基于改进YOLOv7-tiny的红外船舶目标检测

doi:10.19678/j.issn.1000-3428.0069919

摘要/Abstract

摘要： 针对红外场景下的船舶图像检测准确率低和计算量大的问题,提出一种用于红外船舶目标检测的改进YOLOv7-tiny模型。首先,在主干网络采用轻量级模型PP-LCNet,极大降低网络参数量与计算量。然后,改进Fused-MBConv模块和坐标注意力(CA)机制构建ELAN-FM-C模块,将其引入特征融合层,全面关注特征层的空间信息和通道信息,获取更大感受野。接着,使用基于最小点距离的边界框相似度比较的MDPIoU损失函数,简化了计算过程,提高了轻量级模型对红外目标的检测能力。然后,设计R-BiFPN结构来融合更多有效特征,提高了轻量级模型对不同尺度目标的检测效果。最后,利用知识蒸馏技术进一步提高了模型的检测精度。在艾睿光电红外海上船舶数据集上的验证结果表明,相比原始YOLOv7-tiny模型,改进模型检测的均值平均精度(mAP)提高了3.3百分点、参数量和计算量分别降低了23.0%和30.3%、模型大小减小了21.7%。在公开船舶数据集SeaShips和Ship Images上的验证结果表明,与主流和最新检测模型相比,改进模型具有良好的泛化性和鲁棒性,并且在检测精度和轻量化方面表现更优。

关键词: 船舶目标检测, 轻量级, 知识蒸馏, 注意力机制, YOLOv7-tiny网络

Abstract: An improved YOLOv7-tiny-based lightweight infrared ship target detection model is proposed to address the issues of low accuracy and high computation load of ship image detection in the infrared range. First, the lightweight model PP-LCNet is employed in the backbone network, which significantly reduces both the number of parameters and computational requirements. Second, an improved Fused-MBConv module and a Coordinate Attention (CA) mechanism are incorporated to construct the ELAN-FM-C module, which is then integrated into the feature fusion layer to comprehensively focus on the spatial and channel information of the feature layer to obtain a large receptive field. Subsequently, the Minimum Distance Points Intersection over Union (MDPIoU) loss function, which compares the bounding box similarity based on the minimum point distance, is adopted to simplify the computation process and improve the detection capability of the lightweight model for infrared targets. Based on this, an R-BiFPN structure is proposed to fuse more effective features, thereby improving the detection performance of the lightweight model across targets of different scales. Finally, a knowledge distillation technique is used to further improve the detection accuracy of the model. The improved model is validated on the Iray Optoelectronics infrared offshore ship dataset, achieving a mean Average Precision (mAP) that is 3.3 percentage points higher than that obtained using the original YOLOv7-tiny model. Simultaneously, the parameter and computational complexities are reduced by 23.0% and 30.3%, respectively, and the model size is reduced by 21.7%. Experiments on publicly available ship datasets, namely SeaShips and Ship Images, reveal that, compared to other mainstream and latest detection models, the improved model demonstrates excellent generalization and robustness and outperforms other models in terms of both detection accuracy and lightweight design.

Key words: ship target detection, lightweight, knowledge distillation, attention mechanism, YOLOv7-tiny network

中图分类号:

TP391

许晓阳, 魏伟, 高重阳. 基于改进YOLOv7-tiny的红外船舶目标检测[J]. 计算机工程, 2026, 52(2): 209-220.

XU Xiaoyang, WEI Wei, GAO Chongyang. Infrared Ship Target Detection Based on Improved YOLOv7-tiny[J]. Computer Engineering, 2026, 52(2): 209-220.

https://www.ecice06.com/CN/Y2026/V52/I2/209

参考文献

[1] 王炳德, 杨柳涛. 基于YOLOv3的船舶目标检测算法[J]. 中国航海, 2020, 43(1): 67-72. WANG B D, YANG L T. Ship target detection algorithm based on YOLOv3[J]. Navigation of China, 2020, 43(1): 67-72. (in Chinese)
[2] LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2): 91-110.
[3] NOBLE W S. What is a support vector machine?[J]. Nature Biotechnology, 2006, 24(12): 1565-1567.
[4] 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.
[5] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2016: 779-788.
[6] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[M]. Berlin, Germany: Springer International Publishing, 2016.
[7] 刘皓皎, 刘力双, 张明淳. 基于YOLOv5改进的红外目标检测算法[J]. 激光技术, 2024, 48(4): 534-541. LIU H J, LIU L S, ZHANG M C. An improved infrared object detection algorithm based on YOLOv5[J]. Laser Technology, 2024, 48(4): 534-541. (in Chinese)
[8] 陈永麟, 王恒涛, 张上. 基于YOLO v7的轻量级红外目标检测算法[J]. 红外技术, 2024, 46(12): 1380-1389. CHEN Y L, WANG H T, ZHANG S. Lightweight infrared target detection algorithm based on YOLO v7[J]. Infrared Technology, 2024, 46(12): 1380-1389. (in Chinese)
[9] 张莉莉, 王修晖. 基于FMF-YOLOv5的光伏组件红外图像故障诊断[J]. 计算机工程与应用, 2025, 61(2): 327-334. ZHANG L L, WANG X H. Infrared image fault diagnosis of photovoltaic modules based on FMF-YOLOv5[J]. Computer Engineering and Applications, 2025, 61(2): 327-334. (in Chinese)
[10] 袁亚剑, 毛力. 一种增强前景的轻量级交通标志检测模型[J]. 计算机工程, 2025, 51(3): 54-63. YUAN Y J, MAO L. A lightweight traffic sign detection model with enhanced foregrounds[J]. Computer Engineering, 2025, 51(3): 54-63. (in Chinese)
[11] 张上, 黄俊锋, 王恒涛, 等. 低空轻量级红外弱小目标检测算法[J]. 激光与红外, 2024, 54(1): 122-129. ZHANG S, HUANG J F, WANG H T, et al. Low altitude lightweight infrared weak small target detection algorithm[J]. Laser & Infrared, 2024, 54(1): 122-129. (in Chinese)
[12] 常凯旋, 黄建华, 孙希延, 等. 基于双模态图像融合的无人机光学小目标检测算法[J]. 激光与光电子学进展, 2025, 62(4): 0428001. CHANG K X, HUANG J H, SUN X Y, et al. Optical small target detection method by drone based on dual-modal image fusion[J]. Laser & Optoelectronics Progress, 2025, 62(4): 0428001. (in Chinese)
[13] 李琳, 靳志鑫, 俞晓磊, 等. Haar小波下采样优化YOLOv9的道路车辆和行人检测[J]. 计算机工程与应用, 2024, 60(20): 207-214. LI L, JIN Z X, YU X L, et al. Road vehicle and pedestrian detection based on YOLOv9 for Haar wavelet downsampling[J]. Computer Engineering and Applications, 2024, 60(20): 207-214. (in Chinese)
[14] WANG C Y, YEH I H, LIAO H Y M. YOLOv9: learning what you want to learn using programmable gradient information[EB/OL].[2024-04-05]. https://arxiv.org/abs/2402.13616.
[15] WANG C Y, BOCHKOVSKIY A, LIAO H M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2023: 7464-7475.
[16] CUI C, GAO T, WEI S, et al. PP-LCNet: a lightweight CPU convolutional neural network[EB/OL].[2024-04-05]. https://arxiv.org/pdf/2109.15099.
[17] TAN M, LE Q. EfficientNetV2: smaller models and faster training[C]//Proceedings of the International Conference on Machine Learning.[S. l.]: PMLR, 2021: 10096-10106.
[18] SILIANG M, YONG X. MPDIoU: a loss for efficient and accurate bounding box regression[EB/OL].[2024-04-05]. https://arxiv.org/abs/2307.07662.
[19] ELFWING S, UCHIBE E, DOYA K. Sigmoid-weighted linear units for neural network function approximation in reinforcement learning[J]. Neural Networks, 2018, 107: 3-11.
[20] CHEN G, CHOI W, YU X, et al. Learning efficient object detection models with knowledge distillation[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2017: 742-751.
[21] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA: IEEE Press, 2018: 7132-7141.
[22] HOWARD A G, ZHU M, CHEN B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[EB/OL].[2024-04-05]. https://arxiv.org/abs/1704.04861.
[23] HOWARD A, SANDLER M, CHEN B, et al. Searching for MobileNetV3[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). Washington D.C, USA: IEEE Press, 2019: 1314-1324.
[24] TAN M, LE Q. EfficientNet: rethinking model scaling for convolutional neural networks[C]//Proceedings of the International Conference on Machine Learning.[S. l.]: PMLR, 2019: 6105-6114.
[25] ZHENG Z H, WANG P, LIU W, et al. Distance-IoU loss: faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2020: 12993-13000.
[26] HE J, ERFANI S, MA X, et al. Alpha-IoU: a family of power intersection over union losses for bounding box regression[J]. Advances in Neural Information Processing Systems, 2021, 34: 20230-20242.
[27] TAN M X, PANG R M, LE Q V. EfficientDet: scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D.C., USA: IEEE Press, 2020: 10778-10787.
[28] WANG A, CHEN H, LIU L, et al. YOLOv10: real-time end-to-end object detection[EB/OL].[2024-04-05]. https://arxiv.org/abs/2405.14458.

选择文件类型/文献管理软件名称

选择包含的内容