CN114510961A - Ship behavior intelligent monitoring algorithm based on recurrent neural network and Beidou positioning - Google Patents
Ship behavior intelligent monitoring algorithm based on recurrent neural network and Beidou positioning Download PDFInfo
- Publication number
- CN114510961A CN114510961A CN202210000265.6A CN202210000265A CN114510961A CN 114510961 A CN114510961 A CN 114510961A CN 202210000265 A CN202210000265 A CN 202210000265A CN 114510961 A CN114510961 A CN 114510961A
- Authority
- CN
- China
- Prior art keywords
- neural network
- recurrent neural
- ship
- data
- beidou
- 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
- 230000000306 recurrent effect Effects 0.000 title claims abstract description 35
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 25
- 238000012544 monitoring process Methods 0.000 title claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000003062 neural network model Methods 0.000 claims abstract description 13
- 238000004140 cleaning Methods 0.000 claims abstract description 7
- 238000010606 normalization Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000000605 extraction Methods 0.000 claims description 6
- DMBHHRLKUKUOEG-UHFFFAOYSA-N diphenylamine Chemical compound C=1C=CC=CC=1NC1=CC=CC=C1 DMBHHRLKUKUOEG-UHFFFAOYSA-N 0.000 claims 1
- 230000019771 cognition Effects 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 21
- 238000000034 method Methods 0.000 description 9
- 238000010276 construction Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 238000004088 simulation Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/42—Determining position
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/52—Determining velocity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Signal Processing (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an intelligent ship behavior monitoring algorithm based on a recurrent neural network and Beidou positioning, which comprises the steps of firstly carrying out normalization processing on returned Beidou data, cleaning the data and constructing a sample with a ship operation mode label; then training a recurrent neural network by using the sample to obtain a recurrent neural network model for predicting the ship operation intention; and finally, normalizing and cleaning the received Beidou data, inputting the data into a recurrent neural network model for prediction, and obtaining a prediction result of the operation behavior of the ship. The attribute characteristics of the ship such as the navigational speed, the course, the navigation way shortcut and the like and the corresponding operation behavior labels are used as samples to be input into the recurrent neural network model, and the network node weight training model is adjusted, so that the model can simulate the dynamic cognitive process with sequence characteristics, and the real-time monitoring of the ship behavior is realized.
Description
Technical Field
The invention belongs to the fields of computer technology and artificial intelligence, and judges the ship track through real-time Beidou positioning returned by a ship so as to judge the current behavior intention of the ship.
Background
For marine monitoring, the position of the ship can be determined according to real-time positioning information returned by the Beidou after the ship is offshore, but the method can only monitor the position, the place and the state of the ship and cannot effectively restore the operation intention of the ship at the moment. For specific ships, such as fishing vessels, small ships and other objects, if the behavioral intention of the specific ships can be analyzed through the ship-parking track, the maritime affairs can be effectively monitored according to the presumed result of the intention, and the corresponding maritime affairs activities in the regional sea area can be reasonably planned.
Taking an application scene of fishery fishing and mining as an example, most fishing boats in China are equipped with Beidou navigation equipment at present, a maritime department can realize real-time butt joint of the data and obtain corresponding boat positions, and meanwhile, the running track of the fishing boat can be restored through space-time link of the positions, and the prior art can only restore the operation track and the action of the fishing boat on the sea by using navigation data. However, in the prior art, the operation track of the fishing boat on the sea can be restored only by using the Beidou navigation data, the restoration of the time-space position data and the track reproduction are only realized, further data mining analysis is lacked, the ship navigation data and the navigation intention are not effectively associated, and the utilization degree of the maritime data is low.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides an intelligent ship behavior monitoring algorithm based on a recurrent neural network and Beidou positioning, which is characterized in that attribute characteristics of ship speed, course, navigation path shortcut, space-time relative position coordinates and the like and corresponding operation behavior labels are used as samples and input into a recurrent neural network model, network node weights are adjusted, and the trained model is handed to a maritime department, so that monitoring on one class of ship operation behaviors in a unified sea area is realized, a dynamic cognitive process with sequence characteristics can be simulated by the model, and real-time monitoring on the ship behaviors in maritime operation is realized.
The technical scheme adopted by the invention for solving the technical problem comprises the following steps:
and 4, normalizing and cleaning the received Beidou data according to the steps 1 and 2, inputting the data into the recurrent neural network model generated in the step 3 for prediction, and obtaining the prediction result of the operation behavior of the ship.
And 2, when the sliding regression is carried out in the step 2, if the number of the residual samples is less than one window length, discarding the residual sample data.
And 3, calculating a loss value by using the cross entropy as a cost function, and stopping training when the value of the cross entropy is less than 0.2.
The invention has the beneficial effects that:
according to the invention, the operation behavior of the ship is effectively mapped and associated with the navigation state of the ship by learning the big data of the marine ship, so that the learning of nonlinear mapping between the ship behavior and the ship state attribute and the integration of prior information are realized, the recursive neural network algorithm can be converged more quickly, the training result can be multiplexed, and the method has good adaptability to different scenes.
The method is different from the traditional method in that only the attribute characteristic of a single moment is considered, the time sequence change characteristic is added, and the characteristic change before and after the time sequence change is related.
The invention can effectively realize the correlation of the current time-space behavior with the operation intention, judge the state of the fishing vessel, analyze the position relation between the position of the fishing vessel and the exclusive economic area, the fishery agreement area and the fishing permitted area after the analysis result is returned to the shore base, can identify the fishing in the sea area beyond the range specified by the fishing license of the fishing vessel, and can identify the illegal fishing in the fishing area through the time position.
The invention can predict the operation behavior of the ship through Beidou positioning, and can effectively monitor and plan fishery exploitation in the intelligent monitoring application of the fishery.
Drawings
FIG. 1 is a basic flow chart of the algorithm of the present invention;
FIG. 2 is a schematic diagram of a neural network framework.
Detailed Description
The present invention will be further described with reference to the following drawings and examples, which include, but are not limited to, the following examples.
The idea of the invention is to train a marine department to realize associated fishing boat behavior samples in a certain range of sea area by using a recurrent neural network, and predict the operation behavior of the position boat by using a training model after obtaining the training model.
The specific implementation steps of the invention comprise:
Normalizing the selected attribute data to obtain
Wherein m is the monitored characteristic attribute, including speed, course angle, relative position, time, and travel area relative to the shore-based observation point; m isminIs the minimum value of the feature, mmaxAnd finally, normalizing the range of each feature to be 0-1 for the maximum value of the feature.
And setting L as the time window length of the sample construction, wherein the L is selected according to the actual use condition.
The input structure of the recurrent neural network model is generally [ sample number, time window length, characteristic number ]]A three-dimensional array of shapes. The invention introduces a sliding regression method. I.e. in the original time series S ═ { x ═ x1,x2,…xi…xTWhere T is the length of the total time series, xiThe data representing the ith time point is a d-dimensional vector, i.e. xi＝{n1,n2,n3…ndAnd d represents the number of features. Given a sliding window length L, i.e. a time window L, a sliding regression is performed on the original time series with t as a step length to construct a sliding regression model consisting of { x }1,x2,…xL,y1},{xt+1,xt+2,…xt+L,yt… …, when the number of samples is less than one window length,discarding superfluous data, wherein ytAnd (4) a label representing the t sample, namely the ship operation mode.
And (3) inputting the ship operation mode labels corresponding to the data samples processed in the step (2) by adopting One-hot codes and corresponding multi-dimensional characteristics as input into a recurrent neural network, extracting spatial characteristics and temporal characteristics of the data by utilizing the structural characteristics of the recurrent neural network, connecting a Softmax classifier behind a recurrent neural network layer, converting the ship operation behavior prediction problem into a classification problem, calculating a loss value by using cross entropy as a cost function in network training, and stopping the training when the value of the cross entropy is less than 0.2 to obtain a recurrent neural network model for predicting the ship operation intention.
And (3) carrying out normalization and cleaning on the subsequently received Beidou sequential data according to the steps 1 and 2, and then inputting the recursive neural network model generated in the step 3 for prediction to obtain the operation behavior prediction result of the unknown ship at the moment.
The specific steps of the present invention will be further described with reference to fig. 1.
And obtaining ship position points and corresponding operation type data through a ship Beidou terminal.
And 2, normalizing the sample characteristics.
Normalization processing is carried out on returned Beidou data
Wherein m is a monitored characteristic such as speed, heading, route shortcut, space-time relative position coordinate, etc., mminIs the minimum value of the feature, mmaxIs the maximum value of the feature. Finally, the range of each feature is normalized to 0-1.
The method comprises the following steps of finishing sample segmentation and data set construction by using a sliding method:
setting L as the time window length of the sample construction, where L is selected as appropriate according to the actual use situation, and the selection of L in this embodiment is 200.
The input structure of the recurrent neural network model is generally [ sample number, time window length, characteristic number ]]A three-dimensional array of shapes. This patent introduces a method of sliding regression. I.e. in the original time series S ═ { x ═ x1,x2,…xi…xTWhere T is the length of the total time series, xiThe data representing the ith time point is a d-dimensional vector, i.e. xi＝{n1,n2,n3…ndAnd d represents the number of features. Given a sliding window length L, i.e. a time window L, a sliding regression is performed on the original time series with t as a step length to construct a sliding regression model consisting of { x }1,x2,…xL,y1},{xt+1,xt+2,…xt+L,yt… …, etc. Wherein y istAnd (4) a label representing the t sample, namely the ship operation mode. Thus, the construction of the ship operation sample is completed.
And 3, performing one-hot coding on the corresponding intention of the processed data by data training, inputting the coded data and the corresponding multidimensional characteristics into a recurrent neural network, wherein the recurrent neural network designed in the practice has the input characteristic number of 5, the hidden layer number of 20 and the network layer number of 2, performing spatial characteristic extraction and time characteristic extraction on the data by utilizing the structural characteristics of the recurrent neural network, connecting a Softmax classifier behind the recurrent neural network layer, converting the ship operation behavior prediction problem into classification problem network training, calculating a loss value by using cross entropy as a cost function, converging the model when the cross entropy is less than 0.2, stopping training, and obtaining a model result of the recurrent neural network for prediction.
And 4, testing the simulation result.
The sequence inertial Beidou information of unknown ships is obtained through Beidou navigation, the operation behavior result of the unknown ships is predicted by using the model, and the real-time monitoring of the operation behavior of the ships in the designated sea area is realized.
The fishing boat fishing behavior and the Beidou data samples which are externally published in 2019 by Fujian maritime are selected as research objects to be practiced, and the Fujian maritime fishery management department publishes a group of samples which are correspondingly related between fishing boat time sequence Beidou positioning information and corresponding net-pulling fishing operation modes in 2019. In the simulation process, five characteristics of the navigational speed, the course angle, the relative position, the time and the traveling area relative to a shore-based observation point of a fishing boat are selected as sample attributes extracted from Beidou return data, corresponding Beidou data of 3w fishing boats are cleaned and structurally processed to form a sample set, a recurrent neural network is used for training, a training model is used for predicting the residual 2000 samples, the training and predicting accuracy is 0.93, and the testing accuracy reaches 0.91.
Claims (4)
1. The utility model provides a boats and ships action intelligent monitoring algorithm based on recurrent neural network and big dipper location which characterized in that includes following step:
step 1, normalization processing is carried out on the returned Beidou data to obtain an original time sequence S ═ x1,x2,…xi…xTWhere T is the length of the time series, xiData representing the ith time point, xi＝{n1,n2,n3…ndD represents the number of features including speed, heading angle, relative position, time, and area of travel relative to a shore-based viewpoint;
step 2, a sliding window length L is given, sliding regression is carried out on the original time sequence by taking t as a step length, and a sequence formed by { x }1,x2,…xL,y1},{xt+1,xt+2,…xt+L,yt… …, where ytA label representing the t-th sample, namely a ship operation mode;
step 3, after the ship operation mode corresponding to the sample is coded by One-hot, the ship operation mode and corresponding multidimensional characteristics are jointly used as the input of a recurrent neural network, the data are subjected to spatial characteristic extraction and time characteristic extraction, a Softmax classifier is connected behind a recurrent neural network layer, the ship operation behavior prediction problem is converted into a classification problem, cross entropy is used as a cost function for calculating a loss value in network training, and a recurrent neural network model for predicting the ship operation intention is obtained;
and 4, normalizing and cleaning the received Beidou data according to the steps 1 and 2, inputting the data into the recurrent neural network model generated in the step 3 for prediction, and obtaining the prediction result of the operation behavior of the ship.
2. The vessel behavior intelligent monitoring algorithm based on the recurrent neural network and Beidou positioning as claimed in claim 1, wherein the step 1 normalizes the returned Beidou data to obtain
Where m is the monitored characteristic, mminIs the minimum value of the feature, mmaxIs the maximum value of the feature.
3. The vessel behavior intelligent monitoring algorithm based on the recurrent neural network and the Beidou positioning as claimed in claim 1, wherein when the step 2 performs the sliding regression, if the number of the remaining samples is less than a window length, the remaining sample data is discarded.
4. The vessel behavior intelligent monitoring algorithm based on the recurrent neural network and the Beidou positioning as claimed in claim 1, wherein the step 3 uses cross entropy as a cost function to calculate a loss value, and when the value of the cross entropy is less than 0.2, the training is stopped.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210000265.6A CN114510961A (en) | 2022-01-03 | 2022-01-03 | Ship behavior intelligent monitoring algorithm based on recurrent neural network and Beidou positioning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210000265.6A CN114510961A (en) | 2022-01-03 | 2022-01-03 | Ship behavior intelligent monitoring algorithm based on recurrent neural network and Beidou positioning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114510961A true CN114510961A (en) | 2022-05-17 |
Family
ID=81549245
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210000265.6A Pending CN114510961A (en) | 2022-01-03 | 2022-01-03 | Ship behavior intelligent monitoring algorithm based on recurrent neural network and Beidou positioning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114510961A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116310601A (en) * | 2023-05-19 | 2023-06-23 | 广州健新科技有限责任公司 | Ship behavior classification method based on AIS track diagram and camera diagram group |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018000748A1 (en) * | 2016-06-28 | 2018-01-04 | 深圳大学 | Maritime navigation rescue system based on beidou satellite |
| WO2018026310A1 (en) * | 2016-08-03 | 2018-02-08 | Акционерное общество "Российская корпорация ракетно-космического приборостроения и информационных систем" (АО "Российские космические системы") | Method of monitoring the fishing of aquatic biological resources |
| 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 |
| RU2735357C1 (en) * | 2020-01-28 | 2020-10-30 | Акционерное общество «Российская корпорация ракетно-космического приборостроения и информационных систем» (АО «Российские космические системы») | Information-analytical system for monitoring of aquatic biological resources and method of use thereof |
| CN111899450A (en) * | 2020-07-24 | 2020-11-06 | 宁波盛洋电子科技有限公司 | Method and system for monitoring ships entering and exiting port and finding dangerous ships |
| WO2021017577A1 (en) * | 2019-07-29 | 2021-02-04 | 南京莱斯网信技术研究院有限公司 | Ship-type-spoofing detection method employing ensemble learning |
| CN112561148A (en) * | 2020-12-09 | 2021-03-26 | 北京京航计算通讯研究所 | Ship track prediction method and system based on one-dimensional convolutional neural network and LSTM |
| CN113467250A (en) * | 2021-07-27 | 2021-10-01 | 浙江中裕通信技术有限公司 | Ship course and region estimation method based on Beidou navigation and data driving |
| CN113627359A (en) * | 2021-08-13 | 2021-11-09 | 上海磐启微电子有限公司 | Fishing boat behavior intelligent monitoring algorithm based on bidirectional long-time and short-time neural network |
-
2022
- 2022-01-03 CN CN202210000265.6A patent/CN114510961A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018000748A1 (en) * | 2016-06-28 | 2018-01-04 | 深圳大学 | Maritime navigation rescue system based on beidou satellite |
| WO2018026310A1 (en) * | 2016-08-03 | 2018-02-08 | Акционерное общество "Российская корпорация ракетно-космического приборостроения и информационных систем" (АО "Российские космические системы") | Method of monitoring the fishing of aquatic biological resources |
| 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 |
| WO2021017577A1 (en) * | 2019-07-29 | 2021-02-04 | 南京莱斯网信技术研究院有限公司 | Ship-type-spoofing detection method employing ensemble learning |
| RU2735357C1 (en) * | 2020-01-28 | 2020-10-30 | Акционерное общество «Российская корпорация ракетно-космического приборостроения и информационных систем» (АО «Российские космические системы») | Information-analytical system for monitoring of aquatic biological resources and method of use thereof |
| CN111899450A (en) * | 2020-07-24 | 2020-11-06 | 宁波盛洋电子科技有限公司 | Method and system for monitoring ships entering and exiting port and finding dangerous ships |
| CN112561148A (en) * | 2020-12-09 | 2021-03-26 | 北京京航计算通讯研究所 | Ship track prediction method and system based on one-dimensional convolutional neural network and LSTM |
| CN113467250A (en) * | 2021-07-27 | 2021-10-01 | 浙江中裕通信技术有限公司 | Ship course and region estimation method based on Beidou navigation and data driving |
| CN113627359A (en) * | 2021-08-13 | 2021-11-09 | 上海磐启微电子有限公司 | Fishing boat behavior intelligent monitoring algorithm based on bidirectional long-time and short-time neural network |
Non-Patent Citations (2)
| Title |
|---|
| 刘娇;史国友;杨学钱;朱凯歌;: "基于DE-SVM的船舶航迹预测模型", 上海海事大学学报, no. 01, 26 March 2020 (2020-03-26) * |
| 王胜正;申心泉;赵建森;冀宝仙;杨平安;: "基于ASAE深度学习预测海洋气象对船舶航速的影响", 交通运输工程学报, no. 02, 15 April 2018 (2018-04-15) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116310601A (en) * | 2023-05-19 | 2023-06-23 | 广州健新科技有限责任公司 | Ship behavior classification method based on AIS track diagram and camera diagram group |
| CN116310601B (en) * | 2023-05-19 | 2023-08-04 | 广州健新科技有限责任公司 | Ship behavior classification method based on AIS track diagram and camera diagram group |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108460382B (en) | Optical remote sensing image ship detection method based on deep learning single-step detector | |
| CN111428789A (en) | Network traffic anomaly detection method based on deep learning | |
| CN102799900B (en) | Target tracking method based on supporting online clustering in detection | |
| CN110232319A (en) | A kind of ship Activity recognition method based on deep learning | |
| CN112613478B (en) | Data active selection method for robot grabbing | |
| CN115730263A (en) | Ship behavior pattern detection method and device | |
| CN114510961A (en) | Ship behavior intelligent monitoring algorithm based on recurrent neural network and Beidou positioning | |
| CN115512152A (en) | Ship track classification method and system combining CNN (CNN) neural network and LSTM neural network | |
| CN115690568A (en) | Unmanned ship target detection method based on incremental learning | |
| Wang et al. | Object detection using transfer learning for underwater robot | |
| CN113516232B (en) | Self-attention mechanism-based wall-penetrating radar human body posture reconstruction method | |
| CN114155491A (en) | Ship behavior identification method and system based on AIS data | |
| CN117494057A (en) | Ship motion prediction method based on hierarchical multi-scale decomposition | |
| Xia et al. | Analysis of vessel anomalous behavior based on Bayesian recurrent neural network | |
| CN117349748A (en) | Active learning fault diagnosis method based on cloud edge cooperation | |
| CN108153519A (en) | A kind of Intelligent target tracking universal design frame | |
| CN117494871A (en) | Ship track prediction method considering ship interaction influence | |
| CN112418149A (en) | Abnormal behavior detection method based on deep convolutional neural network | |
| CN115931429A (en) | Intelligent robot soil sampling method based on image recognition | |
| CN111651930B (en) | Extreme learning machine-based flow field vortex region detection method | |
| Huang et al. | Ship trajectory anomaly detection based on multi-feature fusion | |
| Jia et al. | Lightweight CNN-Based Image Recognition with Ecological IoT Framework for Management of Marine Fishes | |
| CN117475387B (en) | Method for carrying out double-towing analysis on fishing boat by utilizing Hive off-line analysis technology and accompanying ideas | |
| CN115761537B (en) | Power transmission line foreign matter intrusion identification method oriented to dynamic feature supplementing mechanism | |
| CN115293297B (en) | Method for predicting track of ship driven by intention |
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 |