WO2022252398A1 - 基于船舶轨迹特征点提取的时空dp方法 - Google Patents

基于船舶轨迹特征点提取的时空dp方法 Download PDF

Info

Publication number
WO2022252398A1
WO2022252398A1 PCT/CN2021/112036 CN2021112036W WO2022252398A1 WO 2022252398 A1 WO2022252398 A1 WO 2022252398A1 CN 2021112036 W CN2021112036 W CN 2021112036W WO 2022252398 A1 WO2022252398 A1 WO 2022252398A1
Authority
WO
WIPO (PCT)
Prior art keywords
trajectory
ship
ais
point
points
Prior art date
Application number
PCT/CN2021/112036
Other languages
English (en)
French (fr)
Inventor
马勇
江海洋
严新平
Original Assignee
武汉理工大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 武汉理工大学 filed Critical 武汉理工大学
Priority to US17/615,111 priority Critical patent/US11851147B2/en
Publication of WO2022252398A1 publication Critical patent/WO2022252398A1/zh

Links

Images

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B63SHIPS OR OTHER WATERBORNE VESSELS; RELATED EQUIPMENT
    • B63BSHIPS OR OTHER WATERBORNE VESSELS; EQUIPMENT FOR SHIPPING 
    • B63B79/00Monitoring properties or operating parameters of vessels in operation
    • B63B79/40Monitoring properties or operating parameters of vessels in operation for controlling the operation of vessels, e.g. monitoring their speed, routing or maintenance schedules
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G3/00Traffic control systems for marine craft
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/0206Control of position or course in two dimensions specially adapted to water vehicles
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/23Clustering techniques
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20Instruments for performing navigational calculations
    • G01C21/203Specially adapted for sailing ships
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G3/00Traffic control systems for marine craft
    • G08G3/02Anti-collision systems

