Abstract
Ship detection is a key research task for ship identification, monitoring and management to ensure the safety of shipping lanes and harbour. Object detection methods based on computer vision and deep learning has the potential for real-time ship detection, but has the challenges of accuracy, real-time, and lack of high-quality ship datasets. This paper proposes an object detection network named YOLOv4-MobileNetV2-DW-LCARM for ship detection, which is a hybrid application of YOLOv4, MobileNetV2, Depthwise separable convolution and a proposed Lightweight Channel Attention Residual Module (LCARM). To verify the effectiveness of the network, we built a ship dataset with 10 categories and 20216 samples for model training, ablation experiments, and comparative experiments. Compared with YOLOv4, the results show that our network reduces the number of parameters by 82.58% and improves the forward inference speed by 53.6%, reaching the accuracy of 62.75% ArP@0.75 and 94% AP@0.5. The proposed network can be deployed on edge devices for real-time ship detection because it has 24.82 FPS of processing speed. The proposed dataset construction methods also contribute to similar object detection tasks.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References s
Lecun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436 (2015)
Tang, J., Deng, C., et al.: Compressed-domain ship detection on spaceborne optical image using deep neural network and extreme learning machine. IEEE Trans. Geosci. Remote Sens. 53(3), 1174–1185 (2015)
Wang, X., Chen, C.: Ship detection for complex background SAR images based on a multiscale variance weighted image entropy method. IEEE Geosci. Remote Sens. Lett. 14(2), 184–187 (2017)
Yang, X., Sun, H., Fu, K., et al.: Automatic ship detection of remote sensing images from Google earth in complex scenes based on multi-scale rotation dense feature pyramid networks. Remote Sens. 10(1), 132 (2018)
Li, Q., Mou, L., Liu, Q., et al.: HSF-Net: multiscale deep feature embedding for ship detection in optical remote sensing imagery. IEEE Trans. Geosci. Remote Sens. 56(12), 7147–7161 (2018)
Yuan, Y., Jiang, Z., Zhang, H., et al.: Ship detection in optical remote sensing images based on deep convolutional neural networks. J. Appl. Remote Sens. 11(4), 1 (2017)
Li, J., Qu, C., Shao, J.: Ship detection in SAR images based on an improved faster R-CNN. In: SAR in Big Data Era: Models, Methods & Applications. IEEE (2017)
Lee, W.-J., Roh, M.-I., Lee, H.-W., et al.: Detection and tracking for the awareness of surroundings of a ship based on deep learning. J. Comput. Des. Eng. 8(5), 1407–1430 (2021)
Li, H., Deng, L., Yang, C., et al.: Enhanced YOLOv3 tiny network for real-time ship detection from visual image. IEEE Access 9, 16692–16706 (2021)
Ma, Z.: High-speed lightweight ship detection algorithm based on YOLO-v4 for three-channels RGB SAR image. Remote Sens. 13, 1909 (2021)
Bochkovskiy, A., Wang, C.Y., Liao, H.: YOLOv4: optimal speed and accuracy of object detection (2020)
Wang, C.Y., Liao, H., Wu, Y.H., et al.: CSPNet: a new backbone that can enhance learning capability of CNN. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). IEEE (2020)
He, K., Ren, S., et al.: Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Trans. Pattern Anal. Mach. Intell. 37(9), 1904–1916 (2015)
Liu, S., et al.: Path aggregation network for instance segmentation. IEEE (2018)
Zhang, X., Zhou, X., Lin, M., et al.: ShuffleNet: an extremely efficient convolutional neural network for mobile devices (2017)
Iandola, F.N., Han, S., Moskewicz, M.W., et al.: SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size (2016)
Howard, A.G., Zhu, M., Chen, B., et al.: MobileNets: efficient convolutional neural networks for mobile vision applications (2017)
Chollet, F.: Xception: deep learning with depthwise separable convolutions. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE (2017)
Szegedy, C., Vanhoucke, V., Ioffe, S., et al.: Rethinking the inception architecture for computer vision, pp. 2818–2826. IEEE (2016)
Sandler, M., Howard, A., Zhu, M., et al.: MobileNetV2: inverted residuals and linear bottlenecks. In: Conference on Computer Vision and Pattern Recognition. IEEE (2018)
Jie, H., Li, S., Gang, S., et al.: Squeeze-and-excitation networks. IEEE Trans. Pattern Anal. Mach. Intell. (2017)
Veit, A., Matera, T., Neumann, L., et al.: COCO-text: dataset and benchmark for text detection and recognition in natural images (2016)
Everingham, M., Eslami, S., Gool, L.V., et al.: The Pascal visual object classes challenge: a retrospective. Int. J. Comput. Vis. 111(1), 98–136 (2015)
Acknowledgments
This research was supported by the National Key R&D Program of China under Grant No. 2020YFB1707700.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Xie, P., Tao, R., Luo, X., Shi, Y. (2022). YOLOv4-MobileNetV2-DW-LCARM: A Real-Time Ship Detection Network. In: Uden, L., Ting, IH., Feldmann, B. (eds) Knowledge Management in Organisations. KMO 2022. Communications in Computer and Information Science, vol 1593. Springer, Cham. https://doi.org/10.1007/978-3-031-07920-7_22
Download citation
DOI: https://doi.org/10.1007/978-3-031-07920-7_22
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-07919-1
Online ISBN: 978-3-031-07920-7
eBook Packages: Computer ScienceComputer Science (R0)