZATOFIGHI H, TSOI N, GWEAK J Y, et al. Generalized intersection over union: a metric and a loss for bounding box regression[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 658-666.
28	HE J B, SARAH E, MA X J, et al. Alpha-IOU: a family of power intersection over union losses for bounding box regression[EB/OL]. [2022-03-31]. http://arxiv.org/abs/2110.16375.
29	李晨瑄, 顾佼佼, 王磊, 等. 多尺度特征融合的anchor-free轻量化舰船要害检测算法[J]. 北京航空航天大学学报, 2022, 48 (10): 2006- 2919.
	LI C X , GU J J , WANG L , et al. Key detection algorithm of anchor-free light weight ship based on multiscale feature fusion[J]. Journal of Beijing University of Aeronautics and Astronautics, 2022, 48 (10): 2006- 2919.
30	PADILA R, NETTO S L, DA S E. A survey on performance metrics for object-detection algorithms[C]//Proc. of the International Conference on Systems, Signals and Image Processing, 2020: 237-242.

参数	取值
学习速度	0.01
优化器	Adam
基本大小	8
轮次	200

序号	DSC	CA	ASFF	参数量/10⁶	mAP/%	FPS
1	-	-	-	7.3	85.56	43.3
2	√	-	-	6.1	85.12	48.1
3	-	√	-	7.7	87.78	40.5
4	-	-	√	7.5	88.61	41.7
5	√	√	-	6.5	87.43	45.0
6	√	-	√	6.3	87.92	45.2
7	-	√	√	7.8	89.79	39.6
8	√	√	√	6.7	89.35	44.6

算法	P/%	R/%	mAP/%	FPS
Faster R-CNN	87.32	86.37	86.73	15.2
SSD	81.75	77.46	76.98	31.1
YOLOX	87.39	82.33	82.21	36.5
YOLOv4	83.93	79.67	79.06	36.7
Centernet	85.66	85.30	84.62	43.7
YOLOv5s	86.47	86.28	85.56	43.3
本文算法	91.64	89.71	89.35	44.6

[1]	王慧赢, 王春平, 付强, 韩子硕, 张冬冬. 基于图像特征的红外与低照度图像融合[J]. 系统工程与电子技术, 2023, 45(8): 2395-2404.
[2]	张涛, 杨小冈, 卢瑞涛, 谢学立, 刘闯. 基于关键点的遥感图像舰船目标检测[J]. 系统工程与电子技术, 2022, 44(8): 2437-2447.
[3]	赵晓枫, 吴飞, 徐叶斌, 牛家辉, 蔡伟, 张志利. 基于背景还原的红外伪装效果评价方法[J]. 系统工程与电子技术, 2022, 44(8): 2554-2561.
[4]	陈冬, 句彦伟. 基于语义分割实现的SAR图像舰船目标检测[J]. 系统工程与电子技术, 2022, 44(4): 1195-1201.
[5]	李永刚, 朱卫纲, 黄琼男, 李云涛, 何永华. 复杂背景下SAR图像近岸舰船目标检测[J]. 系统工程与电子技术, 2022, 44(10): 3096-3103.
[6]	李旭, 丁萌, 魏东辉, 吴晓舟, 曹云峰. VDAS中基于单目红外图像的深度估计方法[J]. 系统工程与电子技术, 2021, 43(5): 1210-1217.
[7]	崔邦彦, 田润澜, 王东风, 崔钢, 石静苑. 基于注意力机制和改进CLDNN的雷达辐射源识别[J]. 系统工程与电子技术, 2021, 43(5): 1224-1231.
[8]	陈冬, 句彦伟. 基于改进型YOLOv3的SAR图像舰船目标检测[J]. 系统工程与电子技术, 2021, 43(4): 937-943.
[9]	李晨瑄, 钱坤, 胥辉旗. 基于深浅层特征融合的舰船要害关键点检测算法[J]. 系统工程与电子技术, 2021, 43(11): 3239-3249.
[10]	孙嘉赤, 邹焕新, 邓志鹏, 李美霖, 曹旭, 马倩. 基于级联卷积神经网络的港口多方向舰船检测与分类[J]. 系统工程与电子技术, 2020, 42(9): 1903-1910.
[11]	花思齐, 赵伟, 刘建业. 基于改进滤波器和图像多尺度变换的背景抑制算法[J]. 系统工程与电子技术, 2020, 42(8): 1679-1684.
[12]	李亚兰, 金炜东, 葛鹏. 基于VMD和特征融合的辐射源信号识别[J]. 系统工程与电子技术, 2020, 42(7): 1499-1503.
[13]	张红颖, 胡文博. 多特征融合的尺度自适应相关滤波跟踪算法[J]. 系统工程与电子技术, 2019, 41(5): 951-957.
[14]	熊兴隆, 陈楠, 李永东, 马愈昭, 李猛, 冯帅. 基于卷积神经网络的低空风切变类型识别[J]. 系统工程与电子技术, 2019, 41(4): 772-779.
[15]	李健伟, 曲长文, 彭书娟, 邓兵. 基于卷积神经网络的SAR图像舰船目标检测[J]. 系统工程与电子技术, 2018, 40(9): 1953-1959.