Analysis and optimization of ship trajectory dissimilarity models based on multi-feature fusion
-
摘要: 基于船舶自动识别系统轨迹，构建了船舶轨迹静态相异度模型、动态相异度模型以及组合相异度模型，包括轨迹起点和终点相异度模型、轨迹长度相异度模型、轨迹空间分布相异度模型、轨迹航速均值相异度模型、轨迹航向均值相异度模型、轨迹航速标准差相异度模型和轨迹航向标准差相异度模型；采用KNN分类算法进行轨迹分类，分析了单个相异度模型的有效性和时效性，对比了单个相异度模型和组合相异度模型下轨迹分类效果，研究了组合相异度模型中相异度模型的类别和权重对轨迹分类的影响；分别以内河航道和港口水域船舶轨迹进行试验。试验结果显示：在采用单个相异度的情况下，就分类效果而言，轨迹起点和终点相异度模型和轨迹航向均值相异度模型在内河航道和港口水域船舶轨迹分类效果均优于其他模型，而基于轨迹航速均值相异度模型和轨迹航速标准差相异度模型的轨迹分类效果最低，就分类效率而言，基于航速、航向均值和标准差的相异度模型耗时明显低于其他3个相异度模型；采用组合相异度进行轨迹分类，内河航道和港口水域船舶轨迹分类结果的基于精确率和召回率的宏平均值和微平均值均能接近99%；将组合相异度中相异度类别数由4个增加到7个，轨迹分类评估结果进一步得到提高。因此，单个相异度模型中以轨迹起点和终点相异度模型、轨迹航向均值相异度模型以及轨迹空间分布相异度模型分类效果最优且稳定，而轨迹空间分布相异度模型和轨迹长度相异度模型耗时明显高于其他方式，各相异度模型在不同场景中的适应性基本相似，通过增加组合相异度中相异度类别能够提高轨迹识别效果。Abstract: Based on ship automatic identification system (AIS) trajectories, static dissimilarity models, dynamic dissimilarity models, and a combined dissimilarity model of ship trajectories were constructed, including the following dissimilarity models: trajectory departure and destination, trajectory length, trajectory spatial distribution, trajectory mean speed, trajectory mean course, trajectory speed standard deviation, and trajectory course standard deviation. Trajectories were classified using the KNN classification algorithm, the effectivenesses and efficiencies of each single dissimilarity model were analyzed, the effect of trajectory classification under different unique dissimilarity models and the combined dissimilarity model were compared, and the influence of the categories and weights of dissimilarity models on trajectory classification in the combined dissimilarity model was studied. Experiments were conducted using ship trajectories in inland waterways and port waters. Experimental results show that under the condition of adopting a single dissimilarity, in terms of the classification effect, the ship trajectory classification based on the dissimilarity model of trajectory departure and destination and the dissimilarity model of trajectory mean course is better than that using other dissimilarity models in inland waterways and port waters, whereas the trajectory classification effect based on the dissimilarity model of trajectory mean speed and the dissimilarity model of trajectory speed standard deviation is worse. In terms of classification efficiency, the time consumed by the dissimilarity models based on mean value and standard deviation is significantly lower than that of the other dissimilarity models. Through the analysis and optimization of trajectory dissimilarity models based on the trajectory classification results of the KNN classification algorithm, when the trajectory classification is conducted using the combined dissimilarity model, macro and micro averages based on accuracy and recall of ship trajectory classification results in the inland waterway and port waters can both reach 99%; moreover, by increasing the number of dissimilarity categories in the combined dissimilarity from 4 to 7, the evaluation result of trajectory classification is further improved. Therefore, in the single dissimilarity model, the classification effects of the dissimilarity model of trajectory departure and destination, the dissimilarity model of trajectory mean course, and the dissimilarity model of trajectory spatial distribution are optimal and stable, whereas the time consumption of the dissimilarity model of trajectory spatial distribution and the dissimilarity model of trajectory length are significantly higher than those of other models. The adaptabilities of each dissimilarity are similar in different scenarios. By increasing the dissimilarity category in the combined dissimilarity model, the trajectory recognition effect can be improved. 3 tabs, 12 figs, 30 refs.
-
图 1 船舶轨迹相异度有效性研究流程
Figure 1. Flow of effectiveness research of ship trajectory dissimilarity
图 6 类型A~F和a~f船舶轨迹特征分布
Figure 6. Feature distributions of ship trajectories A-F and a-f
图 9 武汉段水域k值选取试验结果
Figure 9. Experimental results of k value selection in Wuhan waterway
图 10 湛江港水域k值选取试验结果
Figure 10. Experimental results of k value selection in Zhanjiang Port
图 11 单个相异度模型KNN分类试验评估结果
Figure 11. Evaluation results of KNN classification experiments based on single dissimilarity model
图 12 武汉段水域内分类错误轨迹特征分布及实例
Figure 12. Feature distribution and example of wrong classification trajectories in Wuhan waterway
表 1 船舶原始轨迹类别与数目
Table 1. Categories and numbers of ship original trajectories
区域 武汉段水域船舶轨迹 湛江港水域船舶轨迹 类型 A B C D E F a(L) b(U) c(L) d(U) e f 数目 110 1 825 100 28 52 27 706 465 513 125 104 111 
下载: 导出CSV
表 2 基于单个相异度模型的KNN分类耗时统计
Table 2. Time-consuming statistics of KNN classification based on single dissimilarity model
相异度模型 D1 D2 D3 D4 D5 D6 D7 时间/s 武汉段水域 4.571 104.961 1 101.140 1.059 1.138 1.260 3.054 湛江港水域 3.311 24.695 228.054 0.854 0.991 1.089 1.318
下载: 导出CSV
表 3 相异度模型组合下轨迹分类评估结果
Table 3. Evaluation results of KNN classification experiments based on combined dissimilarity models
组合相异度 Mi Ma 耗时/s 武汉段水域 湛江港水域 武汉段水域 湛江港水域 武汉段水域 湛江港水域 Dc, 1 0.993 0.991 0.980 0.991 2.057×103 304.180 Dc, 2 0.990 0.997 0.998 0.996 2.058×103 357.749 Dc, 3 0.996 0.996 0.993 0.995 2.006×103 371.963
下载: 导出CSV
-
[1] 马枫, 初秀民, 严新平. AIS基站短消息特性[J]. 交通运输工程学报, 2012, 12(6): 111-118. http://jtysgcxb.xml-journal.net/article/id/201206017MA Feng, CHU Xiu-min, YAN Xin-ping. Short message characteristics of AIS base stations[J]. Journal of Traffic and Transportation Engineering, 2012, 12(6): 111-118. (in Chinese) http://jtysgcxb.xml-journal.net/article/id/201206017 [2] ZHANG Li-ye, MENG Qiang, XIAO Zhe, et al. A novel ship trajectory reconstruction approach using AIS data[J]. Ocean Engineering, 2018, 159: 165-174. doi: 10.1016/j.oceaneng.2018.03.085 [3] 刘兴龙, 初秀民, 马枫, 等. AIS报文异常动态信息甄别方法[J]. 交通运输工程学报, 2016, 16(5): 142-150. http://jtysgcxb.xml-journal.net/article/id/201605016LIU Xing-long, CHU Xiu-min, MA Feng, et al. Discriminating method of abnormal dynamic information in AIS messages[J]. Journal of Traffic and Transportation Engineering, 2016, 16(5): 142-150. (in Chinese) http://jtysgcxb.xml-journal.net/article/id/201605016 [4] 雷进宇, 初秀民, 何伟, 等. 桥区船舶交通流可视分析系统[J]. 上海交通大学学报, 2017, 51(7): 840-845. https://www.cnki.com.cn/Article/CJFDTOTAL-SHJT201707011.htmLEI Jin-yu, CHU Xiu-min, HE Wei, et al. Visual analytic system of vessel traffic in bridge waterway[J]. Journal of Shanghai Jiaotong University, 2017, 51(7): 840-845. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-SHJT201707011.htm [5] SVANBERG M, SANTÉN V, HÖRTEBORN A, et al. AIS in maritime research[J]. Marine Policy, 2019, 106: 103520. doi: 10.1016/j.marpol.2019.103520 [6] 向哲, 胡勤友, 施朝健, 等. 基于AIS数据的受限水域船舶领域计算方法[J]. 交通运输工程学报, 2015, 15(5): 110-117. http://jtysgcxb.xml-journal.net/article/id/201505014XIANG Zhe, HU Qin-you, SHI Chao-jian, et al. Computation method of ship domains in restricted waters based on AIS data[J]. Journal of Traffic and Transportation Engineering, 2015, 15(5): 110-117. (in Chinese) http://jtysgcxb.xml-journal.net/article/id/201505014 [7] LEI Jin-yu, LIU Lei, CHU Xiu-min, et al. Automatic identification system data-driven model for analysis of ship domain near bridge-waters[J]. Journal of Navigation, 2021, DOI: 10.1017/S0373463321000461. [8] YANG Dong, WU Ling-xiao, WANG Shuai-an, et al. How big data enriches maritime research—a critical review of automatic identification system (AIS) data applications[J]. Transport Reviews, 2019, 39(6): 755-773. doi: 10.1080/01441647.2019.1649315 [9] ZHANG Li-ye, MENG Qiang, FWA T F. Big AIS data based spatial-temporal analyses of ship traffic in Singapore port waters[J]. Transportation Research Part E: Logistics and Transportation Review, 2019, 129: 287-304. doi: 10.1016/j.tre.2017.07.011 [10] 牟军敏, 张新生, 姚鑫, 等. 基于航行数据的北极地区船舶排放清单[J]. 交通运输工程学报, 2019, 19(5): 116-124. http://jtysgcxb.xml-journal.net/article/id/201905012MOU Jun-min, ZHANG Xin-sheng, YAO Xin, et al. Emission inventory of ship based on navigation data in Arctic region[J]. Journal of Traffic and Transportation Engineering, 2019, 19(5): 116-124. (in Chinese) http://jtysgcxb.xml-journal.net/article/id/201905012 [11] BORKOWSKI P. The ship movement trajectory prediction algorithm using navigational data fusion[J]. Sensors, 2017, DOI: 10.3390/s17061432. [12] ZHEN Rong, JIN Yong-xing, HU Qin-you, et al. Maritime anomaly detection within coastal waters based on vessel trajectory clustering and Naïve Bayes classifier[J]. Journal of Navigation, 2017, 70(3): 648-670. doi: 10.1017/S0373463316000850 [13] 何正伟, 杨帆, 刘力荣. 基于AIS数据的船舶安全航行水深参考图[J]. 交通运输工程学报, 2018, 18(4): 171-181. http://jtysgcxb.xml-journal.net/article/id/201804018HE Zheng-wei, YANG Fan, LIU Li-rong. Ship safe navigation depth reference map based on AIS data[J]. Journal of Traffic and Transportation Engineering, 2018, 18(4): 171-181. (in Chinese) http://jtysgcxb.xml-journal.net/article/id/201804018 [14] 徐垚, 李卓然, 孟金龙, 等. 基于大规模船舶轨迹数据的航道边界提取方法[J]. 计算机应用, 2019, 39(1): 105-112. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201901021.htmXU Yao, LI Zhuo-ran, MENG Jin-long, et al. Extraction method of marine lane boundary from exploiting trajectory big data[J]. Journal of Computer Application, 2019, 39(1): 105-112. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201901021.htm [15] ZHANG Shu-kai, SHI Guo-you, LIU Zheng-jiang, et al. Data-driven based automatic maritime routing from massive AIS trajectories in the face of disparity[J]. Ocean Engineering, 2018, 155: 240-250. doi: 10.1016/j.oceaneng.2018.02.060 [16] YAN Zhao-jin, XIAO Yi-jia, CHENG Liang, et al. Exploring AIS data for intelligent maritime routes extraction[J]. Applied Ocean Research, 2020, DOI: 10.1016/j.apor.2020.102271. [17] FILIPIAK D, W$ \mathop {\rm{E}}\limits_{\rm{c}} $CEL K, STRÓŻYNA M, et al. Extracting maritime traffic networks from AIS data using evolutionary algorithm[J]. Business and Information Systems Engineering, 2020, 62(5): 435-450. doi: 10.1007/s12599-020-00661-0 [18] 赵梁滨, 史国友, 杨家轩. 基于DBSCAN算法的船舶轨迹自适应层次聚类[J]. 中国航海, 2018, 41(3): 53-58. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGHH201803011.htmZHAO Liang-bin, SHI Guo-you, YANG Jia-xuan. Adaptive hierarchical clustering of ship trajectory with DBSCAN algorithm[J]. Navigation of China, 2018, 41(3): 53-58. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGHH201803011.htm [19] 牟军敏, 陈鹏飞, 贺益雄, 等. 船舶AIS轨迹快速自适应谱聚类算法[J]. 哈尔滨工程大学学报, 2018, 39(3): 428-432. https://www.cnki.com.cn/Article/CJFDTOTAL-HEBG201803005.htmMOU Jun-min, CHEN Peng-fei, HE Yi-xiong, et al. Fast self-tuning spectral clustering algorithm for AIS ship trajectory[J]. Journal of Harbin Engineering University, 2018, 39(3): 428-432. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HEBG201803005.htm [20] 刘磊, 初秀民, 蒋仲廉, 等. 基于KNN的船舶轨迹分类算法[J]. 大连海事大学学报, 2018, 44(3): 15-21. https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201803005.htmLIU Lei, CHU Xiu-min, JIANG Zhong-lian, et al. Ship trajectory classification algorithm based on KNN[J]. Journal of Dalian Maritime University, 2018, 44(3): 15-21. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DLHS201803005.htm [21] SHENG Kai, LIU Zhong, ZHOU De-chao, et al. Research on ship classification based on trajectory features[J]. The Journal of Navigation, 2018, 71(1): 100-116. doi: 10.1017/S0373463317000546 [22] LEE J, HAN J, WHANG K. Trajectory clustering: a partition-and-group framework[C]//ACM. 2007 ACM SIGMOD International Conference on Management of Data. New York: ACM Press, 2007: 593-604. [23] 江玉玲, 熊振南, 唐基宏. 基于轨迹段DBSCAN的船舶轨迹聚类算法[J]. 中国航海, 2019, 42(3): 1-5. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGHH201903001.htmJIANG Yu-ling, XIONG Zhen-nan, TANG Ji-hong. Ship trajectory clustering algorithm based on DBSCAN[J]. Navigation of China, 2019, 42(3): 1-5. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGHH201903001.htm [24] SHENG Pan, YIN Jing-bo. Extracting shipping route patterns by trajectory clustering model based on automatic identification system data[J]. Sustainability, 2018, 10(7): 2327. doi: 10.3390/su10072327 [25] LI Huan-huan, LIU Jing-xian, LIU R W, et al. A dimensionality reduction-based multi-step clustering method for robust vessel trajectory analysis[J]. Sensors, 2017, DOI: 10.3390/s17081792. [26] 彭祥文, 高曙, 初秀民, 等. 基于Spark的船舶航行轨迹聚类方法[J]. 中国航海, 2017, 40(3): 49-53. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGHH201703011.htmPENG Xiang-wen, GAO Shu, CHU Xiu-min, et al. Clustering method of ship's navigation trajectory set based on spark[J]. Navigation of China, 2017, 40(3): 49-53. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGHH201703011.htm [27] FU Pei-guo, WANG Hao-zhou, LIU Kui-en, et al. Finding abnormal vessel trajectories using feature learning[J]. IEEE Access, 2017, 5: 7898-7909. doi: 10.1109/ACCESS.2017.2698208 [28] ZHAO Liang-bin, SHI Guo-you. A trajectory clustering method based on Douglas-Peucker compression and density for marine traffic pattern recognition[J]. Ocean Engineering, 2019, 172: 456-467. doi: 10.1016/j.oceaneng.2018.12.019 [29] 高邈, 史国友, 李伟峰. 改进的Sliding Window在线船舶AIS轨迹数据压缩算法[J]. 交通运输工程学报, 2018, 18(3): 218-227. http://jtysgcxb.xml-journal.net/article/id/201803022GAO Miao, SHI Guo-you, LI Wei-feng. Online compression algorithm of AIS trajectory data based on improved sliding window[J]. Journal of Traffic and Transportation Engineering, 2018, 18(3): 218-227. (in Chinese) http://jtysgcxb.xml-journal.net/article/id/201803022 [30] ZHENG Zu-duo, SU Dong-cai. Short-term traffic volume forecasting: a k-nearest neighbor approach enhanced by constrained linearly sewing principle component algorithm[J]. Transportation Research Part C: Emerging Technologies, 2014, 43: 143-157. doi: 10.1016/j.trc.2014.02.009 -
点击查看大图
图(12) / 表(3)
计量
- 文章访问数: 1138
- HTML全文浏览量: 227
- PDF下载量: 136
- 被引次数: 0