Definitions

  • the invention belongs to the technical field of ship track compression, and more particularly relates to a space-time Douglas-Peucker (DP) method based on feature point extraction of ship track.
  • DP space-time Douglas-Peucker
  • the current automatic identification system (Automatic Identification System, AIS) base station network framework has been basically formed.
  • AIS Automatic Identification System
  • the shipborne AIS data collected by AIS receiving points in various ports can be received in real time, and maritime authorities can obtain massive Ship AIS trajectory data.
  • Massive ship AIS trajectory data contains a lot of information, including ship static information, dynamic information, human factors of ship pilots, ship collision avoidance behavior, crew's usual practices, customary routes, etc.
  • effective and potential information that can reflect the rules of ships can be obtained, and then provide effective data support for maritime authorities to supervise ship violations, revise navigation rules, and implement ship routing systems.
  • there are some data points with extremely low utilization value in the massive AIS data When these data points are removed, the ship trajectory will not change greatly. Therefore, in order to improve the utilization efficiency of data, it is necessary to compress the redundant ship AIS trajectory data.
  • Conventional ship trajectory compression algorithms often only consider the distance offset of the trajectory to compress the trajectory, and the obtained trajectory often ignores the dynamic information of the ship.
  • the track feature points such as the ship's speed, course change, and entry and exit of a certain area boundary are discarded during the compression process.
  • Reduce the utilization value of the data a few compression algorithms retain the characteristic points of the ship's trajectory through the average value of the course and speed change rate, but ignore the small-scale fluctuations in the speed and course due to sensor errors, and will retain the fluctuation points.
  • compression Too many data points are reserved; a few compression algorithms consider the space-time characteristics of ships, but often only use the time characteristics of ships as an index for classification and sorting.
  • the present invention proposes a spatio-temporal DP method based on ship track feature point extraction, which supplements the deficiencies of the current ship track compression method and solves the problem of retaining feature track points in the process of ship track compression.
  • a spatio-temporal DP method based on ship track feature point extraction, which supplements the deficiencies of the current ship track compression method and solves the problem of retaining feature track points in the process of ship track compression.
  • the shape of the trajectory is better preserved.
  • the present invention provides a spatio-temporal DP method based on ship track feature point extraction, comprising:
  • the AIS data is compressed.
  • step (4) includes:
  • step (5) includes:
  • step (6) includes:
  • the distance between the coordinates of the Mercator coordinate system at the time point is taken as the space-time distance d from the AIS data point to the virtual straight line space-time trajectory, find the maximum distance d max among all space-time distances, and compare the maximum distance with the preset distance threshold d T the size of;
  • step (6.4) If d max >d T , the AIS data point corresponding to the maximum distance should be reserved as the data point on the resulting trajectory, and at the same time, the sub-trajectory is divided into two parts by the AIS data point corresponding to the maximum distance, and the two parts Curves are processed by step (6.2) and step (6.3) respectively, until all d max ⁇ d T ;
  • the trajectory formed by connecting the segmentation points in turn is the approximate trajectory after the compression of the original trajectory.
  • x r 0 ⁇
  • the present invention fully considers the reservation of characteristic track points when compressing the ship's AIS track, and at the same time uses the space-time distance to compress the track to better preserve the shape of the track, and the simplified data has greater secondary use value.
  • Fig. 1 is a kind of ship trajectory compression flowchart provided by the embodiment of the present invention.
  • Fig. 2 is a schematic diagram of calculation of a speed change rate and a course change rate provided by an embodiment of the present invention
  • Fig. 3 is a kind of time-space DP method schematic diagram based on ship trajectory feature point extraction that the embodiment of the present invention provides;
  • Fig. 4 is a kind of overall ship trajectory compression result diagram provided by the embodiment of the present invention, wherein, (a) represents the original trajectory point diagram, (b) represents the compressed trajectory point diagram;
  • Fig. 5 is a kind of single-vessel trajectory compression result figure provided by the embodiment of the present invention.
  • Fig. 6 is a feature point of an access bridge area provided by an embodiment of the present invention.
  • S1 Eliminate noise points, use clustering algorithm to cluster and analyze AIS raw data, identify outliers in AIS data, and then remove noise points to construct single-ship AIS time series data records;
  • the position noise points of the ship are mainly eliminated, and after eliminating the data points with large position deviation, a single ship AIS time series data record is constructed.
  • r 0 indicates the radius of the parallel circle at the standard latitude
  • q indicates the equidistant latitude
  • a represents the major radius of the earth ellipsoid
  • e represents the first eccentricity of the earth ellipsoid
  • (x, y) represents the coordinates of the Mercator coordinate system after latitude and longitude conversion.
  • this example uniformly converts the latitude and longitude coordinates of each AIS data point into the coordinates of the Mercator coordinate system, and uniformly converts the time into seconds.
  • S3 the speed change rate and course change rate of each AIS data point, and the average speed change rate and average course change rate during the entire navigation process;
  • S cri and C cri represent the speed change rate and heading change rate of the i-th AIS data point respectively, represent the average speed change rate and the average heading change rate during the entire voyage, respectively, Indicates the speed of the i+1th AIS data point, Indicates the speed of the i-1th AIS data point, Indicates the heading of the i+1th AIS data point, Indicates the heading of the i-1th AIS data point, ⁇ t represents the time interval between the i+1th AIS data point and the i-1th AIS data point, and n represents the number of AIS data points, as shown in Figure 2.
  • the ship's speed and heading data will fluctuate in a small range. If the average rate of change is directly used as the threshold, these fluctuation points may be retained as trajectory feature points of speed changes and heading changes, resulting in compression After the data volume is still huge, the expansion coefficient M is introduced in the present invention, and the average speed change rate and the average heading change rate are enlarged.
  • Ship entry/exit behaviors include entry/exit of docks, anchorages, bridge area waters, fishing area waters, roundabouts and other closed areas, as well as trajectory points in non-closed areas such as waterways, danger lines, and boundary lines, and determine two adjacent AIS Whether the product of the values after the data points are substituted into the boundary line equation is less than 0, if it is less than 0, it will be marked and reserved as the track point of the ship entering and leaving a certain area to form a point set E of entering and exiting a certain area.
  • B i (x i , y i ) is the coordinate of the i-th AIS data point in the sub-track segment in the Mercator coordinate system
  • B′ i (x′ i , y′ i ) is the coordinate of B i in the virtual straight line space-time The coordinates in the Mercator coordinate system of the points on the trajectory at the same time.
  • the distance threshold d T is set to 80.
  • each step/component described in this application can be split into more steps/components, and two or more steps/components or part of the operations of steps/components can also be combined into a new Step/component, to realize the object of the present invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Ocean & Marine Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Data Mining & Analysis (AREA)
  • Mechanical Engineering (AREA)
  • Combustion & Propulsion (AREA)
  • Chemical & Material Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Navigation (AREA)
  • Traffic Control Systems (AREA)
  • Instructional Devices (AREA)

Abstract

本发明公开了一种基于船舶轨迹特征点提取的时空DP方法,属于船舶轨迹压缩技术领域,包括:步骤1:利用聚类算法对AIS原始数据进行聚类分析,识别AIS数据中的离群点,进而对噪声点进行剔除;步骤2:对船舶航向改变、航速改变、船舶进出某区域等特征轨迹点进行识别与保留;步骤3:以船舶轨迹的起点、终点以及步骤2保留的特征轨迹点为初始点,同时考虑AIS数据的时空特性,对AIS数据进行压缩。利用本发明可以对冗杂AIS数据进行有效压缩,压缩后船舶轨迹与原轨迹差异极小,同时能够保留船舶运动状态改变点,船舶进出区域边界点的信息,再利用价值空间大,能够为船舶历史数据分析,船舶行为识别奠定数据处理基础。

