改进RTMDet的SAR舰船检测算法

doi:10.3778/j.issn.1002-8331.2307-0175

摘要/Abstract

摘要： 针对合成孔径雷达（synthetic aperture radar，SAR）图像中小目标舰船和复杂背景下舰船检测精度低的问题，提出一种改进RTMDet（real-time models for object detection）的SAR舰船检测算法。优化主干网络结构中的基本构建单元，并引入全局注意力机制SimAM（simple，parameter-free attention module），在不增加额外参数的情况下提高模型对关键特征信息的提取能力；为了在特征融合过程中减少小目标特征信息流失和增加其在浅层特征图中的融合占比，构建新的轻量级特征融合模块SPD-RPAFPN（space to depth reverse path aggregation feature pyramid network）；在预测模块中将回归损失函数替换为KFIoU（Kalman filter based intersection over union），提高模型对小目标舰船的检测能力。在公开数据集RSDD上进行了实验对比，改进后的算法相比RTMDet，模型参数量和计算量分别下降4.5%和10.8%，同时近岸AP提高14.6个百分点，总AP提高2.7个百分点，达到90.7%。与目前的主流算法对比，SAR舰船检测精度也明显提升，证明了改进RTMDet算法的有效性。

关键词: SAR图像, 舰船检测, RTMDet, SimAM, SPD-RPAFPN, KFIoU

Abstract: A synthetic aperture radar (SAR) ship detection algorithm with improved RTMDet (real-time models for object detection) is proposed to address the problem of low detection accuracy in small target ships and complex backgrounds in SAR images. Firstly, the basic building blocks in backbone network structure are optimized, and the global attention mechanism SimAM (simple, parameter-free attention module) is introduced, which improves the ability of the model to extract key feature information without adding additional parameters. In order to reduce the loss of small target feature information and increase its share in shallow feature map during feature fusion, a new lightweight feature fusion module SPD-RPAFPN (space to depth reverse path aggregation feature pyramid network) is constructed. Finally, the regression loss function is replaced with KFIoU (Kalman filter based intersection over union) in the prediction module to improve the detection capability of the model for small target ships. Experimental comparisons are conducted on the publicly available dataset RSDD. Compared with RTMDet, the improved model improves the inshore AP value by 14.6 percentage points and the total AP value by 2.7 percentage points to 90.7%, while the number of model parameters and computational effort are decreased by 4.5% and 10.8%. Compared with the current mainstream algorithm, the SAR ship detection accuracy is also significantly improved, which proves the effectiveness of the improved RTMDet algorithm.

Key words: SAR images, ship detection, RTMDet, SimAM, SPD-RPAFPN, KFIoU

张玉宁, 贾渊, 陈越. 改进RTMDet的SAR舰船检测算法[J]. 计算机工程与应用, 2024, 60(22): 314-322.

ZHANG Yuning, JIA Yuan, CHEN Yue. Improved RTMDet for SAR Ship Detection[J]. Computer Engineering and Applications, 2024, 60(22): 314-322.

