CN112907632A - Single-towing ship target identification method and device - Google Patents
Single-towing ship target identification method and device Download PDFInfo
- Publication number
- CN112907632A CN112907632A CN202110265932.9A CN202110265932A CN112907632A CN 112907632 A CN112907632 A CN 112907632A CN 202110265932 A CN202110265932 A CN 202110265932A CN 112907632 A CN112907632 A CN 112907632A
- Authority
- CN
- China
- Prior art keywords
- result
- ship
- target
- data
- clustering
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- 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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Abstract
The invention provides a method and a device for identifying a single-towed-vessel target, wherein the method comprises the following steps: acquiring real-time running track data of a ship; preprocessing the real-time driving track data to obtain a preprocessing result; carrying out track clustering on the preprocessing result to obtain a clustering result; extracting the characteristics of the clustering result to obtain characteristic data; and inputting the characteristic data into a preset model for identifying the single-towing ship target for processing to obtain an identification result of the single-towing ship target. The scheme of the invention is based on the motion characteristics of a single-towed ship, abstracts the characteristics of sailing direction change, turning times, segmental straight-going time and the like, does not directly set a threshold value, greatly reduces the false alarm rate and improves the information acquisition capability of illegal behaviors involving sea and fish.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a method and a device for identifying a single-towed-vessel target.
Background
Accurate identification of a single-towing fishing vessel (a fishing vessel for fishing operation by towing a net by the single-towing vessel) can improve information acquisition capacity of illegal behaviors involving sea and fish, and further enhance accurate attack on illegal behaviors involving fish and fine management of bent-season fishing.
The prior art has at least the following problems:
while a typical single-drag operation can last for more than several hours, the conventional rule-based single-drag operation can be used for identifying a track with a short time (1 hour), and cannot completely express the track characteristics of the single-drag operation. The identification based on the rules needs to set a fixed threshold value, has certain limitation, cannot adapt to variable motion tracks, and causes false alarm and missing report.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method and a device for identifying a target of a single-towed ship, abstracting characteristics such as sailing direction change, turning times, segmental straight-ahead time and the like based on the movement characteristics of the single-towed ship, learning a large number of characteristics of single-towed tracks without directly setting a threshold, extracting historical tracks for at least 3 hours, classifying the tracks, greatly reducing the false alarm rate and improving the information acquisition capability of illegal behaviors of sea-fishing-wading.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a method of identifying a single tow vessel target, comprising:
acquiring real-time running track data of a ship;
preprocessing the real-time driving track data to obtain a preprocessing result;
carrying out track clustering on the preprocessing result to obtain a clustering result;
extracting the characteristics of the clustering result to obtain characteristic data;
and inputting the characteristic data into a preset model for identifying the single-towing ship target for processing to obtain an identification result of the single-towing ship target.
Optionally, the preprocessing the real-time driving trajectory data includes:
deleting the real-time running track data of which the number of points of the track is less than a first value in the real-time running track data;
deleting the real-time running track data of which the average speed is less than a second value in the real-time running track data; and/or
And deleting the real-time running track data of which the average speed is greater than a third value in the real-time running track data.
Optionally, performing track clustering on the preprocessing result to obtain a clustering result, including:
performing track clustering on the preprocessing result through a clustering algorithm to obtain an intermediate clustering result;
calculating the moving range of the ship to be identified according to the intermediate clustering result, and acquiring a moving range result;
deleting the active range results that are less than or equal to the first threshold; otherwise, the activity range is reserved to obtain the final clustering result.
Optionally, performing feature extraction on the clustering result to obtain feature data, including:
performing feature extraction on the clustering result to obtain at least one of the following feature data: course change result, turning frequency result, straight-going time result, circle rate result, navigation speed and course.
Optionally, the course change result is the sum of course changes of the ship to be identified, and the course change is the course difference of the ship to be identified at two moments;
the turning times result is the turning times of the ship to be identified;
the straight-going time result is the sectional statistical straight-going time of the ship to be identified;
the circle rate result is the circle rate of the ship to be identified; the circle rate is the total course of the vessel to be identified divided by the displacement.