Description

基于船舶轨迹特征点提取的时空DP方法 技术领域
本发明属于船舶轨迹压缩技术领域,更具体地,涉及一种基于船舶轨迹特征点提取的时空道格拉斯-普克(Douglas–Peucker,DP)方法。
背景技术
经过多年的建设,目前船舶自动识别系统(Automatic Identification System,AIS)基站网络框架已经基本形成,通过AIS信息采集系统实时地接收各港口AIS接收点采集到的船载AIS数据,海事机关可以获得海量的船舶AIS轨迹数据。海量的船舶AIS轨迹数据中,蕴含着大量信息,包括船舶的静态信息,动态信息,船舶驾驶员的人为因素,船舶避碰行为,船员通常做法,习惯航路等。通过对船舶轨迹的分析研究,可以从中获取能够反映船舶规律的、有效的、潜在的信息,进而为海事机关对船舶违章行为监管,修订航行规则,推行船舶定线制提供有效的数据支持。然而海量的AIS数据中存在一些利用价值极低的数据点,当移除这些数据点后船舶轨迹不会产生较大的改变。因此,为提高数据的利用效率,需要对冗杂的船舶AIS轨迹数据进行压缩处理。
常规的船舶轨迹压缩算法往往只考虑轨迹的距离偏移量来压缩轨迹,得到的轨迹往往忽略船舶的动态信息,船舶航速、航向改变、进出某区域边界等航迹特征点在压缩过程被舍弃,降低了数据的利用价值;少数压缩算法通过航向、航速变化率均值对船舶轨迹特征点进行保留,但忽略由于传感器的误差,航速、航向会出现小范围波动,会将波动点进行保留,压缩后保留分数据点过多;少数压缩算法考虑船舶的时空特性,但往往只将船舶的时间特性仅作为分类和排序的指标。
发明内容
针对现有技术的以上缺陷或改进需求,本发明提出了一种基于船舶轨迹特征点提取的时空DP方法,补充目前船舶轨迹压缩方法的不足,解决船舶轨迹压缩过程中特征轨迹点的保留问题,同时兼顾AIS数据的时空特性较好的保留了轨迹的形状。
为实现上述目的,本发明提供了一种基于船舶轨迹特征点提取的时空DP方法,包括:
(1)对AIS原始数据进行聚类分析,识别AIS数据中的离群点,进而对噪声点进行剔除,构建单船AIS时序性数据记录;
(2)将单船AIS时序性数据记录中各AIS数据点的经纬度坐标转化为墨卡托投影坐标;
(3)获取各AIS数据点的航速变化率、航向变化率以及整个航行过程中的平均航速变化率、平均航向变化率;
(4)识别并保留单船AIS时序性数据记录中的船舶航向和航速的改变点;
(5)识别并保留单船AIS时序性数据记录中的船舶进出某区域轨迹点;
(6)以船舶轨迹的起点、终点以及保留的船舶航向和航速的改变点、船舶进出某区域轨迹点为初始点,同时考虑AIS数据的时空特性,对AIS数据进行压缩。
在一些可选的实施方案中,由
Figure PCTCN2021112036-appb-000001
得到第i个AIS数据点的航速变化率S cri,由
Figure PCTCN2021112036-appb-000002
得到第i个AIS数据点的航向变化率C cri,由
Figure PCTCN2021112036-appb-000003
得到整个航行过程中的平均航速变化率
Figure PCTCN2021112036-appb-000004
Figure PCTCN2021112036-appb-000005
得到整个航行过程中的平均航向变化率
Figure PCTCN2021112036-appb-000006
表示第i+1个AIS数据点的航速,
Figure PCTCN2021112036-appb-000007
表示第i-1个AIS数据点的航速,
Figure PCTCN2021112036-appb-000008
表示 第i+1个AIS数据点的航向,
Figure PCTCN2021112036-appb-000009
表示第i-1个AIS数据点的航向,Δt表示第i+1个AIS数据点和第i-1个AIS数据点的时间间隔,n表示AIS数据点个数。
在一些可选的实施方案中,步骤(4)包括:
设置船舶航速改变的阈值
Figure PCTCN2021112036-appb-000010
依次判断各个AIS数据点B i的航速变化率S cri与S tre的大小,如果S cri≥S tre,则航速改变点集合S=S∪B i
设置船舶航向改变的阈值
Figure PCTCN2021112036-appb-000011
依次判断各个AIS数据点P i的航向变化率C cri与C tre的大小,如果C cri≥C tre,则航向改变点集合C=C∪P i,M和N表示系数。
在一些可选的实施方案中,步骤(5)包括:
判断相邻两个AIS数据点分别代入区域边界线方程后值的乘积是否小于0,若小于0,则将该相邻两个AIS数据点标记并保留为船舶进出某区域轨迹点,构成进出某区域点集合E。
在一些可选的实施方案中,步骤(6)包括:
(6.1)设置距离阈值d T,以船舶轨迹的起点、终点以及保留的S,E,C中的特征轨迹点为初始点对轨迹进行分段标记,相邻两个轨迹特征点之间的轨迹为一个子轨迹段;
(6.2)连接每个分段航迹的起点和终点,并根据起点与终点的经度,纬度转换后的墨卡托坐标系坐标和时间建立虚拟直线时空轨迹,对每个子轨迹段,计算该子轨迹段AIS数据点在虚拟直线时空轨迹上同时刻点的墨卡托坐标系坐标,将该子轨迹段的AIS数据点的墨卡托坐标系坐标与该AIS数据点在虚拟直线时空轨迹上同时刻点的墨卡托坐标系坐标之间的距离作为该AIS数据点到虚拟直线时空轨迹的时空距离d,找到所有时空距离中的最大距离d max,比较该最大距离与预设距离阈值d T的大小;
(6.3)如果d max<d T,则该子轨迹段上所有中间数据点全部舍掉,舍掉所有中间点后,连接该子轨迹段起点和终点的直线就作为该子轨迹段的近似,该段子轨迹处理完毕;
(6.4)如果d max>d T,则对应最大距离的AIS数据点应保留为结果轨迹上的数据点,同时通过对应最大距离的AIS数据点将该段子轨迹分为两部分,对这两部分曲线分别采用步骤(6.2)和步骤(6.3)进行处理,直到所有的d max<d T
(6.5)当所有子轨迹段处理完后,依次连接各分割点形成的轨迹,即为原轨迹压缩后的近似轨迹。
在一些可选的实施方案中,由
Figure PCTCN2021112036-appb-000012
Figure PCTCN2021112036-appb-000013
x=r 0×λ,y=r 0×q将单船AIS时序性数据记录中各AIS数据点的经纬度坐标转化为墨卡托投影坐标,其中,
Figure PCTCN2021112036-appb-000014
表示AIS数据点的经纬度坐标,r 0表示标准纬度的平行圆半径,q表示等距纬度,
Figure PCTCN2021112036-appb-000015
表示墨卡托投影的标准纬度,a表示地球椭球的长半径,e表示地球椭球的第一偏心率,(x,y)表示经纬度转换后的墨卡托坐标系坐标。
在一些可选的实施方案中,M∈[9,11],N∈[3,5]。
总体而言,通过本发明所构思的以上技术方案与现有技术相比,能够取得下列有益效果:
本发明在对船舶AIS轨迹压缩时,充分考虑了特征轨迹点的保留问题,同时利用时空距离压缩轨迹较好的保留了轨迹的形状,简化后的数据有较大的二次利用价值。
附图说明
图1是本发明实施例提供的一种船舶轨迹压缩流程图;
图2是本发明实施例提供的一种航速变化率和航向变化率计算示意图;
图3是本发明实施例提供的一种基于船舶轨迹特征点提取的时空DP方 法原理图;
图4是本发明实施例提供的一种总体船舶轨迹压缩结果图,其中,(a)表示原始轨迹点图,(b)表示压缩后的轨迹点图;
图5是本发明实施例提供的一种单船轨迹压缩结果图;
图6是本发明实施例提供的一种进出桥区特征点。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白,以下结合附图及实施例,对本发明进行进一步详细说明。应当理解,此处所描述的具体实施例仅仅用以解释本发明,并不用于限定本发明。此外,下面所描述的本发明各个实施方式中所涉及到的技术特征只要彼此之间未构成冲突就可以相互组合。
本实施例采用长江武汉段2016年8月9日当天所采集AIS数据作为原始数据进行压缩。如图1所示,本发明采取的技术方案是:
S1:噪声点剔除,利用聚类算法对AIS原始数据进行聚类分析,识别AIS数据中的离群点,进而对噪声点进行剔除,构建单船AIS时序性数据记录;
本实例中主要对船舶的位置噪音点进行剔除,剔除位置偏差较大的数据点后,构建单船AIS时序性数据记录。
S2:为方便计算距离,将单船AIS时序性数据记录中各AIS数据点的经纬度坐标转化为墨卡托投影坐标;
Figure PCTCN2021112036-appb-000016
Figure PCTCN2021112036-appb-000017
x=r 0×λ
y=r 0×q
其中,
Figure PCTCN2021112036-appb-000018
表示AIS数据点的经纬度坐标,r 0表示标准纬度的平行圆半径,q表示等距纬度,
Figure PCTCN2021112036-appb-000019
表示墨卡托投影的标准纬度,a表示地球椭球的长半径,e表示地球椭球的第一偏心率,(x,y)表示经纬度转换后的墨卡托坐标系坐标。
本实例为了方便计算,提高计算精度,统一将各AIS数据点的经纬度坐标转化为墨卡托坐标系坐标,时间统一转化为以秒为单位。
S3:各AIS数据点的航速变化率、航向变化率以及整个航行过程中的平均航速变化率、平均航向变化率;
Figure PCTCN2021112036-appb-000020
Figure PCTCN2021112036-appb-000021
Figure PCTCN2021112036-appb-000022
Figure PCTCN2021112036-appb-000023
其中,S cri、C cri分别表示第i个AIS数据点的航速变化率、航向变化率,