参考文献

[1] MOREIRA A, PRATS-IRAOLA P, YOUNIS M, et al. A tutorial on synthetic aperture radar[J]. IEEE Geoscience and Remote Sensing Magazine, 2013, 1(1): 6-43.
[2] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision, 2015: 1440-1448.
[3] REN S Q, HE K, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.
[4] HAN J M, DING J, XUE N, et al. REDet: a rotation-equivariant detector for aerial object detection[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 2786-2795.
[5] XIE X X, CHENG G, WANG J B, et al. Oriented R-CNN for object detection[C]//Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision, 2021: 3520-3529.
[6] LIN T, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2): 318-327.
[7] YANG X, YAN J C, FENG Z M, et al. R3Det: refined single-stage detector with feature refinement for rotating object[C]//Proceedings of the 2021 AAAI Conference on Artificial Intelligence, 2021: 3163-3171.
[8] HAN J M, DING J, LI J, et al. Align deep features for oriented object detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-11.
[9] GLENN J. YOLOv5 releases v6.1[EB/OL]. (2022-02-07)[2023-05-25]. https://github.com/ultralytics/yolov5/releases/tag/v6.1.
[10] GE Z, LIU S T, WANG F, et al. YOLOX: exceeding YOLO series in 2021[J]. arXiv:2107.08430, 2021.
[11] WANG C, BOCHKOVSKIY A, LIAO M H. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[J]. arXiv:2207.02696, 2022.
[12] LYU C, ZHANG W W, HUANG H A, et al. RTMDet: an empirical study of designing real-time object detectors[J]. arXiv:2212.07784, 2022.
[13] 赵鹏飞, 谢林柏, 彭力. 融合注意力机制的深层次小目标检测算法[J]. 计算机科学与探索, 2022, 16(4): 927-937.
ZHAO P F, XIE L B, PENG L. Deep small object detection algorithm integrating attention mechanism[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(4): 927-937.
[14] ZHAO H Y, ZHANG H P, ZHAO Y Y. YOLOv7-sea: object detection of maritime UAV images based on improved YOLOv7[C]//Proceedings of the 2023 IEEE/CVF Winter Conference on Applications of Computer Vision, 2023: 233-238.
[15] 刘涛, 丁雪妍, 张冰冰, 等. 改进YOLOv5的遥感图像检测方法[J]. 计算机工程与应用, 2023, 59(10): 253-261.
LIU T, DING X Y, ZHANG B B, et al. Improved YOLOv5 for remote sensing image detection[J]. Computer Engineering and Applications, 2023, 59(10): 253-261.
[16] LI X J, CHEN L, WANG D Y, et al. Detection of ship targets in remote sensing image based on improved YOLOv5[C]//Proceedings of the 2022 International Conference on Electronic Information Engineering and Computer Science, 2022.
[17] CHOLLET F. Xception: deep learning with depthwise separable convolutions[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, 2017: 1251-1258.
[18] LIN T, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.
[19] LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, 2018: 8759-8768.
[20] GHIASI G, LIN T, LE Q V. NAS-FPN: learning scalable feature pyramid architecture for object detection[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 7036-7045.
[21] YANG L X, ZHANG R, LI L D, et al. Simam: a simple, parameter-free attention module for convolutional neural- networks[C]//Proceedings of the 2021 International Conference on Machine Learning, 2021: 11863-11874.
[22] 祁宣豪, 智敏. 图像处理中注意力机制综述[J]. 计算机科学与探索, 2024, 18(2): 345-362.
QI X H, ZHI M. Review of attention mechanisms in image processing[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(2): 345-362.
[23] WOO S, PARK J, LEE J, et al. CBAM: convolutional block attention module[C]//Proceedings of the 2018 European Conference on Computer Vision, 2018: 3-19.
[24] DING X H, ZHANG X Y, HAN J G, et al. Scaling up your kernels to 31x31: revisiting large kernel design in CNNs[C]//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 11963-11975.
[25] WOO S, DEBNATH S, HU R H, et al. ConvNeXt v2: codesigning and scaling convnets with masked autoencoders[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 16133-16142.
[26] SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM: visual explanations from deep networks via gradient-based localization[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision, 2017: 618-626.
[27] SUNKARA R, LUO T. No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects[C]//Proceedings of the 2022 Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 2022: 443-459.
[28] SAJJADI M S M, VEMULAPALLI R, BROWN M. Frame-recurrent video super-resolution[C]//Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, 2018: 6626-6634.
[29] YANG X, YAN J C, MING Q, et al. Rethinking rotated object detection with gaussian wasserstein distance loss[C]//Proceedings of the 2021 International Conference on Machine Learning, 2021: 11830-11841.
[30] YANG X, YANG X J, YANG J R, et al. Learning high-precision bounding box for rotated object detection via Kullback-Leibler divergence[C]//Proceedings of the 2021 Advances in Neural Information Processing Systems, 2021: 18381-18394.
[31] YANG X, ZHOU Y, ZHANG G F, et al. The KFIoU loss for rotated object detection[J]. arXiv:2201.12558, 2022.
[32] 徐从安, 苏航, 李健伟, 等. RSDD-SAR: SAR舰船斜框检测数据集[J]. 雷达学报, 2022, 11(4): 581-599.
XU C A, SU H, LI J W, et al. RSDD-SAR: rotated ship detection dataset in SAR images[J]. Journal of Radars, 2022, 11(4): 581-599.