基于改进DETR模型的SAR图像舰船检测方法

doi:10.3778/j.issn.1002-8331.2402-0204

摘要/Abstract

摘要： 针对DETR（detection transformer）结构在SAR（synthetic aperture radar）图像舰船检测中存在收敛速度慢、弱小目标易漏检等问题，提出了一种基于多尺度注意力模块和标签对抗训练的改进DETR结构。利用骨干网络提取多尺度特征图，构建多尺度注意力权重矩阵和特征图区域采样方式，有效提高了DETR结构的检测能力；在解码器结构中设计了标签对抗训练模块以解决对目标的重复预测，抑制目标预选框的混淆，并大幅加快模型的收敛速度。在公开的HRSID数据集上进行了算法测试，结果表明，训练36批次的mAP_0.5、mAP_0.5：0.95、AR分别为0.921、0.696、0.753，相较于Faster R-CNN分别提升了12.4%、13.3%和14.7%，验证了该方法在提升目标检测精度和收敛速度方面的优越性。

关键词: 合成孔径雷达, 目标检测, DETR, 多尺度注意力

Abstract: Aiming at the slow convergence of DETR (detection transformer) in ship detection of SAR (synthetic aperture radar) images and easy omission of small and dim targets, an improved DETR based on multi-scale attention module and tag countermeasure training is proposed in this paper. The backbone network is used to extract the multi-scale feature map, construct the multi-scale attention weight matrix and feature map region sampling mode, and effectively improve the detection capability of DETR. A label antagonism training module is designed in the decoder structure to solve the repeated prediction of the target, suppress the confusion of the target preselection frame, and greatly accelerate the convergence of the model. The algorithm is tested on the HRSID dataset. The results show that mAP_0.5, mAP_0.5:0.95 and AR of 36 batches of training are 0.921, 0.696 and 0.753, respectively, which are improved by 12.4%, 13.3% and 14.7% compared with Faster R-CNN. The superiority of the proposed method in improving target detection accuracy and convergence speed is verified.

Key words: synthetic aperture radar, target detection, DETR, multi-scale attention

秦伟伟, 迟昊, 朱晓菲, 秦庆强, 梁卓. 基于改进DETR模型的SAR图像舰船检测方法[J]. 计算机工程与应用, 2025, 61(10): 185-191.

QIN Weiwei, CHI Hao, ZHU Xiaofei, QIN Qingqiang, LIANG Zhuo. Ship Detection Method for SAR Images Based on Improved DETR Model[J]. Computer Engineering and Applications, 2025, 61(10): 185-191.

参考文献

[1] CUI Y, ZHOU G Y, YANG J, et al. On the iterative censoring for target detection in SAR images[J]. IEEE Geoscience and Remote Sensing Letters, 2011, 8(4): 641-645.
[2] DAI H, DU L, WANG Y, et al. A modified CFAR algorithm based on object proposals for ship target detection in SAR images[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(12): 1925-1929.
[3] PAPPAS O, ACHIM A, BULL D. Superpixel-level CFAR detectors for ship detection in SAR imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(9): 1397-1401.
[4] GAO F, MA F, WANG J, et al. Visual saliency modeling for river detection in high-resolution SAR imagery[J]. IEEE Access, 2017, 6: 1000-1014.
[5] ZHAO Y, ZHAO L J, XIONG B L, et al. Attention receptive pyramid network for ship detection in SAR images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 2738-2756.
[6] 刘俊明, 孟卫华. 基于深度学习的单阶段目标检测算法研究综述[J]. 航空兵器, 2020, 27(3): 44-53.
LIU J M, MENG W H. Review on single-stage object detection algorithm based on deep learning[J]. Aero Weaponry, 2020, 27(3): 44-53.
[7] SUN Z Z, DAI M C, LENG X G, et al. An anchor-free detection method for ship targets in high-resolution SAR images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 7799-7816.
[8] ZHOU Y, JIANG X, XU G Z, et al. PVT-SAR: an arbitrarily oriented SAR ship detector with pyramid vision transformer[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16: 291-305.
[9] XU X W, ZHANG X L, ZHANG T W. Lite-YOLOv5: a lightweight deep learning detector for on-board ship detection in large-scene sentinel-1 SAR images[J]. Remote Sensing, 2022, 14(4): 1018.
[10] 宋泰年, 秦伟伟, 梁卓, 等. 面向轻量化网络的改进双通道注意力机制图像分类方法[J]. 航空兵器, 2021, 28(5): 81-85.
SONG T N, QIN W W, LIANG Z, et al. Improved dual-channel attention mechanism image classification method for lightweight network[J]. Aero Weaponry, 2021, 28(5): 81-85.
[11] 谭显东, 彭辉. 改进YOLOv5的SAR图像舰船目标检测[J]. 计算机工程与应用, 2022, 58(4): 247-254.
TAN X D, PENG H. Improved YOLOv5 ship target detection in SAR image[J]. Computer Engineering and Applications, 2022, 58(4): 247-254.
[12] 秦伟伟, 宋泰年, 刘洁瑜, 等. 基于轻量化YOLOv3的遥感军事目标检测算法[J]. 计算机工程与应用, 2021, 57(21): 263-269.
QIN W W, SONG T N, LIU J Y, et al. Remote sensing military target detection algorithm based on lightweight YOLOv3[J]. Computer Engineering and Applications, 2021, 57(21): 263-269.
[13] ZHENG Q H, TIAN X Y, YU Z G, et al. MobileRaT: a lightweight radio transformer method for automatic modulation classification in drone communication systems[J]. Drones, 2023, 7(10): 596.
[14] ZHENG Q H, SAPONARA S, TIAN X Y, et al. A real-time constellation image classification method of wireless communication signals based on the lightweight network MobileViT[J]. Cognitive Neurodynamics, 2024, 18(2): 659-671.
[15] CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]//Proceedings of the 16th European Conference on Computer Vision. Cham: Springer, 2020: 213-229.
[16] NEUBECK A, VAN GOOL L. Efficient non-maximum suppression[C]//Proceedings of the 18th International Conference on Pattern Recognition. Piscataway: IEEE, 2006: 850-855.
[17] 李柯泉, 陈燕, 刘佳晨, 等. 基于深度学习的目标检测算法综述[J]. 计算机工程, 2022, 48(7): 1-12.
LI K Q, CHEN Y, LIU J C, et al. Survey of deep learning-based object detection algorithms[J]. Computer Engineering, 2022, 48(7): 1-12.
[18] LI F, ZHANG H, LIU S L, et al. DN-DETR: accelerate DETR training by introducing query denoising[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(4): 2239-2251.
[19] LIU S L, LI F, ZHANG H, et al. DAB-DETR: dynamic anchor boxes are better queries for DETR[C]//Proceedings of the 2021 International Conference on Learning Representations, 2021.
[20] ZHANG H, LI F, LIU S L, et al. DINO: DETR with improved denoising anchor boxes for end-to-end object detection[J]. arXiv:2203.03605, 2022.
[21] WEI S J, ZENG X F, QU Q Z, et al. HRSID: a high-resolution SAR images dataset for ship detection and instance segmentation[J]. IEEE Access, 2020, 8: 120234-120254.
[22] TIAN Z, SHEN C H, CHEN H, et al. FCOS: fully convolutional one-stage object detection[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2019: 9627-9636.
[23] LAW H, DENG J. CornerNet: detecting objects as paired keypoints[C]//Proceedings of the 15th European Conference on Computer Vision. Cham: Springer, 2018: 765-781.
[24] REN S Q, HE K M, 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.
[25] LI F, ZENG A L, LIU S L, et al. Lite DETR: an interleaved multi-scale encoder for efficient DETR[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2023: 18558-18567.
[26] ZHENG D H, DONG W H, HU H L, et al. Less is more: focus attention for efficient DETR[C]//Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2023: 6651-6660.