Figure PCTCN2021112036-appb-000024
分别表示整个航行过程中的平均航速变化率、平均航向变化率,
Figure PCTCN2021112036-appb-000025
表示第i+1个AIS数据点的航速,
Figure PCTCN2021112036-appb-000026
表示第i-1个AIS数据点的航速,
Figure PCTCN2021112036-appb-000027
表示第i+1个AIS数据点的航向,
Figure PCTCN2021112036-appb-000028
表示第i-1个AIS数据点的航向,Δt表示第i+1个AIS数据点和第i-1个AIS数据点的时间间隔,n表示AIS数据点个数,如图2所示。
S4:识别并保留单船AIS时序性数据记录中的船舶航向和航速的改变点;
受限于传感器的精度,船舶的航速,航向数据会存在小范围的波动,如果直接以平均变化率作为阈值,可能会将这些波动点当成航速变化、航 向变化的轨迹特征点进行保留,导致压缩后数据量依然庞大,本发明中引入扩大系数M,N对平均航速变化率和平均航向变化率进行扩大处理。
设初始航速改变点的集合S={},设置船舶航速改变的阈值
Figure PCTCN2021112036-appb-000029
依次判断各个AIS数据点B i的航速变化率S cri与S tre的大小,如果S cri≥S tre,则航速改变点集合S=S∪B i
设初始航向改变点的集合C={},设置船舶航向改变的阈值
Figure PCTCN2021112036-appb-000030
依次判断各个AIS数据点P i的航向变化率C cri与C tre的大小,如果C cri≥C tre,则航向改变点集合C=C∪P i
其中,M∈[9,11],N∈[3,5]。
本实例中取M=10,N=4,构建船舶航速变化率阈值和航向变化率阈值,进行航速特征点和航向特征点的保留。
S5:识别并保留单船AIS时序性数据记录中的船舶进出某区域轨迹点;
大部分船舶轨迹压缩算法都没有将这些点作为轨迹特征点进行保留,但是这些数据点往往包含着驾驶员人为因素,驾驶员通常做法,习惯航路的潜在信息,有一定的利用价值。
船舶驶入/出行为包括驶入/出码头、锚地、桥区水域、渔区水域、环形道等闭合区域以及航道、危险线、边界线等非闭合区域的轨迹点,判断相邻两个AIS数据点分别代入边界线方程后值的乘积是否小于0,若小于0,则标记并保留为船舶进出某区域轨迹点,构成进出某区域点集合E。
本实例对船舶驶入/驶出武汉长江二桥的特征点进行保留。
S6:考虑AIS的时空特性压缩船舶轨迹,具体地,如图3所示:
S6.1:设置距离阈值d T,以船舶轨迹的起点、终点以及以上步骤保留的S,E,C中的轨迹特征点为初始点对轨迹进行分段标记,相邻两个轨迹特征点之间的轨迹为一个子轨迹段;
S6.2:连接每个分段航迹的起点和终点,并根据起点与终点的经度,纬 度转换后的墨卡托坐标系坐标(x,y)和时间建立虚拟直线时空轨迹,对每个子轨迹段,计算各分段AIS数据点B i(x i,y i)在虚拟直线时空轨迹上同时刻点B′ i的墨卡托坐标系坐标(x′ i,y′ i),计算各AIS数据点到虚拟直线时空轨迹的时空距离d,即B iB′ i之间距离,找到所有距离中的最大距离d max,比较该最大距离与预设距离阈值d T的大小;
Figure PCTCN2021112036-appb-000031
其中,B i(x i,y i)为子航迹段第i个AIS数据点墨卡托坐标系下的坐标,B′ i(x′ i,y′ i)为B i在虚拟直线时空轨迹上同时刻点墨卡托坐标系下的坐标。
S6.3:如果d max<d T,则这条轨迹上所有中间数据点全部舍掉,舍掉所有中间点后,连接该子轨迹段起点和终点的直线就作为这条轨迹的近似,该段轨迹处理完毕;
S6.4:如果d max≥d T,则对应最大距离的AIS数据点应保留为结果轨迹上的数据点,同时通过该数据点将该段轨迹分为两部分,对这两部分曲线分别采用S6.2和S6.3进行处理,直到所有的d max<d T
S6.5:当所有子轨迹段处理完后,依次连接各分割点形成的轨迹,即为原轨迹压缩后的近似轨迹。
总体压缩结果如图4所示,其中,图4中(a)为原始轨迹点图,图4中(b)为压缩后的轨迹点图,其总体形状非常相似,可以证明本发明方法在高效压缩的同时保留轨迹的形状特征。单船轨迹压缩结果如图5、图6所示,可以看到轨迹点数量较少,同时特征点保留也较为完整。
本实例中距离阈值d T取80。
需要指出,根据实施的需要,可将本申请中描述的各个步骤/部件拆分为更多步骤/部件,也可将两个或多个步骤/部件或者步骤/部件的部分操作组合成新的步骤/部件,以实现本发明的目的。
本领域的技术人员容易理解,以上所述仅为本发明的较佳实施例而已, 并不用以限制本发明,凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等,均应包含在本发明的保护范围之内。