Optionally, the predictive single tow vessel target model is trained by:
acquiring historical track sample data of a ship;
carrying out data cleaning processing on the historical track sample data of the ship to obtain a cleaning result;
performing data visualization processing on the cleaning result to obtain a visualization result;
performing feature extraction on the visualization result to obtain a feature extraction result;
and training the feature extraction result through a random forest model to obtain the preset model.
Optionally, the parameters of the random forest model include:
n_estimators＝60；
max_depth＝10；
min_samples_split＝10；
wherein n _ estimators is the number of the decision trees, max _ depth is the depth of the decision trees, and min _ samples _ split is the minimum sample number of the node.
Optionally, inputting the feature data into a preset model for identifying the single-towed-vessel target, and processing the feature data to obtain an identification result of the single-towed-vessel target, where the identification result includes:
inputting the characteristic data into a preset model, and outputting a prediction result;
judging the category of the ship to be identified according to the prediction result to obtain a judgment result;
and acquiring a single-towing ship target according to the judgment result.
Optionally, the obtaining a single-towing ship target according to the determination result includes:
and when the judgment result shows that the probability that the ship to be identified is the single-towing ship target is greater than 0.7, acquiring the single-towing ship target.
The embodiment of the invention also provides a device for identifying the target of the single-towed-vessel, which comprises:
the acquisition module is used for acquiring real-time running track data of the ship;
the preprocessing module is used for preprocessing the real-time driving track data to obtain a preprocessing result;
the clustering module is used for carrying out track clustering on the preprocessing result to obtain a clustering result;
the characteristic extraction module is used for extracting the characteristics of the clustering result to obtain characteristic data;
and the identification module is used for inputting the characteristic data into a preset model for identifying the single-towing ship target to be processed, and obtaining the identification result of the single-towing ship target.
The scheme of the invention at least comprises the following beneficial effects:
based on the motion characteristics of a single-towing ship, the characteristics of sailing direction change, turning times, segmented straight-ahead time and the like are abstracted, a threshold value is not directly set, a large amount of single-towing track characteristics are learned, at least 3 hours of historical tracks are extracted, the tracks are classified, the false alarm rate is greatly reduced, missing report is hardly generated, and the information acquisition capability of illegal behaviors caused by sea-fishing is improved.
Drawings
FIG. 1 is a flow chart diagram of a method of identifying a single tow vessel target in accordance with an embodiment of the present invention;
FIG. 2 is a schematic algorithmic flow diagram of a method of identifying a single tow vessel target of an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a single-towed-vessel target recognition apparatus according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may 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 disclosure to those skilled in the art.
As shown in fig. 1, an embodiment of the present invention provides a method for identifying a target of a single-towed ship, including:
step S1: acquiring real-time running track data of a ship;
step S2: preprocessing the real-time driving track data to obtain a preprocessing result;
step S3: carrying out track clustering on the preprocessing result to obtain a clustering result;
step S4: extracting the characteristics of the clustering result to obtain characteristic data;
step S5: and inputting the characteristic data into a preset model for identifying the single-towing ship target for processing to obtain an identification result of the single-towing ship target.
Specifically, as shown in fig. 2, acquiring real-time travel track data of a ship includes:
1) creating a single-drag template at the front end, drawing a single-drag identification area, associating the single-drag identification area with the template to generate an early warning condition, and storing the early warning condition to a Remote Dictionary service (Redis); the back end stores the early warning condition; and stores the data into Redis and Mysql (relational database management system).
2) The target tracks for the last 3 hours are read using Spark (fast general purpose computational engine for large-scale data processing), filtered with the region conditions in Redis, and saved to a local file.
According to the embodiment of the invention, based on the movement characteristics of the single-towing ship, the characteristics of sailing direction change, turning times, segmental straight-ahead time and the like are abstracted, a threshold value is not directly set, a large number of single-towing track characteristics are learned, at least 3 hours of historical tracks are extracted, and the tracks are classified, so that the false alarm rate is greatly reduced, the missing report is hardly caused, and the illegal behavior information acquisition capability of sea-fishing-related violation is improved.
In an alternative embodiment of the present invention, as shown in fig. 2, in step S2, the preprocessing the real-time driving trajectory data includes:
deleting the real-time running track data of which the number of points of the track is less than a first value in the real-time running track data;
deleting the real-time running track data of which the average speed is less than a second value in the real-time running track data; and/or
And deleting the real-time running track data of which the average speed is greater than a third value in the real-time running track data.
Wherein the first value may be 10; the second value may be 2 sections; the third value may be 6 sections;
in an optional embodiment of the present invention, in step S3, performing track clustering on the preprocessing result to obtain a clustering result, including:
performing track clustering on the preprocessing result through a clustering algorithm to obtain an intermediate clustering result;
calculating the moving range of the ship to be identified according to the intermediate clustering result, and acquiring a moving range result;
deleting the active range results that are less than or equal to the first threshold; otherwise, the activity range is reserved to obtain the final clustering result.
Clustering is carried out on the preprocessing result by using a KMeans clustering algorithm, and the first threshold can be 900 meters in a general activity range within 20 minutes;
in an optional embodiment of the present invention, the method for identifying a target of a single-towed vessel further includes:
deleting ships with the ship length not within the range of 0-60 meters;
the characteristics of the track are as follows: moving back and forth in the fixed area.
In an optional embodiment of the present invention, in step S4, performing feature extraction on the clustering result to obtain feature data, where the feature data includes:
performing feature extraction on the clustering result to obtain at least one of the following feature data: course change result, turning frequency result, straight-going time result, circle rate result, navigation speed and course.
In an optional embodiment of the present invention, in step S4, the heading change result is a sum of heading changes of the ship to be identified; the course change is the course difference of the ship to be identified at two moments;
the turning times result is the turning times of the ship to be identified;
the straight-going time result is the sectional statistical straight-going time of the ship to be identified;
the circle rate result is the circle rate of the ship to be identified; the circle rate is the total course of the vessel to be identified divided by the displacement.
In an alternative embodiment of the present invention, as shown in FIG. 2, in step S5, the predictive single tow vessel target model is trained by:
acquiring historical track sample data of a ship;
carrying out data cleaning processing on the historical track sample data of the ship to obtain a cleaning result;
performing data visualization processing on the cleaning result to obtain a visualization result;
performing feature extraction on the visualization result to obtain a feature extraction result;
and training the feature extraction result through a random forest model to obtain the preset model.
Specifically, the python offline portion includes:
1) acquiring historical track sample data of a ship, comprising the following steps: and manually identifying the same number of targets of the single-towed ship and normal targets (which are not targets of the single-towed ship), and inquiring the historical track from the historical track library as sample data.
2) Carrying out data cleaning processing on the historical track sample data of the ship to obtain a cleaning result, wherein the data cleaning processing comprises the following steps: and deleting tracks with too few points, wherein track data are too few and have no reference value.
3) Performing data visualization processing on the cleaning result to obtain a visualization result, including: the method comprises the steps of visualizing the track of each ship to be identified, deleting obvious outliers, wherein the distances between some outliers and a normal track are far, and if the outliers are not deleted, the accuracy of the model is greatly influenced.
4) And performing feature extraction on the visualization result to obtain a feature extraction result, wherein the feature extraction content specifically comprises the following features:
41. and (5) counting the sum of course changes of the ship to be identified (for the same target ID, calculating course differences at two moments before and after, and then accumulating).
42. And (5) counting the turning times of the ship to be identified (if the course difference between the front course and the back course is more than 180 degrees, the ship is turned).
43. And (4) counting the straight-going time of the ship to be identified in a segmented mode (the total track time length of the same target ID is divided by the turning times to be the straight-going time length of each segment).
44. The circle rate (total movement distance divided by displacement) of the ship to be identified is calculated.
45. And counting the maximum value, the minimum value, the average value, the median, the de-duplication statistic value, the standard deviation and the quantile of the navigational speed of the ship to be identified.
46. Dividing the course change of more than 180 and less than 180 into two groups, and combining the two groups with the navigational speed respectively:
specifically, calculating the course difference of the same target ID, wherein the course difference is divided into a group with the number more than or equal to 180 and a group with the number less than 180, and forming the characteristics: the course difference is more than or equal to 180_ navigational speed 1, the navigational phase difference is less than 180_ navigational speed 1, the course difference is more than or equal to 180_ navigational speed 2, and the navigational phase difference is less than 180_ navigational speed 2.
The feature extraction result obtained by the feature extraction specifically includes: course change result, turning frequency result, straight-going time result, circle rate result, navigation speed and course.
5) Training the feature extraction result through a random forest model to obtain the target model of the predicted single-towed ship, wherein the training comprises the following steps:
51. loading an offline model (a preset model) from an HDFS (Hadoop Distributed File System), and randomly sampling n samples from the sample data in a returning way;
52. randomly selecting k features from all feature extraction results obtained in the step 4), and establishing a decision tree (generally, a classification tree CART, or other or mixed) by using the k features for the selected n samples;
53. repeating the steps 51 and 52 for m times to generate m decision trees and form a random forest model;
54. for new data, through each tree decision, the decision is finally voted to confirm which category is assigned.
In an optional embodiment of the present invention, in step S5, the parameters of the random forest model include:
n_estimators＝60；
max_depth＝10；
min_samples_split＝10；
wherein n _ estimators is the number of the decision trees, max _ depth is the depth of the decision trees, and min _ samples _ split is the minimum sample number of the node.
Specifically, the larger the number of the decision trees is, the more the generated decision trees are, the more accurate the result is, but the side length of the training time is also increased; the minimum sample number of the nodes is smaller, so that overfitting is less likely to occur, the depth of the decision tree is larger, the overfitting is more likely to occur when the tree is deeper, and the overfitting is less likely to occur when the value is smaller;
as shown in fig. 2, the best parameters are selected using grid search and cross validation, and the pre-set model is saved to the HDFS.
In an optional embodiment of the present invention, in step S5, inputting the feature data into a preset model for identifying a single-towed-vessel target, and processing the feature data to obtain an identification result of the single-towed-vessel target, where the identification result includes:
inputting the characteristic data into a preset model, and outputting a prediction result;
judging the category of the ship to be identified according to the prediction result to obtain a judgment result;
and acquiring a single-towing ship target according to the judgment result.
Specifically, a feature engineering (feature extraction) method in the python offline process is called, the same features are generated for the real-time data, an offline model (preset model) is loaded, and the prediction result is obtained through real-time prediction.
The trained preset model is a plurality of trees formed by a pile of ifelse sentences, the trees are judged layer by layer from the root node to the leaf nodes, the final category to which the target belongs is obtained, voting is carried out by integrating the results of the plurality of trees, the final category is obtained by few trees obeying majority, prediction is carried out, and the probability that the target is a single-dragged target is output; the output result may be a category to which the target belongs, or may be a probability.
The object categories include: AISB (automatic identification system for ships), RADAR _ AIS _ B, RADAR (RADAR).
In an alternative embodiment of the present invention, as shown in fig. 2, in step S5, the acquiring a single-towed-vessel target according to the determination result includes:
and when the judgment result shows that the probability that the ship to be identified is the single-towing ship target is greater than 0.7, acquiring the single-towing ship target.
Specifically, as shown in fig. 2, the method for identifying a single-towed-vessel target further includes:
1) the ES (file browser) inquires the latest longitude and latitude and other information according to the prediction result so as to obtain the position information of the single-towing ship target and sends the position information to the distributed publishing and subscribing message system Kafka;
2) according to the early warning data of the distributed publish-subscribe message system Kafka, storing results to Redis, Mysql and Hive (data warehouse tool), regularly requesting the early warning results by the front end (predicting whether a single-towed-vessel target model identifies the target as a single-towed-vessel target, if the target is the single-towed-vessel target, storing the target to Kafka, and if not, not storing the target to Kafka), circling out the single-towed-vessel target by using a mark, and obtaining the early warning results and storing the early warning results to Mysql;
3) and displaying the early warning result.
According to the technical scheme of the embodiment of the invention, the accurate identification of the single-towing fishing vessel is realized, the information acquisition capability of illegal behaviors involving sea and fishing is improved, and the accurate striking of illegal behaviors involving fishing and the fine management of fishing in a quarter are further enhanced.
As shown in fig. 3, an embodiment of the present invention further provides a single tow vessel target identification apparatus 30, comprising:
the acquisition module 31 is used for acquiring real-time running track data of a ship;
the preprocessing module 32 is configured to preprocess the real-time travel track data to obtain a preprocessing result;
the clustering module 33 is configured to perform track clustering on the preprocessing result to obtain a clustering result;
a feature extraction module 34, configured to perform feature extraction on the clustering result to obtain feature data;
and the identification module 35 is configured to input the feature data into a preset model for identifying the single-towed-vessel target, and process the feature data to obtain an identification result of the single-towed-vessel target.
Optionally, the preprocessing the real-time driving trajectory data includes:
deleting the real-time running track data of which the number of points of the track is less than a first value in the real-time running track data;
deleting the real-time running track data of which the average speed is less than a second value in the real-time running track data; and/or
And deleting the real-time running track data of which the average speed is greater than a third value in the real-time running track data.
Optionally, performing track clustering on the preprocessing result to obtain a clustering result, including:
performing track clustering on the preprocessing result through a clustering algorithm to obtain an intermediate clustering result;
calculating the moving range of the ship to be identified according to the intermediate clustering result, and acquiring a moving range result;
deleting the active range results that are less than or equal to the first threshold; otherwise, the activity range is reserved to obtain the final clustering result.
Optionally, performing feature extraction on the clustering result to obtain feature data, including:
performing feature extraction on the clustering result to obtain at least one of the following feature data: course change result, turning frequency result, straight-going time result, circle rate result, navigation speed and course.
Optionally, the course change result is the sum of course changes of the ship to be identified, and the course change is the course difference of the ship to be identified at two moments;
the turning times result is the turning times of the ship to be identified;
the straight-going time result is the sectional statistical straight-going time of the ship to be identified;
the circle rate result is the circle rate of the ship to be identified; the circle rate is the total course of the vessel to be identified divided by the displacement.
Optionally, the predictive single tow vessel target model is trained by:
acquiring historical track sample data of a ship;
carrying out data cleaning processing on the historical track sample data of the ship to obtain a cleaning result;
performing data visualization processing on the cleaning result to obtain a visualization result;
performing feature extraction on the visualization result to obtain a feature extraction result;
and training the feature extraction result through a random forest model to obtain the preset model.
Optionally, the parameters of the random forest model include:
n_estimators＝60；
max_depth＝10；
min_samples_split＝10；
wherein n _ estimators is the number of the decision trees, max _ depth is the depth of the decision trees, and min _ samples _ split is the minimum sample number of the node.
Optionally, inputting the feature data into a preset model for identifying the single-towed-vessel target, and processing the feature data to obtain an identification result of the single-towed-vessel target, where the identification result includes:
inputting the characteristic data into a preset model, and outputting a prediction result;
judging the category of the ship to be identified according to the prediction result to obtain a judgment result;
and acquiring a single-towing ship target according to the judgment result.
Optionally, the obtaining a single-towing ship target according to the determination result includes:
and when the judgment result shows that the probability that the ship to be identified is the single-towing ship target is greater than 0.7, acquiring the single-towing ship target.
It should be noted that the apparatus is an apparatus corresponding to the above method, and all the implementations in the above method embodiment are applicable to the embodiment of the apparatus, and the same technical effects can be achieved.
Embodiments of the present invention also provide a computer-readable storage medium comprising instructions which, when executed on a computer, cause the computer to perform the method as described above. All the implementation manners in the above method embodiment are applicable to this embodiment, and the same technical effect can be achieved.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.
Furthermore, it is to be noted that in the device and method of the invention, it is obvious that the individual components or steps can be decomposed and/or recombined. These decompositions and/or recombinations are to be regarded as equivalents of the present invention. Also, the steps of performing the series of processes described above may naturally be performed chronologically in the order described, but need not necessarily be performed chronologically, and some steps may be performed in parallel or independently of each other. It will be understood by those skilled in the art that all or any of the steps or elements of the method and apparatus of the present invention may be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or any combination thereof, which can be implemented by those skilled in the art using their basic programming skills after reading the description of the present invention.
Thus, the objects of the invention may also be achieved by running a program or a set of programs on any computing device. The computing device may be a general purpose device as is well known. The object of the invention is thus also achieved solely by providing a program product comprising program code for implementing the method or the apparatus. That is, such a program product also constitutes the present invention, and a storage medium storing such a program product also constitutes the present invention. It is to be understood that the storage medium may be any known storage medium or any storage medium developed in the future. It is further noted that in the apparatus and method of the present invention, it is apparent that each component or step can be decomposed and/or recombined. These decompositions and/or recombinations are to be regarded as equivalents of the present invention. Also, the steps of executing the series of processes described above may naturally be executed chronologically in the order described, but need not necessarily be executed chronologically. Some steps may be performed in parallel or independently of each other.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (10)
1. A method of identifying a single tow vessel target, comprising:
acquiring real-time running track data of a ship;
preprocessing the real-time driving track data to obtain a preprocessing result;
carrying out track clustering on the preprocessing result to obtain a clustering result;
extracting the characteristics of the clustering result to obtain characteristic data;
and inputting the characteristic data into a preset model for identifying the single-towing ship target for processing to obtain an identification result of the single-towing ship target.
2. The method of claim 1, wherein preprocessing the real-time travel trajectory data comprises:
deleting the real-time running track data of which the number of points of the track is less than a first value in the real-time running track data;
deleting the real-time running track data of which the average speed is less than a second value in the real-time running track data; and/or
And deleting the real-time running track data of which the average speed is greater than a third value in the real-time running track data.
3. The method for identifying the target of the single-towed ship according to claim 1, wherein the track clustering is performed on the preprocessing result to obtain a clustering result, and the method comprises the following steps:
performing track clustering on the preprocessing result through a clustering algorithm to obtain an intermediate clustering result;
calculating the moving range of the ship to be identified according to the intermediate clustering result, and acquiring a moving range result;
deleting the active range results that are less than or equal to the first threshold; otherwise, the activity range is reserved to obtain the final clustering result.
4. The method for identifying the single-towed-vessel target of claim 3, wherein the step of performing feature extraction on the clustering result to obtain feature data comprises the steps of:
performing feature extraction on the clustering result to obtain at least one of the following feature data: course change result, turning frequency result, straight-going time result, circle rate result, navigation speed and course.
5. The method of identifying a single tow vessel target of claim 4,
the course change result is the sum of course changes of the ship to be identified, and the course change is the course difference of the ship to be identified at two moments;
the turning times result is the turning times of the ship to be identified;
the straight-going time result is the sectional statistical straight-going time of the ship to be identified;
the circle rate result is the circle rate of the ship to be identified; the circle rate is the total course of the vessel to be identified divided by the displacement.
6. The method of identifying a single tow vessel target of claim 1, wherein the predictive single tow vessel target model is trained by:
acquiring historical track sample data of a ship;
carrying out data cleaning processing on the historical track sample data of the ship to obtain a cleaning result;
performing data visualization processing on the cleaning result to obtain a visualization result;
performing feature extraction on the visualization result to obtain a feature extraction result;
and training the feature extraction result through a random forest model to obtain the preset model.
7. The method of identifying a single tow vessel target of claim 6, wherein the parameters of the random forest model include:
n_estimators＝60；
max_depth＝10；
min_samples_split＝10；
wherein n _ estimators is the number of the decision trees, max _ depth is the depth of the decision trees, and min _ samples _ split is the minimum sample number of the node.
8. The method for identifying a single-towed-vessel target of claim 7, wherein the step of inputting the characteristic data into a preset model for identifying the single-towed-vessel target and processing the characteristic data to obtain the identification result of the single-towed-vessel target comprises the steps of:
inputting the characteristic data into a preset model, and outputting a prediction result;
judging the category of the ship to be identified according to the prediction result to obtain a judgment result;
and acquiring a single-towing ship target according to the judgment result.
9. The method for identifying a single-towed-vessel target according to claim 8, wherein said obtaining a single-towed-vessel target according to the determination comprises:
and when the judgment result shows that the probability that the ship to be identified is the single-towing ship target is greater than 0.7, acquiring the single-towing ship target.
10. An apparatus for identifying a target of a single tow vessel, comprising:
the acquisition module is used for acquiring real-time running track data of the ship;
the preprocessing module is used for preprocessing the real-time driving track data to obtain a preprocessing result;
the clustering module is used for carrying out track clustering on the preprocessing result to obtain a clustering result;
the characteristic extraction module is used for extracting the characteristics of the clustering result to obtain characteristic data;
and the identification module is used for inputting the characteristic data into a preset model for identifying the single-towing ship target to be processed, and obtaining the identification result of the single-towing ship target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110265932.9A CN112907632A (en) | 2021-03-11 | 2021-03-11 | Single-towing ship target identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110265932.9A CN112907632A (en) | 2021-03-11 | 2021-03-11 | Single-towing ship target identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112907632A true CN112907632A (en) | 2021-06-04 |
Family
ID=76105626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110265932.9A Pending CN112907632A (en) | 2021-03-11 | 2021-03-11 | Single-towing ship target identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112907632A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113849902A (en) * | 2021-08-19 | 2021-12-28 | 中国人民解放军91977部队 | Ship type inference system and method |
| CN114898222A (en) * | 2022-04-21 | 2022-08-12 | 中国人民解放军91977部队 | Ship target track identification method and device |
| CN115563889A (en) * | 2022-12-06 | 2023-01-03 | 三亚海兰寰宇海洋信息科技有限公司 | Target object anchoring prediction method, device and equipment |
-
2021
- 2021-03-11 CN CN202110265932.9A patent/CN112907632A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113849902A (en) * | 2021-08-19 | 2021-12-28 | 中国人民解放军91977部队 | Ship type inference system and method |
| CN114898222A (en) * | 2022-04-21 | 2022-08-12 | 中国人民解放军91977部队 | Ship target track identification method and device |
| CN114898222B (en) * | 2022-04-21 | 2024-01-02 | 中国人民解放军91977部队 | Ship target track identification method and device |
| CN115563889A (en) * | 2022-12-06 | 2023-01-03 | 三亚海兰寰宇海洋信息科技有限公司 | Target object anchoring prediction method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112907632A (en) | Single-towing ship target identification method and device | |
| CN109948669B (en) | Abnormal data detection method and device | |
| Dinkel et al. | Towards duration robust weakly supervised sound event detection | |
| CN112800116A (en) | Method and device for detecting abnormity of service data | |
| CN110018453A (en) | Intelligent type recognition methods based on aircraft track feature | |
| CN113157800B (en) | Identification method for discovering dynamic target in air in real time | |
| CN114254716B (en) | High-risk operation identification method and system based on user behavior analysis | |
| CN112464951B (en) | Fishery safety emergency disposal system and method based on Internet of things | |
| CN113341919B (en) | Computing system fault prediction method based on time sequence data length optimization | |
| CN112116168B (en) | User behavior prediction method and device and electronic equipment | |
| CN115759033A (en) | Method, device and equipment for processing track data | |
| Toloue et al. | Anomalous behavior detection of marine vessels based on Hidden Markov Model | |
| CN112532652A (en) | Attack behavior portrait device and method based on multi-source data | |
| CN115438102A (en) | Space-time data anomaly identification method and device and electronic equipment | |
| CN114863170A (en) | Deep learning-based new energy vehicle battery spontaneous combustion early warning method and device | |
| KR102005952B1 (en) | Apparatus and Method for refining data of removing noise data in Machine learning modeling | |
| CN109740750B (en) | Data collection method and device | |
| Tritsarolis et al. | Online Co-movement Pattern Prediction in Mobility Data | |
| Asaddulloh et al. | Machine Learning Techniques to Predict Rain Tomorrow for Automated Plant Watering System | |
| Chao et al. | Marshalling model inference in video streams | |
| CN114518988A (en) | Resource capacity system, method of controlling the same, and computer-readable storage medium | |
| CN115659282B (en) | GA-KNN-based intelligent extraction method and system for effective motion characteristics of early warning target | |
| CN113743184A (en) | Abnormal behavior crowd detection method and device based on element mining and video analysis | |
| Stavropoulos et al. | Optimizing complex event forecasting | |
| Mittal | Employee Attrition Prediction Using Machine Learning Algorithms |
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 |