CN111949750A - Ship track model building and abnormal track detection method - Google Patents
Ship track model building and abnormal track detection method Download PDFInfo
- Publication number
- CN111949750A CN111949750A CN202010754360.6A CN202010754360A CN111949750A CN 111949750 A CN111949750 A CN 111949750A CN 202010754360 A CN202010754360 A CN 202010754360A CN 111949750 A CN111949750 A CN 111949750A
- Authority
- CN
- China
- Prior art keywords
- track
- ship
- points
- difference
- point
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 50
- 238000001514 detection method Methods 0.000 title claims description 7
- 238000000034 method Methods 0.000 claims abstract description 33
- 239000011159 matrix material Substances 0.000 claims abstract description 24
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 101100152865 Danio rerio thraa gene Proteins 0.000 claims description 5
- 101150117196 tra-1 gene Proteins 0.000 claims description 5
- 101100326684 Caenorhabditis elegans tra-3 gene Proteins 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 101150058668 tra2 gene Proteins 0.000 claims description 3
- 239000002904 solvent Substances 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 10
- 230000005856 abnormality Effects 0.000 description 6
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 238000004873 anchoring Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Probability & Statistics with Applications (AREA)
- Remote Sensing (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Navigation (AREA)
Abstract
The invention relates to a ship track model building and abnormal track detecting method, which comprises the following steps: s1, acquiring an original track data set of the ship, and preprocessing the original track data set to obtain a track data set; s2, calculating course difference, speed difference and average distance between two tracks based on the track data set, and sequentially giving corresponding weights; s3, repeating the step S2 for multiple times, calculating the similarity value of every two tracks through a similarity value formula, and generating a similarity matrix; s4, dividing the track data set into a plurality of track data subsets according to the similarity matrix; s5, establishing a ship track model according to the track data subset; and S6, judging whether the ship behavior state is normal or not according to the ship track model. The invention comprehensively considers factors such as course speed and the like and judges the similarity among the tracks according to the factors, thereby effectively improving the judgment precision, and then establishes a ship track model based on the similarity matrix, thereby providing a data basis for the virtual simulation of the ship and further judging whether the navigation state of the ship is normal.
Description
Technical Field
The invention relates to the technical field of ship data processing, in particular to a ship track model building and abnormal track detection method.
Background
When the intelligent ship executes a navigation task or after the navigation task is finished, the ship database stores a large amount of ship navigation data, including ship track data. And the sailing state of the ship can be determined through the analysis and processing of the ship track data. When ship track data is analyzed, data clustering analysis is needed, the clustering analysis is mainly carried out according to differences among objects, similar objects are aggregated into clusters, so that similarity measurement among tracks is needed, the precision of clustering results can be guaranteed through efficient and reasonable similarity measurement among tracks, and the reliability of the clustering results cannot be guaranteed through a single reference basis in the conventional track measurement method.
Disclosure of Invention
Technical problem to be solved
In view of the above disadvantages and shortcomings of the prior art, the present invention provides a method for establishing a ship track model and detecting an abnormal track, which solves the technical problems that the accuracy of a ship data clustering result is not high and the abnormal track cannot be effectively monitored according to the clustering result.
(II) technical scheme
In order to achieve the purpose, the invention adopts the main technical scheme that:
s1, acquiring an original track data set of the ship, and preprocessing the original track data set to obtain a track data set; the preprocessing comprises invalid data removing, missing data filling and data normalization;
s2, calculating course difference, speed difference and average distance between two tracks based on the track data set, and respectively endowing corresponding weights to the course difference, the speed difference and the average distance according to weight distribution conditions;
s3, repeating the step S2 for multiple times, calculating the similarity value of every two tracks through a similarity value formula, and generating a similarity matrix;
s4, dividing the track data set into a plurality of track data subsets according to behavior characteristics according to the similarity matrix;
s5, establishing a ship track model according to the track data subset;
and S6, adding the ship track model into a ship knowledge base for comparing with actual navigation track data to judge whether the ship behavior state is normal.
Optionally, step S2 includes:
s21, dividing the track into a plurality of track segments based on the angle change of the track point direction;
s22, calculating the course difference of every two track segments through a course difference formula; calculating the speed difference of every two track sections through a speed difference formula; calculating the average distance between all track points of every two track segments by an average distance formula;
and S23, respectively distributing a first weight, a second weight and a third weight for the heading difference, the speed difference and the average distance according to weight distribution conditions.
Optionally, the heading difference formula is:
wherein, cmax(l1-l2) Difference value representing maximum course of track point in two track segments, cmin(l1-l2) Difference value representing minimum course of track point in two track segments, cavg(l1-l2) Representing the difference value of the average course of the track points in the two track segments;
the speed difference formula is as follows:
wherein v ismax(l1-l2) Representing the difference, v, of the maximum navigational speeds of the track points in the two track segmentsmin(l1-l2) Representing the difference, v, of the minimum navigational speeds of the track points in the two track segmentsavg(l1-l2) Representing the difference value of the average navigational speeds of the track points in the two track sections;
the average distance formula is:
wherein, dis (l)1(i)-l2(i) Represents the distance between the ith track points in the two track segments, and n represents the number of the track points;
the weight distribution condition is as follows: distributing weights according to the weight proportion occupied by the three factors of the course difference, the speed difference and the average distance in similarity judgment; in particular, the amount of the solvent to be used,
optionally, the similarity formula is:
wherein, sim (l)1,l2) Representing the similarity of the ith and jth tracks, djRepresenting the sum of the heading differences of all track segments, the sum of the speed differences of all track segments and the sum of the values of the average distances of all track segments,representing the first weight, the second weight, and the third weight;
the similarity matrix is an n-x-n matrix, and n represents the number of tracks in the track data set.
Optionally, step S4 includes:
s41, sorting the tracks in the track data set;
s42, adding a first track in the track data set into a first track data subset, traversing the track data set, and adding the track data of which the track similarity value with the first track data is greater than a similarity threshold into the first track data subset;
and S43, repeating the step S42, and dividing the track data set into a plurality of track data subsets.
Optionally, step S5 includes:
s51, determining clustering tracks in each track data subset;
s52, taking the clustering track points on the clustering track as base points, making a plurality of vertical lines, wherein each vertical line has an intersection point with other track data except the clustering track; the track points comprise ship position information, course information and navigational speed information;
s53, finding track points closest to intersection points under the same vertical line, and calculating the average value of ship position information, course information and navigational speed information of the track points to obtain a plurality of characteristic track points;
and S54, connecting the characteristic track points to obtain a ship track model.
Optionally, the ship trajectory model is:
model={tral,tra2,tra3...,traN}
traN={latN,lonN,cogN,sogN}
the ship track model is composed of N track points which are total of tra1 … traN, the information of each track point comprises ship position information (lat, lon), ship course information (cog) and ship speed information (sog), wherein J represents the number of track data in a ship track data subset.
Optionally, step S6 includes:
s61, comparing each actual track point with each characteristic track point in the track model, and if the actual track point meets the abnormal judgment criterion, determining the actual track point as an abnormal track point;
s62, judging whether the proportion of the number of the abnormal track points of each actual track to the number of the total track points exceeds 0.25;
s63a, if the result exceeds the preset value, determining that the track is an abnormal track;
and S63b, if not, the track is not an abnormal track.
Optionally, the abnormality judgment standard includes a position abnormality judgment standard and a heading and speed comprehensive abnormality judgment standard;
the position abnormity judgment standard is as follows:
wherein, P is an actual track point, PN is a characteristic track point, LAT is a latitude value corresponding to the corresponding point, LON is a longitude value corresponding to the corresponding point, DIST represents the Euclidean distance between the two, and when the value is larger than a specific threshold value, the fact that the actual track point has position abnormity is represented;
the comprehensive course and speed abnormity judgment standard is as follows:
in the formula (1), A is the difference value of the actual track point and the characteristic track point in the heading direction, and the value range is [0,180 ]; MIN (S.P, S.PN) represents the minimum value of the navigation speed in the actual track point and the characteristic track point, MAX (S.P, S.PN) represents the maximum value of the navigation speed in the actual track point and the characteristic track point; SPE represents the comprehensive abnormal condition of the heading and speed of the actual track points and the characteristic track points, the value range is [ -1,1], the larger the SPE value is, the higher the similarity between the two track points is, and the smaller the value is, the higher the abnormal degree of the actual track points is;
and when the formula (2) is satisfied, the actual track point is judged to be an abnormal track point.
(III) advantageous effects
The invention has the beneficial effects that: the method comprehensively considers the factors such as the course speed of the ship track and the like to judge the similarity of the ship track, effectively improves the judgment precision, provides a clustering basis for the track clustering analysis algorithm and ensures the reliability of the clustering algorithm. And a ship track model is established based on the clustering result and added into a ship knowledge base, so that a data basis can be provided for the virtual simulation of the ship, and whether the navigation state of the ship is normal or not can be judged. Then, the abnormal track points are judged through the ship track model, so that whether the abnormal track points exist in a certain section of actual track or not and whether the abnormal track points exist in the section of track or not can be determined, and an accurate theoretical basis is provided for the simulation process of the intelligent ship.
Drawings
FIG. 1 is a schematic flow chart of a ship track model building and abnormal track detection method provided by the invention;
fig. 2 is a schematic flowchart of step S2 of the method for establishing and detecting a track model of an intelligent ship according to the present invention;
FIG. 3 is a schematic diagram of a similarity matrix for a method for building and detecting a track model of an intelligent ship according to the present invention;
fig. 4 is a schematic flowchart of step S4 of the method for establishing and detecting a track model of an intelligent ship according to the present invention;
fig. 5 is a schematic flowchart of step S5 of the method for establishing and detecting a track model of an intelligent ship according to the present invention;
FIG. 6 is a schematic diagram of a clustering track and a typical track for a method for establishing and detecting a track model of an intelligent ship according to the present invention;
fig. 7 is a detailed flowchart of step S6 of the method for establishing and detecting a smart ship trajectory model according to the present invention.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
Fig. 1 is a schematic flow diagram of a ship track model building and abnormal track detecting method provided by the present invention, and as shown in fig. 1, the ship track model building and abnormal track detecting method discloses: firstly, acquiring an original track data set of a ship, and obtaining the track data set through a series of preprocessing; then, calculating course difference, navigational speed difference and average distance between every two tracks, distributing different weights according to weight distribution conditions, calculating similarity values of every two tracks through a similarity value formula based on the course difference, the navigational speed difference, the average distance and the weights, and generating a similarity matrix; clustering the track data according to the similarity matrix, namely dividing the track data set into a plurality of track data subsets according to behavior characteristics, and establishing a ship track model based on the track data subsets; and adding the ship track model into a ship knowledge base for comparing with actual navigation track data to judge whether the ship behavior state is normal or not.
The method comprehensively considers the factors such as the course speed of the ship track and the like to judge the similarity of the ship track, effectively improves the judgment precision, provides a clustering basis for the track clustering analysis algorithm and ensures the reliability of the clustering algorithm. And a ship track model is established based on the clustering result and added into a ship knowledge base, so that a data basis can be provided for the virtual simulation of the ship, and whether the navigation state of the ship is normal or not can be judged. Then, the abnormal track points are judged through the ship track model, so that whether the abnormal track points exist in a certain section of actual track or not and whether the abnormal track points exist in the section of track or not can be determined, and an accurate theoretical basis is provided for the simulation process of the intelligent ship.
For a better understanding of the above-described technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
Specifically, the method for establishing and detecting the intelligent ship track model comprises the following steps:
and S1, acquiring an original track data set of the ship from the ship track database, and preprocessing the original track data set to obtain a track data set. The preprocessing comprises invalid data removing, data noise filtering, missing data filling and data normalization, and aims to ensure the reliability of data.
S2, calculating the course difference, the speed difference and the average distance between the two tracks based on the track data set, and endowing corresponding weights for the course difference, the speed difference and the average distance value according to the weight distribution strip. Fig. 2 is a schematic flowchart of a step S2 of the method for establishing and detecting a track model of an intelligent ship according to the present invention, and as shown in fig. 2, the following steps are the specific steps of step S2:
and S21, dividing the track into track segments based on the angle change of the track point direction. If the change exceeds a certain angle change threshold, the trajectory is segmented.
S22, calculating the course difference of every two track segments through a course difference formula; calculating the speed difference of every two track sections through a speed difference formula; and calculating the distance between all track points of every two track segments by using an average distance formula. The course characteristics comprise a maximum course difference, a minimum course difference and an average course difference, specifically, the course difference of the two track segments is calculated through the course characteristics of the track segments, the maximum course difference and the minimum course difference of the two track segments are integrated, the average course difference is used as the course difference of the two track segments, and the course difference formula is as follows:
wherein, cmax(l1-l2) Difference value representing maximum course of track point in two track segments, cmin(l1-l2) Difference value representing minimum course of track point in two track segments, cavg(l1-l2) Representing the difference value of the average course of the track points in the two track segments;
and calculating the navigational speed distance between the two tracks. The speed characteristics comprise maximum speed difference, minimum speed difference and average speed difference, the speed distance of the two track sections is calculated through the speed characteristics of the track sections, the maximum speed difference and the minimum speed difference of the two track sections are synthesized, the average speed difference is used as the speed distance of the two track sections, and the speed difference formula is as follows:
wherein v ismax(l1-l2) Representing the difference, v, of the maximum navigational speeds of the track points in the two track segmentsmin(l1-l2) Representing the difference, v, of the minimum navigational speeds of the track points in the two track segmentsavg(l1-l2) Representing the difference value of the average navigational speeds of the track points in the two track sections;
and calculating the track average distance. Sequentially calculating the distances between all track points of the two track segments according to the sequence, and judging the similarity of the two tracks by the average value of the distances, wherein the average distance formula is as follows:
wherein, dis (l)1(i)-l2(i) Represents the distance between the ith trace point in the two trace segments and n represents the number of trace points.
And S23, assigning a first weight, a second weight and a third weight to the heading difference, the speed difference and the average distance value according to the weight distribution condition. The weight distribution is mainly based on the weight proportion occupied by three factors of course difference, speed difference and average distance in similarity judgment, the value of the weight distribution is determined according to the navigation state and the empirical value, but the value range of each weight value is (0, 1) and the sum of the three is 1. The weight distribution condition is as follows:
and S3, repeating the step S2 for multiple times, calculating the similarity value of every two tracks through a similarity value formula according to the heading difference, the speed difference, the average distance value and the weight, and generating a similarity matrix. The similarity formula is:
wherein, sim (l)1,l2) Representing the similarity of the ith and jth tracks, d1Representing the sum of course differences of all track segments, d2Sum of speed differences of all track segments, d3The sum of the values of the average distances of all track segments,representing a first weight, a second weight and a third weight.
Fig. 3 is a schematic diagram of a similarity matrix for the method for building and detecting a track model of an intelligent ship according to the present invention, and as shown in fig. 3, the similarity matrix is an n × n matrix, where n represents the number of tracks in a track data set.
And S4, dividing the track data set into a plurality of track data subsets according to the behavior characteristics according to the similarity matrix. The behavior characteristics refer to the driving states of the ship, generally including the states of departure, berthing, anchoring, constant-speed driving, variable-speed driving, collision avoidance and the like. Fig. 4 is a schematic specific flowchart of step S4 of the method for establishing and detecting a track model of an intelligent ship according to the present invention, and as shown in fig. 4, the following is a specific flowchart of step S4:
and S41, sorting the tracks in the track data set.
And S42, adding a first track in the track data set into the first track data subset, traversing the track data set, and adding the track data with the track similarity value larger than the similarity threshold value with the first track data into the first track data subset. Specifically, a first track in the ship track data set is added into a first track data subset, the track data set is traversed, and tracks with the track similarity value larger than a track similarity threshold are added into the first track data subset. And then adding the first track in the rest track data set into the second track data subset, traversing the rest track data set, and adding the tracks with the track similarity value larger than the track similarity threshold into the second track data subset. By analogy, the ship track data set can be divided into a plurality of track subsets with approximate behavior characteristics, and track clustering is completed.
And S43, repeating the step S42, and dividing the track data set into a plurality of track data subsets.
And S5, establishing a ship track model according to the track data subset. By analyzing the behavior characteristics of the data in the same ship track data subset, a virtual ship track model can be obtained, and the virtual ship track model can represent the normal running state of a ship in the navigation scene.
In order to avoid the influence of local abnormal points in ship track modeling, a ship track model mainly selects a series of ship track points with typical representatives, and specifically, when a certain ship track characteristic is modeled, a series of characteristic points are constructed in a corresponding ship track cluster to represent the ship track cluster. Fig. 5 is a detailed flowchart of step S5 of the method for establishing and detecting an intelligent ship track model according to the present invention, and as shown in fig. 5, the feature points in the ship track cluster are constructed through the following steps:
s51 and fig. 6 are schematic diagrams of a clustering track and a typical track for the intelligent ship track model establishing and detecting method provided by the present invention, as shown in fig. 6, a clustering track in each track data subset is determined, in fig. 6, the clustering track is track 2, and track 1 and track 3 are both conventional tracks.
S52, taking the clustering track points on the clustering track as base points, making a plurality of vertical lines, wherein each vertical line has an intersection point with other track data except the clustering track data; the track points comprise ship position information, course information and navigational speed information.
And S53, finding track points closest to the intersection points under the same vertical line, and calculating the average value of the ship position information, the course information and the navigational speed information of each track point to obtain a plurality of characteristic track points.
And S54, connecting the characteristic track points to obtain a typical track, namely the established ship track model under the scene. The ship track model is as follows:
model={tra1,tra2,tra3...,traN}
traN={latN,lonN,cogN,sogN}
the ship track model consists of N track points which are total to tra1 … traN, and the information of each track point comprises ship position information (lat, lon), ship heading information (cog) and ship speed information (sog). Wherein J represents the number of trajectory data in the subset of vessel trajectory data.
And S6, adding the ship track model into a ship knowledge base for comparing with actual navigation track data to judge whether the ship behavior state is normal. The navigation track data collected in the actual navigation process of the ship can be used for detecting the real-time track by comparing with the track model, and the detection comprises the extraction of navigation state abnormity and abnormal track points. Fig. 7 is a schematic specific flowchart of step S6 of the method for establishing and detecting a track model of an intelligent ship according to the present invention, and as shown in fig. 7, the following is a specific flowchart of step S6:
and S61, comparing each actual track point with each characteristic track point in the track model, and if the actual track point meets the abnormity judgment criterion, determining the actual track point as an abnormal track point.
The abnormality judgment standard comprises a position abnormality judgment standard and a course speed comprehensive abnormality judgment standard;
the position abnormity judgment standard is as follows:
wherein, P is the actual track point, PN is the characteristic track point, LAT is the corresponding latitude value of corresponding point, lon is the corresponding longitude value of corresponding point, DIST represents the Euclidean distance between the two, when this value is greater than specific threshold value, represent that there is the position anomaly in actual track point.
The comprehensive abnormal judgment standard of course speed is as follows:
in the formula (1), A is the difference value of the actual track point and the characteristic track point in the heading direction, and the value range is [0,180 ]; MIN (S.P, S.PN) represents the minimum value of the navigation speed in the actual track point and the characteristic track point, MAX (S.P, S.PN) represents the maximum value of the navigation speed in the actual track point and the characteristic track point; SPE represents the comprehensive abnormal condition of the heading and speed of the actual track points and the characteristic track points, the value range is [ -1,1], the larger the SPE value is, the higher the similarity between the two track points is, and the smaller the value is, the higher the abnormal degree of the actual track points is.
And when the formula (2) is satisfied, the actual track point is judged to be an abnormal track point.
And S62, judging whether the proportion of the number of the abnormal track points of each actual track to the number of the total track points exceeds 0.25.
S63a, exceed, is the abnormal track.
And S63b, not exceeding, not being an abnormal track.