Claims (7)

  1. 一种基于船舶轨迹特征点提取的时空DP方法,其特征在于,包括:
    (1)对AIS原始数据进行聚类分析,识别AIS数据中的离群点,进而对噪声点进行剔除,构建单船AIS时序性数据记录;
    (2)将单船AIS时序性数据记录中各AIS数据点的经纬度坐标转化为墨卡托投影坐标;
    (3)获取各AIS数据点的航速变化率、航向变化率以及整个航行过程中的平均航速变化率、平均航向变化率;
    (4)识别并保留单船AIS时序性数据记录中的船舶航向和航速的改变点;
    (5)识别并保留单船AIS时序性数据记录中的船舶进出某区域轨迹点;
    (6)以船舶轨迹的起点、终点以及保留的船舶航向和航速的改变点、船舶进出某区域轨迹点为初始点,同时考虑AIS数据的时空特性,对AIS数据进行压缩。
  2. 根据权利要求1所述的方法,其特征在于,由
    Figure PCTCN2021112036-appb-100001
    得到第i个AIS数据点的航速变化率S cri,由
    Figure PCTCN2021112036-appb-100002
    得到第i个AIS数据点的航向变化率C cri,由
    Figure PCTCN2021112036-appb-100003
    得到整个航行过程中的平均航速变化率
    Figure PCTCN2021112036-appb-100004
    Figure PCTCN2021112036-appb-100005
    得到整个航行过程中的平均航向变化率
    Figure PCTCN2021112036-appb-100006
    Figure PCTCN2021112036-appb-100007
    表示第i+1个AIS数据点的航速,
    Figure PCTCN2021112036-appb-100008
    表示第i-1个AIS数据点的航速,
    Figure PCTCN2021112036-appb-100009
    表示第i+1个AIS数据点的航向,
    Figure PCTCN2021112036-appb-100010
    表示第i-1个AIS数据点的航向,Δt表示第i+1个AIS数据点和第i-1个AIS数据点的时间间隔,n表示AIS数据点个数。
  3. 根据权利要求2所述的方法,其特征在于,步骤(4)包括:
    设置船舶航速改变的阈值
    Figure PCTCN2021112036-appb-100011
    依次判断各个AIS数据点B i 的航速变化率S cri与S tre的大小,如果S cri≥S tre,则航速改变点集合S=S∪B i
    设置船舶航向改变的阈值
    Figure PCTCN2021112036-appb-100012
    依次判断各个AIS数据点P i的航向变化率C cri与C tre的大小,如果C cri≥C tre,则航向改变点集合C=C∪P i,M和N表示系数。
  4. 根据权利要求3所述的方法,其特征在于,步骤(5)包括:
    判断相邻两个AIS数据点分别代入区域边界线方程后值的乘积是否小于0,若小于0,则将该相邻两个AIS数据点标记并保留为船舶进出某区域轨迹点,构成进出某区域点集合E。
  5. 根据权利要求4所述的方法,其特征在于,步骤(6)包括:
    (6.1)设置距离阈值d T,以船舶轨迹的起点、终点以及保留的S,E,C中的特征轨迹点为初始点对轨迹进行分段标记,相邻两个轨迹特征点之间的轨迹为一个子轨迹段;
    (6.2)连接每个分段航迹的起点和终点,并根据起点与终点的经度,纬度转换后的墨卡托坐标系坐标和时间建立虚拟直线时空轨迹,对每个子轨迹段,计算该子轨迹段AIS数据点在虚拟直线时空轨迹上同时刻点的墨卡托坐标系坐标,将该子轨迹段的AIS数据点的墨卡托坐标系坐标与该AIS数据点在虚拟直线时空轨迹上同时刻点的墨卡托坐标系坐标之间的距离作为该AIS数据点到虚拟直线时空轨迹的时空距离d,找到所有时空距离中的最大距离d max,比较该最大距离与预设距离阈值d T的大小;
    (6.3)如果d max<d T,则该子轨迹段上所有中间数据点全部舍掉,舍掉所有中间点后,连接该子轨迹段起点和终点的直线就作为该子轨迹段的近似,该段子轨迹处理完毕;
    (6.4)如果d max>d T,则对应最大距离的AIS数据点应保留为结果轨迹上的数据点,同时通过对应最大距离的AIS数据点将该段子轨迹分为两 部分,对这两部分曲线分别采用步骤(6.2)和步骤(6.3)进行处理,直到所有的d max<d T
    (6.5)当所有子轨迹段处理完后,依次连接各分割点形成的轨迹,即为原轨迹压缩后的近似轨迹。
  6. 根据权利要求1所述的方法,其特征在于,由
    Figure PCTCN2021112036-appb-100013
    Figure PCTCN2021112036-appb-100014
    x=r 0×λ,y=r 0×q将单船AIS时序性数据记录中各AIS数据点的经纬度坐标转化为墨卡托投影坐标,其中,
    Figure PCTCN2021112036-appb-100015
    Figure PCTCN2021112036-appb-100016
    表示AIS数据点的经纬度坐标,r 0表示标准纬度的平行圆半径,q表示等距纬度,
    Figure PCTCN2021112036-appb-100017
    表示墨卡托投影的标准纬度,a表示地球椭球的长半径,e表示地球椭球的第一偏心率,(x,y)表示经纬度转换后的墨卡托坐标系坐标。
  7. 根据权利要求3所述的方法,其特征在于,M∈[9,11],N∈[3,5]。
PCT/CN2021/112036 2021-05-31 2021-08-11 基于船舶轨迹特征点提取的时空dp方法 WO2022252398A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/615,111 US11851147B2 (en) 2021-05-31 2021-08-11 Spatio-temporal DP method based on ship trajectory characteristic point extraction

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110601140.4A CN113450596B (zh) 2021-05-31 2021-05-31 基于船舶轨迹特征点提取的时空dp方法
CN202110601140.4 2021-05-31

Publications (1)

Publication Number Publication Date
WO2022252398A1 true WO2022252398A1 (zh) 2022-12-08

Family

ID=77810472

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/112036 WO2022252398A1 (zh) 2021-05-31 2021-08-11 基于船舶轨迹特征点提取的时空dp方法

Country Status (3)

Country Link
US (1) US11851147B2 (zh)
CN (1) CN113450596B (zh)
WO (1) WO2022252398A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116309708A (zh) * 2023-02-28 2023-06-23 武汉理工大学 一种锚地时空利用效率评价方法和装置

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113901085B (zh) * 2021-09-30 2024-03-15 中远海运科技股份有限公司 一种船舶轨迹动态绘制方法及系统
JP2023060775A (ja) * 2021-10-18 2023-04-28 富士通株式会社 航行監視プログラム、航行監視方法および航行監視装置
CN113870620B (zh) * 2021-10-19 2023-07-21 遨海科技有限公司 一种多ais设备同时开启的船舶识别方法
US20230222919A1 (en) * 2022-01-11 2023-07-13 Shanghai Maritime University Method for vessel traffic pattern recognition via data quality control and data compression
CN114743406B (zh) * 2022-03-11 2023-04-07 中国电子科技集团公司第五十四研究所 一种船舶轨迹去纠缠方法
CN115294802A (zh) * 2022-07-25 2022-11-04 中远海运科技股份有限公司 一种基于ais数据的船舶航行状态智能识别方法及系统
CN115359807B (zh) * 2022-10-21 2023-01-20 金叶仪器(山东)有限公司 一种用于城市噪声污染的噪声在线监测系统
CN116527778B (zh) * 2023-06-26 2023-09-26 广州中海电信有限公司 一种船舶ais数据的压缩方法
CN117073680A (zh) * 2023-07-28 2023-11-17 武汉理工大学 一种船舶航行轨迹修复方法、电子设备及存储介质
CN117215250B (zh) * 2023-11-07 2024-01-30 山东德源电力科技股份有限公司 一种用于磁控开关的远程控制方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20180117933A (ko) * 2017-04-20 2018-10-30 대우조선해양 주식회사 대상 선박의 운항정보 분석 장치 및 방법
KR20200011125A (ko) * 2018-07-24 2020-02-03 대우조선해양 주식회사 환경 하중을 고려한 선박의 속도 증감량 도출 시스템 및 방법과, 동 방법을 컴퓨터에서 실행하기 위한 컴퓨터 프로그램이 기록된, 컴퓨터 판독 가능한 기록 매체
CN110990504A (zh) * 2019-11-14 2020-04-10 中国船舶重工集团公司第七0七研究所 一种基于航向和航速变化率的船舶轨迹压缩方法
CN111291149A (zh) * 2020-02-28 2020-06-16 中国人民解放军国防科技大学 基于航向偏差的船舶ais时空轨迹分段及模式提取方法
CN111353010A (zh) * 2020-02-21 2020-06-30 上海海事大学 一种自适应阈值的船舶轨迹压缩方法
CN111985529A (zh) * 2020-07-06 2020-11-24 东南大学 一种船舶ais轨迹混合聚类方法及系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10048075B2 (en) * 2013-07-19 2018-08-14 Sap Se Trajectory data compression
US9727976B1 (en) * 2015-09-08 2017-08-08 Sandia Corporation Geospatial-temporal semantic graph representations of trajectories from remote sensing and geolocation data
US10891318B2 (en) * 2019-02-22 2021-01-12 United States Of America As Represented By The Secretary Of The Navy Temporal logic fusion of real time data

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20180117933A (ko) * 2017-04-20 2018-10-30 대우조선해양 주식회사 대상 선박의 운항정보 분석 장치 및 방법
KR20200011125A (ko) * 2018-07-24 2020-02-03 대우조선해양 주식회사 환경 하중을 고려한 선박의 속도 증감량 도출 시스템 및 방법과, 동 방법을 컴퓨터에서 실행하기 위한 컴퓨터 프로그램이 기록된, 컴퓨터 판독 가능한 기록 매체
CN110990504A (zh) * 2019-11-14 2020-04-10 中国船舶重工集团公司第七0七研究所 一种基于航向和航速变化率的船舶轨迹压缩方法
CN111353010A (zh) * 2020-02-21 2020-06-30 上海海事大学 一种自适应阈值的船舶轨迹压缩方法
CN111291149A (zh) * 2020-02-28 2020-06-16 中国人民解放军国防科技大学 基于航向偏差的船舶ais时空轨迹分段及模式提取方法
CN111985529A (zh) * 2020-07-06 2020-11-24 东南大学 一种船舶ais轨迹混合聚类方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Doctoral Dissertation", 15 April 2021, DALIAN MARITIME UNIVERSITY, CN, article PENG CHEN: "Research on the Knowledge Base of Ship Collision Avoidance Based on HSSVM and Convolutional Neural Networks", pages: 1 - 150, XP093009746, DOI: 10.26989/d.cnki.gdlhu.2021.000002 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116309708A (zh) * 2023-02-28 2023-06-23 武汉理工大学 一种锚地时空利用效率评价方法和装置
CN116309708B (zh) * 2023-02-28 2023-10-27 武汉理工大学 一种锚地时空利用效率评价方法和装置

Also Published As

Publication number Publication date
US20230118438A1 (en) 2023-04-20
US11851147B2 (en) 2023-12-26
CN113450596B (zh) 2022-07-15
CN113450596A (zh) 2021-09-28

Similar Documents

Publication Publication Date Title
WO2022252398A1 (zh) 基于船舶轨迹特征点提取的时空dp方法
CN109190636B (zh) 一种遥感图像舰船目标信息提取方法
CN110378308B (zh) 改进的基于Faster R-CNN的港口SAR图像近岸舰船检测方法
CN105931295B (zh) 一种地质图专题信息提取方法
CN103546667B (zh) 一种面向海量广播电视监管的自动新闻拆条方法
CN108415975B (zh) 基于bdch-dbscan的出租车载客热点识别方法
CN109325418A (zh) 基于改进YOLOv3的道路交通环境下行人识别方法
CN113553682B (zh) 一种数据驱动的多层次船舶航路网络构建方法
CN109708638B (zh) 一种船舶轨迹点提取方法
CN103632146B (zh) 一种基于头肩距的人体检测方法
CN103578119A (zh) 基于超像素的Codebook动态场景中目标检测方法
CN112906830A (zh) 一种基于ais大数据的船舶最优航线自动生成方法
CN106323301A (zh) 一种道路情报的获取方法及装置
CN111291149A (zh) 基于航向偏差的船舶ais时空轨迹分段及模式提取方法
CN112100435B (zh) 一种基于边缘端交通音视频同步样本的自动标注方法
CN113932821B (zh) 基于连续窗口平均方向特征的轨迹地图匹配方法
CN113763427B (zh) 一种基于从粗到精遮挡处理的多目标跟踪方法
CN109840358A (zh) 一种基于航迹时域差分的航迹分段方法
CN110502596A (zh) 一种基于行人轨迹特征的轨迹在线滑动窗口压缩方法
CN114564545A (zh) 一种基于ais历史数据的船舶经验航线提取系统及方法
CN111681259A (zh) 基于无Anchor机制检测网络的车辆跟踪模型建立方法
CN109800714A (zh) 一种基于人工智能的船舶检测系统及方法
CN106204650A (zh) 一种基于空地视频关联技术的车辆目标跟踪方法
US20230222919A1 (en) Method for vessel traffic pattern recognition via data quality control and data compression
CN110490150A (zh) 一种基于车辆检索的违章图片自动审核系统及方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21943753

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE