基于掩模注意型交互的SAR舰船实例分割

doi:10.12305/j.issn.1001-506X.2024.03.08

摘要/Abstract

摘要：

现有合成孔径雷达(synthetic aperture radar, SAR)舰船实例分割方法未实现掩模交互或交互性能有限, 导致检测精度较低。针对上述问题, 提出了一种基于掩模注意型交互(mask attention interaction, MAI)的SAR舰船实例分割方法MAI-Net。首先, MAI-Net使用了膨胀空间金字塔池化, 来获取多分辨率特征响应, 增强了对背景的鉴别能力。其次, MAI-Net使用了非局部注意力模块来抑制低价值信息, 实现了空间特征自注意。最后, MAI-Net提出了拼接混洗注意力模块来平衡不同特征图的贡献, 进一步提高了实例分割精度。在公开的像素级多边形分割SAR舰船检测数据集(polygon segmentation SAR ship detection dataset, PSeg-SSDD)上的实验结果表明, MAI-Net的SAR舰船实例分割精度高于现有其他11种对比模型, 实例分割精度达到61.1%, 高于次优模型1.5%。

关键词: 合成孔径雷达, 深度学习, 实例分割, 掩模注意型交互

Abstract:

The current synthetic aperture radar (SAR) ship instance segmentation models fail to realize mask interaction or the interaction performance is limited, resulting in low detection accuracy. To solve this problem, a SAR ship instance segmentation method based on mask attention interaction (MAI) is proposed, called MAI-Net. Firstly, MAI-Net uses the atrous spatial pyramid pooling (ASPP) to obtain multi-resolution feature responses and enhance the background identification capability. Secondly, MAI-Net uses a non-local block (NLB) to suppress useless information and realize spatial feature self-attention. Finally, MAI-Net proposes the concatenation shuffle attention block (CSAB), which can balance the contribution of different features and further improve the instance segmentation accuracy. The results on the public polygon segmentation SAR ship detection dataset (PSeg-SSDD) show that the SAR ship instance segmentation accuracy of MAI-Net is higher than that of the other eleven comparison models, the accuracy is 61.1%, 1.5% higher than the suboptimal model.

Key words: synthetic aperture radar (SAR), deep learning, instance segmentation, mask attention interaction

中图分类号:

TP751

张天文, 张晓玲, 邵子康, 曾天娇. 基于掩模注意型交互的SAR舰船实例分割[J]. 系统工程与电子技术, 2024, 46(3): 831-838.

Tianwen ZHANG, Xiaoling ZHANG, Zikang SHAO, Tianjiao ZENG. Mask attention interaction for SAR ship instance segmentation[J]. Systems Engineering and Electronics, 2024, 46(3): 831-838.

图/表 10

图1

图2

图3

图4

图5

表1

表2

表3

图6

表4

参考文献 42

1	ELDHUSET K . An automatic ship and ship wake detection system for spaceborne SAR images in coastal regions[J]. IEEE Trans.on Geoscience and Remote Sensing, 1996, 34 (4): 1010- 1019. doi: 10.1109/36.508418
2	ZHANG T W , ZHANG X L . High-speed ship detection in SAR images based on a grid convolutional neural network[J]. Remote Sensing, 2019, 11 (10): 1206. doi: 10.3390/rs11101206
3	CRISP J D . The state-of-the-art in ship detection in synthetic aperture radar imagery[M]. Edinburgh, Australia: DSTO Information Sciences Laboratory, 2004.
4	NOVAK L M , OWIRKA G J , BROWER W S , et al. The automatic target recognition system in SAIP[J]. The Lincoln Laboratory Journal, 1997, 10 (2): 187- 202.
5	CUI X C , SU Y , CHEN S W . A saliency detector for polarimetric SAR ship detection using similarity test[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12 (9): 3423- 3433. doi: 10.1109/JSTARS.2019.2925833
6	ZHANG T W , ZHANG X L , KE X , et al. LS-SSDD-v1.0: a deep learning dataset dedicated to small ship detection from large-scale Sentinel-1 SAR images[J]. Remote Sensing, 2020, 12 (18): 2997. doi: 10.3390/rs12182997
7	ZHANG T W , ZHANG X L , SHI J , et al. Balance scene learning mechanism for offshore and inshore ship detection in SAR images[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19, 4004905.
8	雷禹, 冷祥光, 周晓艳, 等. 基于改进ResNet网络的复数SAR图像舰船目标识别方法[J]. 系统工程与电子技术, 2022, 44 (12): 3652- 3660.
	LEI Y , LENG X G , ZHOU X Y , et al. Recognition method of ship target in complex SAR image based on improved ResNet network[J]. Systems Engineering and Electronics, 2022, 44 (12): 3652- 3660.
9	李男, 叶晓东, 王昊, 等. 基于改进YOLOv5的复杂场景下SAR图像船舶检测方法[J]. 信号处理, 2022, 38 (5): 1009- 1018.
	LI N , YE X D , WANG H , et al. A ship detection method for SAR images in complex scene based on improved YOLOv5[J]. Journal of Signal Processing, 2022, 38 (5): 1009- 1018.
10	GE Z, LIU S T, WANG F, et al. YOLOX: exceeding yolo series in 2021[EB/OL]. [2022-11-30]. https://doi.org/10.48550/arXiv.2107.08430 (2021).
11	ZHANG T W , ZHANG X , LIU C , et al. Balance learning for ship detection from synthetic aperture radar remote sensing imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 182, 190- 207. doi: 10.1016/j.isprsjprs.2021.10.010
12	ZHANG T W , ZHANG X L . A mask attention interaction and scale enhancement network for SAR ship instance segmentation[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19, 4511005.
13	WEI S J , ZENG X F , QU Q X , et al. HRSID: a high-resolution SAR images dataset for ship detection and instance segmentation[J]. IEEE Access, 2020, 8, 120234- 120254. doi: 10.1109/ACCESS.2020.3005861
14	ZHAO D P , ZHU C B , QI J , et al. Synergistic attention for ship instance segmentation in SAR images[J]. Remote Sensing, 2021, 13 (21): 4384. doi: 10.3390/rs13214384
15	ZHANG T W , ZHANG X . HTC+ for SAR ship instance segmentation[J]. Remote Sensing, 2022, 14 (10): 2395. doi: 10.3390/rs14102395
16	CHEN K, PANG J M, WANG J Q, et al. Hybrid task cascade for instance segmentation[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2019: 4969-4978.
17	LIN T Y, DOLLAR P, GIRSHICK P, et al. Feature pyramid networks for object detection[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.
18	ZHANG T W , ZHANG X L , SHI J , et al. Depthwise separable convolution neural network for high-speed SAR ship detection[J]. Remote Sensing, 2019, 11 (21): 2483. doi: 10.3390/rs11212483
19	ZHANG T W , ZHANG X L , SHI J , et al. HyperLi-Net: a hyper-light deep learning network for high-accurate and high-speed ship detection from synthetic aperture radar imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 167, 123- 153. doi: 10.1016/j.isprsjprs.2020.05.016
20	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 Trans.on Pattern Analysis and Machine Intelligence, 2017, 40 (4): 834- 848.
21	SHAO Z K, ZHANG T W, ZHANG X L. GAN with ASPP for SAR image to optical image conversion[C]//Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2022: 3355-3358.
22	ZHANG T W , ZHANG X . Squeeze-and-excitation Laplacian pyramid network with dual-polarization feature fusion for ship classification in SAR images[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 19, 4019905.
23	WANG Z , WANG B H , XU N , et al. SAR ship detection in complex background based on multi-feature fusion and non-local channel attention mechanism[J]. International Journal of Remote Sensing, 2021, 42 (19): 7519- 7550. doi: 10.1080/01431161.2021.1963003
24	ZHANG T W , ZHANG X L , KE X . Quad-FPN: a novel quad feature pyramid network for SAR ship detection[J]. Remote Sensing, 2021, 13 (14): 2771. doi: 10.3390/rs13142771
25	WANG X L, GIRSHICK R, GUPTA A, et al. Non-local neural networks[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7794-7803.
26	ZHANG T W , ZHANG X L , LI J W , et al. SAR ship detection dataset (SSDD): official release and comprehensive data analysis[J]. Remote Sensing, 2021, 13 (18): 3690. doi: 10.3390/rs13183690
27	KRIZHEVSKY A , SUTSKEVER I , HINTON G E . Imagenet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60 (6): 84- 90. doi: 10.1145/3065386
28	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
29	ZHANG T W , ZHANG X . A polarization fusion network with geometric feature embedding for SAR ship classification[J]. Pattern Recognition, 2022, 123, 108365. doi: 10.1016/j.patcog.2021.108365
30	ZHANG T W , ZHANG X . Injection of traditional hand-crafted features into modern CNN-based models for SAR ship classification: what, why, where, and how[J]. Remote Sensing, 2021, 13 (11): 2091. doi: 10.3390/rs13112091
31	ZHANG T W, ZHANG X, SHI J, et al. High-speed ship detection in SAR images by improved yolov3[C]//Proc. of the 16th International Computer Conference on Wavelet Active Media Technology and Information Processing, 2019: 149-152.
32	ZHANG T W , ZHANG X . ShipDeNet-20: an only 20 convolution layers and < 1-MB lightweight SAR ship detector[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 18 (7): 1234- 1238.
33	LIN T Y, MAIRE M, BELONGIE S, et al. Microsoft COCO: common objects in context[C]//Proc. of the European Confe-rence on Computer Vision, 2014: 740-755.
34	张天文, 张晓玲, 邵子康, 等. 全等级上下文压缩激励的SAR舰船实例分割[J]. 电子科技大学学报, 2023, 52 (3): 357- 365.
	ZHANG T W , ZHANG X L , SHAO Z K , et al. A full-level context squeeze-and-excitation ROI extractor for SAR ship instance segmentation[J]. Journal of University of Electronic Science and Technology of China, 2023, 52 (3): 357- 365.
35	HE K M, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]//Proc. of the IEEE International Conference on Computer Vision, 2017: 2961-2969.
36	HUANG Z J, HUANG L C, GONG Y C, et al. Mask scoring R-CNN[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2019: 6409-6418.
37	CAI Z W, VASCONCELOS N. Cascade R-CNN: delving into high quality object detection[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 6154-6162.
38	LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.
39	BOLYA D, ZHOU C, XIAO F Y, et al. YOLACT: realtime instance segmentation[C]//Proc. of the IEEE International Conference on Computer Vision, 2019: 9156-9165.
40	ROSSI L, KARIMI A, PRATI A. A novel region of interest extraction layer for instance segmentation[C]//Proc. of the 25th International Conference on Pattern Recognition, 2021: 2203-2209.
41	SU H , WEI S J , LIU S , et al. HQ-ISNet: high-quality instance segmentation for remote sensing imagery[J]. Remote Sensing, 2020, 12 (6): 989. doi: 10.3390/rs12060989
42	ZHAO D P , ZHU C B , QI J , et al. Synergistic attention for ship instance segmentation in SAR images[J]. Remote Sensing, 2021, 13 (21): 4384.

项目	信息
传感器	RadarSat-2、TerraSAR-X、Sentinel-1
SAR图像样本数量	1 160
图像平均尺寸	500×500
训练测试比例	8∶2
极化	HH、VV、VH、HV
分辨率/m	1~15
地点	中国烟台、印度维萨卡帕特南
海况	良好、较差
场景	靠岸、离岸
舰船数量	2 587
最小尺寸舰船/像素	66
最大尺寸舰船/像素	78 597

AP₅₀	阈值为0.5时, 实例分割精度
AP₇₅	阈值为0.75时, 实例分割精度
AP_S	阈值为0.5时, 尺寸小于32像素×32像素目标的精度
AP_M	阈值为0.5时, 尺寸大于32像素×32像素小于96像素×96像素目标精度
AP_L	阈值为0.5时, 尺寸大于96像素×96像素数目标的精度
AP	当阈值∈[0.50∶0.05∶0.95]时, AP的平均值

方法	AP₅₀	AP₇₅	AP_S	AP_M	AP_L	AP
Mask R-CNN	88.5	72.1	57.2	60.8	27.4	57.8
Mask Scoring R-CNN	89.4	73.2	58.0	61.4	22.6	58.6
Cascade Mask R-CNN	87.5	70.5	56.3	58.8	22.6	56.6
HTC	91.7	73.1	58.7	61.6	34.8	59.3
PANet	91.1	74.0	59.3	61.0	52.1	59.6
YOLACT	88.0	52.1	47.3	53.5	40.2	48.4
GRoIE	89.8	72.7	58.6	58.7	21.8	58.3
HQ-ISNet-w18	89.3	73.6	58.2	60.4	37.2	58.6
HQ-ISNet-w32	90.4	75.5	58.9	61.1	37.3	59.3
HQ-ISNet-w40	86.0	72.6	56.7	61.3	50.2	57.6
SA R-CNN	90.4	73.3	59.6	60.3	20.2	59.4
MAI-Net	92.1	76.1	60.6	62.4	55.2	61.1

改进模块			AP₅₀	AP₇₅	AP_S	AP_M	AP_L	AP
ASPP	NLB	CSAB	AP₅₀	AP₇₅	AP_S	AP_M	AP_L	AP
✘	✘	✘	91.7	73.1	58.7	61.6	34.8	59.3
$ \checkmark$	✘	✘	92.0	73.9	59.3	62.1	60.2	59.9
$ \checkmark$	$ \checkmark$	✘	92.2	75.4	60.2	63.2	40.1	60.8
$ \checkmark$	$ \checkmark$	$ \checkmark$	92.1	76.1	60.6	62.4	55.2	61.1

[1]	方小宇, 黄丽佳. 基于全局位置信息和残差特征融合的SAR船舶检测算法[J]. 系统工程与电子技术, 2024, 46(3): 839-848.
[2]	张亚丽, 冯伟, 全英汇, 邢孟道. 基于多源遥感图像多级协同融合的舰船识别算法[J]. 系统工程与电子技术, 2024, 46(2): 407-418.
[3]	刘燊文, 崔兴超, 陈思伟. 结合空时上下文信息的视频SAR图像相干斑滤波[J]. 系统工程与电子技术, 2024, 46(2): 446-458.
[4]	陈洋, 肖国尧, 全英汇, 任爱锋, 别博文, 邢孟道. 基于多核DSP的星载双基FMCW SAR成像算法实现[J]. 系统工程与电子技术, 2024, 46(1): 121-129.
[5]	施端阳, 林强, 胡冰, 杜小帅. 基于YOLO的航管一次雷达目标检测方法[J]. 系统工程与电子技术, 2024, 46(1): 143-151.
[6]	刘鸣谦, 徐仲秋, 陈天成, 张冰尘, 吴一戎. 基于L₁ & TV正则化的低过采样Staggered SAR成像方法[J]. 系统工程与电子技术, 2023, 45(9): 2718-2726.
[7]	王中宝, 尹奎英. 基于联合域滤波的无人机载SAR图像块效应抑制方法[J]. 系统工程与电子技术, 2023, 45(9): 2768-2776.
[8]	汪萌, 诸兵. 不确定性建模在2D和3D目标检测中的应用[J]. 系统工程与电子技术, 2023, 45(8): 2370-2376.
[9]	寇鹏, 刘永祥, 张弛, 李玮杰, 张双辉, 霍凯. 序列ISAR像复杂结构航天器在轨姿态估计[J]. 系统工程与电子技术, 2023, 45(8): 2438-2445.
[10]	王宁, 贺鹏超, 卢景月, 刘曦. 基于DOA估计的前视多通道SAR成像方法[J]. 系统工程与电子技术, 2023, 45(8): 2471-2478.
[11]	邵凯, 杜自群, 王光宇. 基于模型剪枝动态调整压缩率的CSI反馈方法[J]. 系统工程与电子技术, 2023, 45(8): 2615-2622.
[12]	王鹏飞, 詹珩艺, 孙洪忠. 双基前视雷达二维空变补偿频域成像方法[J]. 系统工程与电子技术, 2023, 45(7): 1990-2001.
[13]	朱瀚神, 胡文华, 郭宝锋, 焦丽婷, 朱晓秀, 朱常安. 双基地ISAR稀疏孔径机动目标MTRC补偿成像算法[J]. 系统工程与电子技术, 2023, 45(7): 2022-2030.
[14]	崔天舒, 王栋, 黄振. 面向空间认知通信的轻量化网络自动调制分类方法[J]. 系统工程与电子技术, 2023, 45(7): 2220-2226.
[15]	汪俊澎, 邢世其, 李永祯, 黄大通, 宋少秋. 基于时频交叉乘积的调频连续波SAR干扰方法研究[J]. 系统工程与电子技术, 2023, 45(6): 1651-1657.