| 一种基于DBSCAN的船舶会遇实时识别方法 |
| |
| 引用本文： | 甄荣,RIVEIRO Mari,金永兴.一种基于DBSCAN的船舶会遇实时识别方法[J].上海海事大学学报,2018,39(1):1-5. |
| |
| 作者姓名： | 甄荣 RIVEIRO Mari 金永兴 |
| |
| 作者单位： | 1.上海海事大学商船学院； 2.舍夫德大学信息研究中心,舍夫德大学信息研究中心,上海海事大学商船学院 |
| |
| 基金项目： | 国家自然科学基金（51709165）；国家留学基金管理委员会联合培养博士生项目（201608310093）；上海市科学技术委员会地方院校能力建设项目（15590501600）；上海海事大学研究生创新基金(2016ycx077)；上海海事大学优秀博士学位论文培育项目（2017bxlp003） |
| |
| 摘 要： | 针对海上交通监控中船舶数量众多,且对具有潜在碰撞危险的船舶识别效率不高的问题,提出一种基于DBSCAN(带噪声的基于密度的空间聚类)的船舶会遇实时识别方法。根据海上交通风险监控的研究需求,分析船舶会遇局面的定义。运用墨卡托算法计算船舶之间的距离,采用DBSCAN算法进行船舶会遇聚类识别。基于浙江舟山群岛西南海域航行船舶的AIS数据,对设置不同船舶会遇距离的试验结果进行比较分析,结果表明:当船舶会遇距离为1 n mile时,可以将56艘船划分为7个会遇船舶类,占船舶总数的32.1%,每个会遇船舶类包括2~3艘船。将该方法运用到实际海上交通监控中,可对每个会遇船舶类中的船舶航行动态进行重点关注,降低海上交通监控人员的工作负担,提高海上交通监控的效率。
|
| 关 键 词： | 海上交通监控 DBSCAN 船舶会遇 航行风险 |
| 收稿时间： | 2017/5/9 0:00:00 |
| 修稿时间： | 2017/7/14 0:00:00 |
A real-time identification method to ship encounter based on DBSCAN |
| |
| Institution: | Merchant Marine College,Shanghai Maritime University,Informatics Research Centre, University of Skovde and Merchant Marine College, Shanghai Maritime University |
| |
| Abstract: | Aiming at the problem that there are a large number of ships in maritime traffic monitoring and the recognition efficiency for ships with potential collision risk is low, a real time identification method to ship encounter based on DBSCAN(density based spatial clustering of applications with noise) is proposed. The definition of ship encounter situation is analyzed according to the research demand of maritime traffic risk monitoring. The distance between ships is calculated using the Mercator algorithm, and the DBSCAN algorithm is applied to the cluster identification of ship encounter. Based on AIS data of ships sailing on southwest sea area of Zhoushan Islands in Zhejiang Province of China, the test results are compared for different ship encounter radii. The results show that, when the ship encounter radius is 1 n mile, 56 ships can be divided into 7 clusters of ship encounter which compose 32.1% of the total number of ships, and each cluster includes 2 or 3 ships. The method is applied to practical maritime traffic monitoring to pay more attention to the dynamic motion of ships in encounter clusters, which can reduce the workload of maritime traffic monitoring personnel and improve the efficiency of maritime traffic monitoring. |
| |
| Keywords: | maritime traffic monitoring DBSCAN ship encounter sailing risk |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《上海海事大学学报》浏览原始摘要信息 |
| 点击此处可从《上海海事大学学报》下载免费的PDF全文 |
|