In the specific embodiment of establishing the state track model of the ship sailing away from a certain port, firstly, ship track data from each departure of the ship to a certain position is collected to form a ship track data set; then, data preprocessing is performed. Preprocessing the extracted track set, including removing invalid data, filling missing data and the like to ensure the reliability of the data to be subjected to clustering analysis; establishing a track similarity matrix, calculating the similarity value between every two tracks according to the ship track similarity measurement method, and establishing a similarity matrix; thirdly, completing ship track clustering, realizing ship track data clustering according to the established ship track similarity matrix, and dividing an original data set into a plurality of track data subsets with the same characteristic points; then, establishing a ship departure track model, finding a track data subset meeting the requirement from all the completed track data subsets, and establishing a typical ship departure track model; finally, warehousing the model, adding the established ship departure track model into a ship knowledge base, and judging whether the ship navigation state is abnormal or not by comparing the model data with the model data every time the ship departs from the port; and in the process of ship virtual simulation, carrying out virtual ship departure simulation through the track model.
In summary, the invention obtains the original track data set of the ship, eliminates invalid data through preprocessing, fills in missing data, and normalizes the data for the convenience of subsequent calculation, and finally obtains the track data set capable of effectively ensuring track clustering precision. And then, calculating the similarity value of every two tracks, generating a similarity matrix according to the similarity value, and dividing the track data set into a plurality of track data subsets according to behavior characteristics according to the similarity matrix. And then, establishing a ship track model based on the track data subset, and simultaneously adding the ship track model into a ship knowledge base for comparing with actual navigation track data to judge whether the ship behavior state is normal.
According to the ship track data characteristics, the ship track similarity is judged by combining multiple factors, and the track judgment precision is improved. And the ship track data are subjected to clustering analysis by taking the track similarity as a basis, so that the clustering precision is improved, and the reliability of a result is ensured. And establishing a ship track model according to the clustering result, wherein the obtained model has high reference degree and better accords with the actual running condition of the ship. By applying the track model, the behavior state of the ship can be researched, the abnormal behavior of the ship can be automatically detected, and a theoretical basis is provided for the simulation process of the intelligent ship.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions.
It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the terms first, second, third and the like are for convenience only and do not denote any order. These words are to be understood as part of the name of the component.
Furthermore, it should be noted that in the description of the present specification, the description of the term "one embodiment", "some embodiments", "examples", "specific examples" or "some examples", etc., means that a specific feature, structure, material or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, the claims should be construed to include preferred embodiments and all changes and modifications that fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include such modifications and variations.
Claims (9)
1. A ship track model building and abnormal track detection method comprises the following steps:
s1, acquiring an original track data set of the ship, and preprocessing the original track data set to obtain a track data set; the preprocessing comprises invalid data removing, missing data filling and data normalization;
s2, calculating course difference, speed difference and average distance between two tracks based on the track data set, and respectively endowing corresponding weights to the course difference, the speed difference and the average distance according to weight distribution conditions;
s3, repeating the step S2 for multiple times, calculating the similarity value of every two tracks through a similarity value formula, and generating a similarity matrix;
s4, dividing the track data set into a plurality of track data subsets according to behavior characteristics according to the similarity matrix;
s5, establishing a ship track model according to the track data subset;
and S6, adding the ship track model into a ship knowledge base for comparing with actual navigation track data to judge whether the ship behavior state is normal.
2. The method for building a ship track model and detecting abnormal tracks according to claim 1, wherein step S2 includes:
s21, dividing the track into a plurality of track segments based on the angle change of the track point direction;
s22, calculating the course difference of every two track segments through a course difference formula; calculating the speed difference of every two track sections through a speed difference formula; calculating the average distance between all track points of every two track segments by an average distance formula;
and S23, respectively distributing a first weight, a second weight and a third weight for the heading difference, the speed difference and the average distance according to weight distribution conditions.
3. The ship track model building and abnormal track detecting method according to claim 2,
the course difference formula is as follows:
wherein, cmax(l1-l2) Difference value representing maximum course of track point in two track segments, cmin(l1-l2) Difference value representing minimum course of track point in two track segments, cavg(l1-l2) Representing the difference value of the average course of the track points in the two track segments;
the speed difference formula is as follows:
wherein v ismax(l1-l2) Representing the difference, v, of the maximum navigational speeds of the track points in the two track segmentsmin(l1-l2) Representing the difference, v, of the minimum navigational speeds of the track points in the two track segmentsavg(l1-l2) Representing the difference value of the average navigational speeds of the track points in the two track sections;
the average distance formula is:
wherein, dis (l)1(i)-l2(i) Represents the distance between the ith track points in the two track segments, and n represents the number of the track points;
the weight distribution condition is as follows: distributing weights according to the weight proportion occupied by the three factors of the course difference, the speed difference and the average distance in similarity judgment; in particular, the amount of the solvent to be used,
4. the ship track model building and abnormal track detecting method according to claim 3,
the similarity formula is as follows:
wherein, sim (l)1,l2) Representing the similarity of the ith and jth tracks, djRepresenting the sum of the heading differences of all track segments, the sum of the speed differences of all track segments and the sum of the values of the average distances of all track segments,representing the first weight, the second weight, and the third weight;
the similarity matrix is an n-x-n matrix, and n represents the number of tracks in the track data set.
5. The method for building a ship track model and detecting abnormal tracks according to claim 1, wherein step S4 includes:
s41, sorting the tracks in the track data set;
s42, adding a first track in the track data set into a first track data subset, traversing the track data set, and adding the track data of which the track similarity value with the first track data is greater than a similarity threshold into the first track data subset;
and S43, repeating the step S42, and dividing the track data set into a plurality of track data subsets.
6. The method for building a ship track model and detecting abnormal tracks according to claim 1, wherein step S5 includes:
s51, determining clustering tracks in each track data subset;
s52, taking the clustering track points on the clustering track as base points, making a plurality of vertical lines, wherein each vertical line has an intersection point with other track data except the clustering track; the track points comprise ship position information, course information and navigational speed information;
s53, finding track points closest to intersection points under the same vertical line, and calculating the average value of ship position information, course information and navigational speed information of the track points to obtain a plurality of characteristic track points;
and S54, connecting the characteristic track points to obtain a ship track model.
7. The method for building the ship track model and detecting the abnormal track according to claim 6, wherein the ship track model comprises:
model={tra1,tra2,tra3...,traN}
traN={latN,lonN,cogN,sogN}
the ship track model is composed of N track points which are total of tra1 … traN, the information of each track point comprises ship position information (lat, lon), ship course information (cog) and ship speed information (sog), wherein J represents the number of track data in a ship track data subset.
8. The ship track model building and abnormal track detection method as claimed in claim 1, wherein step S6 includes
S61, comparing each actual track point with each characteristic track point in the track model, and if the actual track point meets the abnormal judgment criterion, determining the actual track point as an abnormal track point;
s62, judging whether the proportion of the number of the abnormal track points of each actual track to the number of the total track points exceeds 0.25;
s63a, if the result exceeds the preset value, determining that the track is an abnormal track;
and S63b, if not, the track is not an abnormal track.
9. The method according to claim 8, wherein the anomaly determination criteria include a location anomaly determination criterion and a course speed synthetic anomaly determination criterion;
the position abnormity judgment standard is as follows:
wherein, P is an actual track point, PN is a characteristic track point, LAT is a latitude value corresponding to the corresponding point, LON is a longitude value corresponding to the corresponding point, DIST represents the Euclidean distance between the two, and when the value is larger than a specific threshold value, the fact that the actual track point has position abnormity is represented;
the comprehensive course and speed abnormity judgment standard is as follows:
in the formula (1), A is the difference value of the actual track point and the characteristic track point in the heading direction, and the value range is [0,180 ]; MIN (S.P, S.PN) represents the minimum value of the navigation speed in the actual track point and the characteristic track point, MAX (S.P, S.PN) represents the maximum value of the navigation speed in the actual track point and the characteristic track point; SPE represents the comprehensive abnormal condition of the heading and speed of the actual track points and the characteristic track points, the value range is [ -1,1], the larger the SPE value is, the higher the similarity between the two track points is, and the smaller the value is, the higher the abnormal degree of the actual track points is;
and when the formula (2) is satisfied, the actual track point is judged to be an abnormal track point.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010754360.6A CN111949750B (en) | 2020-07-30 | 2020-07-30 | Ship track model building and abnormal track detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010754360.6A CN111949750B (en) | 2020-07-30 | 2020-07-30 | Ship track model building and abnormal track detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111949750A true CN111949750A (en) | 2020-11-17 |
CN111949750B CN111949750B (en) | 2024-06-11 |
Family
ID=73338317
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010754360.6A Active CN111949750B (en) | 2020-07-30 | 2020-07-30 | Ship track model building and abnormal track detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111949750B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112464951A (en) * | 2020-11-24 | 2021-03-09 | 中国水产科学研究院南海水产研究所 | Fishery safety emergency disposal system and method based on Internet of things |
CN112465041A (en) * | 2020-12-01 | 2021-03-09 | 大连海事大学 | AIS data quality evaluation method based on analytic hierarchy process |
CN112556698A (en) * | 2020-12-17 | 2021-03-26 | 广州博进信息技术有限公司 | Ship behavior change and abnormality recognition method, equipment and medium |
CN112907632A (en) * | 2021-03-11 | 2021-06-04 | 三亚海兰寰宇海洋信息科技有限公司 | Single-towing ship target identification method and device |
CN112947516A (en) * | 2021-02-02 | 2021-06-11 | 三亚海兰寰宇海洋信息科技有限公司 | Ship motion state discrimination method and system |
CN113032502A (en) * | 2021-02-09 | 2021-06-25 | 北京工业大学 | Ship anomaly detection method based on improved track segment DBSCAN clustering |
CN113239025A (en) * | 2021-04-23 | 2021-08-10 | 四川大学 | Ship track classification method based on feature selection and hyper-parameter optimization |
CN113448953A (en) * | 2021-06-03 | 2021-09-28 | 宁波大学 | AIS data processing method |
CN114219039A (en) * | 2021-12-17 | 2022-03-22 | 杭州电子科技大学 | Ship track compression method and system based on ship behavior characteristics |
CN114357322A (en) * | 2022-03-16 | 2022-04-15 | 交通运输部规划研究院 | Ship track anomaly detection method based on space-time decomposition |
CN114494358A (en) * | 2022-04-07 | 2022-05-13 | 中航信移动科技有限公司 | Data processing method, electronic equipment and storage medium |
CN115438247A (en) * | 2022-06-23 | 2022-12-06 | 山东天地通数码科技有限公司 | Track-based one-ship multi-bucket distinguishing method, device and equipment |
CN115630133A (en) * | 2022-12-22 | 2023-01-20 | 亿海蓝(北京)数据技术股份公司 | Abnormal ship searching method and device based on track knowledge graph |
CN116309708A (en) * | 2023-02-28 | 2023-06-23 | 武汉理工大学 | Anchor ground space-time utilization efficiency evaluation method and device |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103593430A (en) * | 2013-11-11 | 2014-02-19 | 胡宝清 | Clustering method based on mobile object spatiotemporal information trajectory subsections |
CN103942533A (en) * | 2014-03-24 | 2014-07-23 | 河海大学常州校区 | Urban traffic illegal behavior detection method based on video monitoring system |
US20160139977A1 (en) * | 2013-07-01 | 2016-05-19 | Agent Video Intelligence Ltd. | System and method for abnormality detection |
WO2017143631A1 (en) * | 2016-02-22 | 2017-08-31 | 江苏大学 | Automatic navigation route tracking method for aquaculture vessel |
CN108804539A (en) * | 2018-05-08 | 2018-11-13 | 山西大学 | A kind of track method for detecting abnormality under time and space double-visual angle |
CN110069585A (en) * | 2017-12-05 | 2019-07-30 | 腾讯科技(深圳)有限公司 | Treating method and apparatus, storage medium and the electronic device of track point data |
CN110175633A (en) * | 2019-04-29 | 2019-08-27 | 杭州电子科技大学 | A kind of fishing boat track method for detecting abnormality based on Beidou satellite transmission data |
CN110363094A (en) * | 2019-06-20 | 2019-10-22 | 珠海云航智能技术有限公司 | A kind of ship abnormal behaviour recognition methods, device and terminal device |
CN111291149A (en) * | 2020-02-28 | 2020-06-16 | 中国人民解放军国防科技大学 | Ship AIS space-time trajectory segmentation and mode extraction method based on course deviation |
-
2020
- 2020-07-30 CN CN202010754360.6A patent/CN111949750B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160139977A1 (en) * | 2013-07-01 | 2016-05-19 | Agent Video Intelligence Ltd. | System and method for abnormality detection |
CN103593430A (en) * | 2013-11-11 | 2014-02-19 | 胡宝清 | Clustering method based on mobile object spatiotemporal information trajectory subsections |
CN103942533A (en) * | 2014-03-24 | 2014-07-23 | 河海大学常州校区 | Urban traffic illegal behavior detection method based on video monitoring system |
WO2017143631A1 (en) * | 2016-02-22 | 2017-08-31 | 江苏大学 | Automatic navigation route tracking method for aquaculture vessel |
CN110069585A (en) * | 2017-12-05 | 2019-07-30 | 腾讯科技(深圳)有限公司 | Treating method and apparatus, storage medium and the electronic device of track point data |
CN108804539A (en) * | 2018-05-08 | 2018-11-13 | 山西大学 | A kind of track method for detecting abnormality under time and space double-visual angle |
CN110175633A (en) * | 2019-04-29 | 2019-08-27 | 杭州电子科技大学 | A kind of fishing boat track method for detecting abnormality based on Beidou satellite transmission data |
CN110363094A (en) * | 2019-06-20 | 2019-10-22 | 珠海云航智能技术有限公司 | A kind of ship abnormal behaviour recognition methods, device and terminal device |
CN111291149A (en) * | 2020-02-28 | 2020-06-16 | 中国人民解放军国防科技大学 | Ship AIS space-time trajectory segmentation and mode extraction method based on course deviation |
Non-Patent Citations (3)
Title |
---|
丁建立;黄天镜;徐俊洁;王静;: "基于时间序列的多维距离聚类异常检测方法", 计算机工程与设计, no. 07, 16 July 2020 (2020-07-16) * |
刘磊: "基于KNN 的船舶轨迹分类算法", 大连海事大学学报, no. 3, 31 August 2018 (2018-08-31), pages 1 * |
胡智辉: "基于轨迹聚类的船舶异常行为识别研究", 中国硕士学位论文全文数据库 工程科技Ⅰ辑, 15 January 2018 (2018-01-15), pages 3 - 4 * |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112464951A (en) * | 2020-11-24 | 2021-03-09 | 中国水产科学研究院南海水产研究所 | Fishery safety emergency disposal system and method based on Internet of things |
CN112464951B (en) * | 2020-11-24 | 2024-01-12 | 中国水产科学研究院南海水产研究所 | Fishery safety emergency disposal system and method based on Internet of things |
CN112465041A (en) * | 2020-12-01 | 2021-03-09 | 大连海事大学 | AIS data quality evaluation method based on analytic hierarchy process |
CN112465041B (en) * | 2020-12-01 | 2024-01-05 | 大连海事大学 | AIS data quality assessment method based on analytic hierarchy process |
CN112556698A (en) * | 2020-12-17 | 2021-03-26 | 广州博进信息技术有限公司 | Ship behavior change and abnormality recognition method, equipment and medium |
CN112556698B (en) * | 2020-12-17 | 2023-12-15 | 广州博进信息技术有限公司 | Ship behavior change and abnormality identification method, equipment and medium |
CN112947516B (en) * | 2021-02-02 | 2022-10-21 | 三亚海兰寰宇海洋信息科技有限公司 | Ship motion state discrimination method and system |
CN112947516A (en) * | 2021-02-02 | 2021-06-11 | 三亚海兰寰宇海洋信息科技有限公司 | Ship motion state discrimination method and system |
CN113032502A (en) * | 2021-02-09 | 2021-06-25 | 北京工业大学 | Ship anomaly detection method based on improved track segment DBSCAN clustering |
CN113032502B (en) * | 2021-02-09 | 2024-02-02 | 北京工业大学 | Ship anomaly detection method based on improved track segment DBSCAN clustering |
CN112907632A (en) * | 2021-03-11 | 2021-06-04 | 三亚海兰寰宇海洋信息科技有限公司 | Single-towing ship target identification method and device |
CN113239025A (en) * | 2021-04-23 | 2021-08-10 | 四川大学 | Ship track classification method based on feature selection and hyper-parameter optimization |
CN113448953A (en) * | 2021-06-03 | 2021-09-28 | 宁波大学 | AIS data processing method |
CN114219039A (en) * | 2021-12-17 | 2022-03-22 | 杭州电子科技大学 | Ship track compression method and system based on ship behavior characteristics |
CN114219039B (en) * | 2021-12-17 | 2024-05-24 | 杭州电子科技大学 | Ship track compression method and system based on ship behavior characteristics |
CN114357322A (en) * | 2022-03-16 | 2022-04-15 | 交通运输部规划研究院 | Ship track anomaly detection method based on space-time decomposition |
CN114494358A (en) * | 2022-04-07 | 2022-05-13 | 中航信移动科技有限公司 | Data processing method, electronic equipment and storage medium |
CN114494358B (en) * | 2022-04-07 | 2022-06-21 | 中航信移动科技有限公司 | Data processing method, electronic equipment and storage medium |
CN115438247B (en) * | 2022-06-23 | 2023-10-10 | 山东天地通数码科技有限公司 | Method, device and equipment for discriminating multiple vessels based on track |
CN115438247A (en) * | 2022-06-23 | 2022-12-06 | 山东天地通数码科技有限公司 | Track-based one-ship multi-bucket distinguishing method, device and equipment |
CN115630133A (en) * | 2022-12-22 | 2023-01-20 | 亿海蓝(北京)数据技术股份公司 | Abnormal ship searching method and device based on track knowledge graph |
CN116309708B (en) * | 2023-02-28 | 2023-10-27 | 武汉理工大学 | Anchor ground space-time utilization efficiency evaluation method and device |
CN116309708A (en) * | 2023-02-28 | 2023-06-23 | 武汉理工大学 | Anchor ground space-time utilization efficiency evaluation method and device |
Also Published As
Publication number | Publication date |
---|---|
CN111949750B (en) | 2024-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111949750B (en) | Ship track model building and abnormal track detection method | |
CN108921083B (en) | Illegal mobile vendor identification method based on deep learning target detection | |
CN103366602A (en) | Method of determining parking lot occupancy from digital camera images | |
CN107330469B (en) | Travel mode identification method based on GPS track data | |
CN112508105A (en) | Method for detecting and retrieving faults of oil extraction machine | |
CN115457395A (en) | Lightweight remote sensing target detection method based on channel attention and multi-scale feature fusion | |
CN106845496B (en) | Fine target identification method and system | |
CN111582380A (en) | Ship track density clustering method and device based on space-time characteristics | |
CN114333423B (en) | Narrow water channel navigation collision risk assessment method | |
CN111783844A (en) | Target detection model training method and device based on deep learning and storage medium | |
CN112329605B (en) | City appearance random pasting and random drawing behavior identification method, storage device and server | |
CN113836999A (en) | Tunnel construction risk intelligent identification method and system based on ground penetrating radar | |
CN115050214B (en) | AIS data-based ship collision risk prediction method | |
CN104038792A (en) | Video content analysis method and device for IPTV (Internet Protocol Television) supervision | |
CN110562261A (en) | Method for detecting risk level of driver based on Markov model | |
CN110263836B (en) | Bad driving state identification method based on multi-feature convolutional neural network | |
CN113255533B (en) | Method for identifying forbidden zone intrusion behavior, storage device and server | |
CN117912301A (en) | Ship abnormal behavior monitoring and alarming method based on isolated forest | |
CN101996312A (en) | Method and device for tracking targets | |
CN106548195A (en) | A kind of object detection method based on modified model HOG ULBP feature operators | |
CN112507867A (en) | Lane line detection method based on EDLines line characteristics | |
CN111882135B (en) | Internet of things equipment intrusion detection method and related device | |
Grigoropoulos et al. | Detection and classification of bicyclist group behavior for automated vehicle applications | |
CN114485681A (en) | Method for evaluating accuracy map data consistency rate by using DR track | |
CN113128568A (en) | Excavator activity identification method, system, device and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |