视觉目标跟踪在各种海事应用中发挥着重要作用。然而，现有的跟踪方法大多属于生成模型，只关注对象的特征，忽略背景信息。因此，对目标的视觉显著性有更高的要求。本文将深度学习方法应用于船舶跟踪，提出使用孪生网络和区域推荐网络的海上船舶跟踪方法。为进一步提高跟踪性能，参照AlexNet网络对孪生网络的CNN模块进行修改，并提出一种基于历史轨迹的自适应搜索区域提取方法，以适应不同的运动场景。利用数据集对所提出的跟踪器进行评估。结果表明，在使用Intel Xeon CPU E5-2620，GTX TITAN的PC机上可以达到58%的平均精度和124.21 FPS。
Visual target tracking plays an important role in various maritime applications. However, most of the existing tracking methods belong to the generation model, focusing only on the characteristics of the target, ignoring the background information. Therefore, there is a higher requirement for the visual salience of the target. This paper applied the deep learning method to ship tracking. This paper proposed a method of tracking ships at sea using a twin network and a regional recommendation network. In order to further improve the tracking performance, we modified AlexNet on the CNN module of the siamese network, and proposed an adaptive search region extraction method based on historical trajectory to adapt to different motion scenarios. Using our data set to evaluate the proposed tracker, the results showed that the tracker achieved an average accuracy of 58% and 124.21 FPS on a PC using Intel Xeon CPU E5-2620, GTX TITAN. Our tracking model has better performance than existing ship tracking algorithms, and we can continue to study more accurate deep learning ship tracking models in the future.
2019,41(12): 103-108 收稿日期：2019-09-30
DOI：10.3404/j.issn.1672-7649.2019.12.021
分类号：U664.82
基金项目：广东重点研发计划资助项目（2018B00108004）；南方海洋科学与工程广东省实验室资助项目
作者简介：张云飞(1984-),男,博士,教授级高级工程师,研究方向为智能科技
参考文献：
[1] 李晓飞. 基于视觉的船舶跟踪与三维定位技术研究与应用[D]. 广州：广东工业大学, 2014.
[2] 闫雨石, 鞠文博. 船舶无人驾驶技术中的数据融合应用技术[J]. 舰船科学技术, 2019(20): 5–7
[3] 张彩虹, 胡彦军. 无人船远距离目标追踪与自动避障算法优化[J]. 舰船科学技术, 2019(16)
[4] ZHANG Y, LI Q Z, ZANG F N. Ship detection for visual maritime surveillance from non-stationary platforms[J]. Ocean Engineering, 2017, 141: 53–63
[5] BLOISI D, IOCCHI L. Argos—a video surveillance system for boat traffic monitoring in venice[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2009, 23(7): 1477–1502
[6] PRASAD D K, RAJAN D, RACHMAWATI L, et al. Video processing from electro-optical sensors for object detection and tracking in a maritime environment: a survey[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(8): 1993–2016
[7] 滕飞, 刘清, 朱琳. 一种快速鲁棒的内河CCTV系统船舶跟踪算法[J]. 武汉理工大学学报, 2014(5): 86–91
[8] KONG X, LIU L, QIAN Y, et al. Automatic detection of sea-sky horizon line and small targets in maritime infrared imagery[J]. Infrared Physics & Technology, 2016, 76: 185–199
[9] HELD D, THRUN S, SAVARESE S. Learning to track at 100 fps with deep regression networks[C]//European Conference on Computer Vision. Springer, Cham, 2016: 749–765.
[10] TANG S, ANDRILUKA M, ANDRES B, et al. Multiple people tracking by lifted multicut and person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 3539–3548.
[11] TELLO M, LÓPEZ-MARTÍNEZ C, MALLORQUI J J. A novel algorithm for ship detection in SAR imagery based on the wavelet transform[J]. IEEE Geoscience and remote sensing letters, 2005, 2(2): 201–205
[12] ZHU C, ZHOU H, WANG R, et al. A novel hierarchical method of ship detection from spaceborne optical image based on shape and texture features[J]. IEEE Transactions on geoscience and remote sensing, 2010, 48(9): 3446–3456
[13] 刘静, 刘以安, 杨新刚. 基于Kalman滤波的船舶跟踪技术[J]. 微计算机信息, 2007, 23(31): 187–189
[14] KÜNZNER N, KUSHAUER J, KATZENBEIßER S, et al. Modern electro-optical imaging system for maritime surveillance applications[C]//2010 International WaterSide Security Conference. IEEE, 2010: 1–4.
[15] HU W C, YANG C Y, HUANG D Y. Robust real-time ship detection and tracking for visual surveillance of cage aquaculture[J]. Journal of Visual Communication and Image Representation, 2011, 22(6): 543–556
[16] TENG F, LIU Q. Robust multi-scale ship tracking via multiple compressed features fusion[J]. Signal Processing: Image Communication, 2015, 31: 76–85
[17] SZPAK Z L, TAPAMO J R. Maritime surveillance: Tracking ships inside a dynamic background using a fast level-set[J]. Expert systems with applications, 2011, 38(6): 6669–6680
[18] FROST D, TAPAMO J R. Detection and tracking of moving objects in a maritime environment using level set with shape priors[J]. EURASIP Journal on Image and Video Processing, 2013, 2013(1): 42
[19] XIN J, DU X, ZHANG J. Deep learning for robust outdoor vehicle visual tracking[C]//2017 IEEE International Conference on Multimedia and Expo (ICME). IEEE, 2017: 613–618.
[20] CAI Y, WANG H, SUN X, et al. Visual Vehicle Tracking Based on Deep Representation and Semisupervised Learning[J]. Journal of Sensors, 2017, 2017
[21] BRUNETTI A, BUONGIORNO D, TROTTA G F, et al. Computer vision and deep learning techniques for pedestrian detection and tracking: A survey[J]. Neurocomputing, 2018, 300: 17–33
[22] BOUSETOUANE F, MORRIS B. Fast CNN surveillance pipeline for fine-grained vessel classification and detection in maritime scenarios[C]//201613th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). IEEE, 2016: 242–248.
[23] GIRSHICK R. Fast r-cnn[C]//Proceedings of the IEEE international conference on computer vision. 2015: 1440–1448.
[24] LECLERC M, THARMARASA R, FLOREA M C, et al. Ship classification using deep learning techniques for maritime target tracking[C]//201821st International Conference on Information Fusion (FUSION). IEEE, 2018: 737-744.
[25] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 1–9.
[26] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770–778.
[27] GEIGER A, LENZ P, STILLER C, et al. Vision meets robotics: The KITTI dataset[J]. The International Journal of Robotics Research, 2013, 32(11): 1231–1237
[28] KRISTAN M, LEONARDIS A, MATAS J, et al. The sixth visual object tracking vot2018 challenge results[C]//Proceedings of the European Conference on Computer Vision (ECCV). 2018: 0–0.
[29] LI B, YAN J, WU W, et al. High performance visual tracking with siamese region proposal network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018: 8971–8980.
