CN109389186B - VMS fixed gill net fishing boat net number extraction method based on DBSCAN - Google Patents
VMS fixed gill net fishing boat net number extraction method based on DBSCAN Download PDFInfo
- Publication number
- CN109389186B CN109389186B CN201811405919.3A CN201811405919A CN109389186B CN 109389186 B CN109389186 B CN 109389186B CN 201811405919 A CN201811405919 A CN 201811405919A CN 109389186 B CN109389186 B CN 109389186B
- Authority
- CN
- China
- Prior art keywords
- net
- fixed
- time interval
- time
- dbscan
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention relates to a VMS fixed gill net fishing boat net number extraction method based on DBSCAN, which comprises the following steps: counting the frequency of the instant navigational speed returned by each point, judging the net collecting and releasing states of the fixed barbed net, and extracting the ship position data in the net collecting state; and obtaining a time interval threshold value between the network levels by using the return time, calculating clustering parameters according to the navigational speed of each point and the time interval between the network levels, substituting a DBSCAN algorithm to judge the network level of the fixed barbed net, calculating the distance of each point in the network level in sequence, and extracting the length of each network level of the fixed barbed net. The method can calculate the number of the operation net times of the fixed gill net fishing boat and obtain the net length of each net time.
Description
Technical Field
The invention relates to the technical field of fishing vessel position data mining, in particular to a VMS fixed position gill net fishing vessel net number extraction method based on DBSCAN.
Background
Fishing boat Monitoring System (VMS, Vessel Monitoring System) can provide fishing boat time, position and dynamic information, and at present, the fishing boat of China installs automatic identification equipment (AIS) of boats and ships, big dipper satellite position Monitoring System, CDMA public mobile communication equipment and the like, and the real-time contact and tracking Monitoring to the marine fishing boat position are preliminarily realized. The time resolution of longitude and latitude ship position data transmitted by the Beidou satellite is recorded once in 3 minutes, the spatial resolution is about 10 meters, the space-time precision is high, the real-time performance is strong, and the method has a strong application value in the aspect of fishing behavior research of fishing boats. The net times are important statistical parameters in fishing effort force calculation, fishery resource investigation and fishery production management, most of the existing methods for extracting the net times of the fixed barbed net are statistical analysis based on course and speed characteristics, and a method for identifying the net times through clustering based on position information does not exist.
Disclosure of Invention
The invention aims to solve the technical problem of providing a VMS fixed gill net fishing boat net number extraction method based on DBSCAN, which can calculate the number of operation net numbers of the fixed gill net fishing boat and obtain the net length of each net.
The technical scheme adopted by the invention for solving the technical problems is as follows: the utility model provides a VMS fixed gill net fishing boat net number extraction method based on DBSCAN, which comprises the following steps:
(1) counting the frequency of the instant navigational speed returned by each point, judging the net collecting and releasing states of the fixed barbed net, and extracting the ship position data in the net collecting state;
(2) and obtaining a time interval threshold value between the network levels by using the return time, calculating clustering parameters according to the navigational speed of each point and the time interval between the network levels, substituting a DBSCAN algorithm to judge the network level of the fixed barbed net, calculating the distance of each point in the network level in sequence, and extracting the length of each network level of the fixed barbed net.
The step (1) specifically comprises: counting the frequency of ship points at each navigational speed according to the navigational speed of the fixed barbed wire fishing ship, finding out a frequency peak value, obtaining valley values at two sides of the peak value as the minimum value and the maximum value of a navigational speed threshold value of the fixed barbed wire net collection, and screening out the points which are possibly in the net collection state by utilizing the navigational speed threshold value.
The obtaining of the inter-network time interval threshold by using the reported time in the step (2) is specifically as follows: calculating the time interval of the adjacent ship position points in the points possibly in the network receiving state, counting the frequency of occurrence in each time interval, taking logarithm of the time interval frequency, and finding out the minimum value when the change in the distribution diagram tends to be stable as the threshold value of the time interval between networks.
The step (2) specifically comprises: calculating parameters of a DBSCAN algorithm according to the time resolution, the speed threshold value in the fixed net stabbing collection state and the time interval threshold value between the net steps, clustering VMS ship position data to extract each net step, merging different net steps belonging to the same net step by calculating the distance and time of starting and stopping points between adjacent net steps, sequentially calculating the distance of each point in the net steps, and extracting the length of the net steps of single net stabbing.
The radius range in the clustering parameters of the DBSCAN algorithm in the step (2) is the product of the time resolution and the speed threshold value in the network receiving state of the fixed position gill net, and the number of designated points in the radius is 3.
Advantageous effects
Due to the adoption of the technical scheme, compared with the prior art, the invention has the following advantages and positive effects: the method extracts the number of the fixed barbed net operation nets based on the position information, can count the net releasing length, has high accuracy and is suitable for all fixed barbed net operation fishing boats. The result can be used as the fishing effort force of the fixed barbed net for deeply analyzing the fishing economic benefit, and can also be used as important statistics to participate in production management such as limited fishing and fishery resource investigation and evaluation.
Drawings
FIG. 1 is a schematic diagram of track points of sea going operation of a fixed gill net fishing boat;
FIG. 2 is a statistical chart of the frequency of occurrence of fixed stinger ship sites at various speeds;
FIG. 3 is a schematic diagram of a possible fixed-position gill net collection point (thick black dot) for speed threshold screening;
FIG. 4 is a schematic view of a fixed barbed wire mesh secondary overlapping phenomenon;
FIG. 5 is a log-log of the frequency of occurrence of time intervals for adjacent ship sites;
FIG. 6 is a conceptual diagram of the DBSCAN algorithm;
FIG. 7 is a diagram illustrating a phenomenon that the same network is determined to be multiple networks;
FIG. 8 is a schematic diagram of the net extraction of the fixed barbed net;
FIG. 9 is a statistical chart of the net-releasing length of each net of the fixed barbed wire.
Detailed Description
The invention will be further illustrated with reference to the following specific examples. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and such equivalents may fall within the scope of the present invention as defined in the appended claims.
The embodiment of the invention relates to a VMS fixed gill net fishing boat net secondary extraction method based on DBSCAN, which comprises the following steps: counting the frequency of the instant navigational speed returned by each point, judging the net collecting and releasing states of the fixed barbed net, and extracting the ship position data in the net collecting state; and obtaining a time interval threshold value between the network levels by using the return time, calculating clustering parameters according to the navigational speed of each point and the time interval between the network levels, substituting a DBSCAN algorithm to judge the network level of the fixed barbed net, calculating the distance of each point in the network level in sequence, and extracting the length of each network level of the fixed barbed net. The method comprises the following specific steps:
counting the frequency of ship points at each navigational speed according to the navigational speed of the fixed barbed net fishing ship, finding out a frequency peak value, and obtaining valley values at two sides of the peak value as the minimum value and the maximum value of a navigational speed threshold value of the fixed barbed net collection; screening out points which are possibly in a net collecting state by using a navigational speed threshold, then calculating time intervals of adjacent ship points, counting frequency occurring in each time interval, taking logarithm of the frequency of each time interval, and finding out the minimum value when the change in a distribution diagram tends to be stable, wherein the minimum value is used as the threshold value of the time interval between nets; calculating parameters of a DBSCAN algorithm according to the time resolution of the VMS ship position data, the speed threshold value in the fixed barbed net receiving state and the time interval threshold value between network orders, and clustering the VMS ship position data to extract each network order; combining different network times belonging to the same network time by calculating the distance and time of starting and stopping points between adjacent network times; and sequentially calculating the distance between the points in each net to obtain the length of each net of the fixed barbed net.
The DBSCAN algorithm used in this embodiment is a density-based clustering method, and can find clusters of any shape, automatically determine the number of clusters, and effectively remove noise points. The method is characterized in that if the number of objects contained in a certain point and a certain area in a clustering space is not less than a given threshold value, a current object and nearby objects are marked as a class, and the same judgment is carried out on the objects which are not retrieved in the class, so that the class is expanded. Two parameters are mainly involved in the general DBSCAN algorithm: neighborhood (Eps) and neighborhood density threshold (MinPs). Because the operation points of the fixed barbed wire are gathered around the netting gear, the data of different months are often overlapped together, and therefore, the time interval T between the wires is added for restriction when the wires are judged so as to distinguish the wires.
The invention is further explained by specific embodiments, and spatial distribution of ship positions of a fixed gill net fishing ship in 2017 is shown in fig. 1, and spatial track points are formed according to time sequence. When the fixed gill-net fishing boat is operated, the fishnet fishing boat usually moves to a fishing ground at a higher speed to prepare for net release, the fishing boat sails at a higher speed when the net is released, and then the fishing boat is in anchoring or drifting waiting, and the sailing speed is basically 0 m/s. And the network collection is started after a period of time, and the network collection speed is slow and the duration is long. If the fish catch amount is less in the net collecting process, an operation mode of collecting and releasing the net is adopted, otherwise, the net is collected firstly, and the net is released after the fish is taken. The net-releasing speed of the fixed barbed wire fishing boat is higher, the speed range is overlapped with the speed range during navigation, and the speed of the net-collecting process is greatly different from that of other states, so that the net-collecting starting and stopping points of the fixed barbed wire fishing boat are expressed by the net-collecting starting and stopping points.
1. Speed frequency statistics
Counting the frequency of ship points at each navigational speed according to the navigational speed of the fixed fishnet fishing ship, and when the fishing ship receives the net, keeping the fishnet at a relatively close lower navigational speed, so that the peak value of the navigational speed frequency is found, and the valley values at two sides of the peak value are obtained and can be used as the minimum value and the maximum value of the navigational speed threshold value of the fishnet receiving; it can be considered that only the ship position with the navigational speed within the navigational speed threshold value is the ship position which is carrying out the net collecting operation. Fig. 2 is a curve of the speed frequency of the fishing vessel, the first peak is the net collecting state of the fishing vessel, the threshold values of the net collecting state are obtained on both sides of the peak, namely the minimum value is 0.4m/s, the maximum value is 2m/s, and possible net stabbing and collecting points (thick black points in fig. 3) are screened through the speed threshold value.
2. Inter-network time interval frequency statistics
According to the characteristics of fixed net stabbing operation, a fishing boat usually carries out multiple operations in a fixed sea area, and after net collection is finished, operation of immediately taking off the net is frequently carried out, so that multiple operation conditions often exist in the same sea area, ship sites among net passes densely exist at small distance intervals (figure 4), and points of different net passes can be identified into the same type when clustering is carried out through the distance between the points. To avoid such a situation, the mesh must be distinguished first. The Beidou VMS data comprises the time of each ship site, a certain time difference exists between each network of the fixed barbed net, after possible fixed barbed net collection points are extracted, the time intervals of adjacent ship sites can be calculated, and the frequency of occurrence of each time interval is counted. As shown in fig. 5, the change is observed by taking the logarithm of the frequency of the time interval. The time interval is mainly concentrated on 0-30 minutes, and the corresponding frequency shows a descending trend within 30 minutes; the time interval spans a large span after 30 minutes and the frequency fluctuates mainly in the low value range. The fixed net stabbing operation is usually to put the net or continue to sail to the next net after the net is lifted, because the operation point data in the net lifting state is extracted in the early stage, the larger value of the time interval mostly represents the time difference between the end point of the operation of the previous net and the start point of the operation of the next net, the time interval of the net is random, the corresponding frequency of any time interval is not high, and the characteristic is combined to know that the time interval of the adjacent nets in the net receiving state is mainly concentrated after 30 minutes, so the net times are distinguished by taking 30 minutes as the threshold value.
Determining DBSCAN algorithm parameters and preliminarily extracting fixed position gill net times
Selecting thresholds of 3min and 2m/s by combining the Beidou data time resolution and the speed extraction threshold, and obtaining a radius Eps of 360 m; before and after 3min, each search point should include one point, so MinPts is 3; the inter-network time interval threshold T is 30 min. The important step in the algorithm execution is to judge the time interval between the retrieval point and other points in the Eps neighborhood and count the number of effective points. Fig. 6 is a conceptual diagram of the DBSCAN algorithm, in which a core point indicates that there are no less than MinPts in the Eps neighborhood of the point. As shown in point a, the neighborhood includes 5 points (including point a), and since the time interval of point a is greater than T, the number of actual valid points is 4, and thus it is determined as a core point. A boundary point indicates that the point is in the neighborhood of some core point, but is not itself a core point, as shown by point B; the noise points are points with less than 3 effective points in the neighborhood. The core points are mostly network-time internal points, and the boundary points are network-time starting and stopping points.
In the algorithm execution process, the distance between the surrounding point and the core point needs to be calculated, whether the distance is within 360m or not is judged, and the earth is an irregular sphere with two-stage parts which are slightly flat, so that the earth can be generally used as a sphere for calculation when a geodetic coordinate system is calculated. Therefore, between two points on the earth with known longitude and latitude, the distance can be calculated by using a spherical distance formula. Suppose two ship positions A and B with longitude and latitude Alon，Alat，Blon，Blat. The distance D between the two points A, B can be obtained by using the Haversine formula:
wherein R is the mean radius of the earth 6378145 m.
In this embodiment, a DBSCAN algorithm is implemented by using Java programming, and the minimum and maximum values of each type of clustered time are extracted as the starting point and the ending point of the network, and stored in the MySQL database, where the table structure is as follows.
TABLE 1 fixed gill net number extracting table
4. Mesh merging
After the preliminary extraction of the net times, it is found that the ship position information may be lost due to the problems of time slot collision and the like, and a complete net time is divided into a plurality of adjacent net times (fig. 7) due to the distance between points during clustering, so that errors occur in statistics of the net times. In order to reduce the misjudgment of the network times, the starting point and the ending point of the current network time are respectively calculated with the ending point of the previous network time and the starting point of the next network time to obtain the distance and the time interval, if the time interval is less than 30min and the distance between the two points is less than the product of the time interval and the maximum speed value (2 m/s in the example), the two network times are combined, and finally the extracted network times are shown in fig. 8.
5. Fixed gill net sub-length extraction
And sequentially accumulating the space distances of two adjacent points according to the longitude and latitude of the ship position point in each network during network collection to obtain the network length of the single network of the fixed gill net fishing ship. The fishing boat 2017 in the example has 201 meshes in total all the year, and the length of each mesh is shown in FIG. 9.
Claims (4)
1. A VMS fixed gill net fishing boat net number extraction method based on DBSCAN is characterized by comprising the following steps:
(1) counting the frequency of the instant navigational speed returned by each point, judging the net collecting and releasing states of the fixed barbed net, and extracting the ship position data in the net collecting state;
(2) obtaining a time interval threshold value between the grids by using the return time, calculating clustering parameters according to the navigational speed of each point and the time interval between the grids, substituting a DBSCAN algorithm to judge the grids of the fixed barbed net, calculating the distance of each point in the grids in sequence, and extracting the length of each grid of the fixed barbed net; the method specifically comprises the following steps: calculating parameters of a DBSCAN algorithm according to the time resolution, the speed threshold value in the fixed net stabbing collection state and the time interval threshold value between the net stabbing modes, clustering VMS ship position data to extract each net time, judging whether the adjacent net times belong to the same net time or not by calculating the distance and the time of starting and stopping points between the adjacent net times, combining the adjacent net times if the adjacent net times belong to the same net time, sequentially calculating the distance of each point in the combined net times, and extracting the length of the net times of single net stabbing.
2. The method for extracting the net times of the VMS fixed gill-net fishing boat based on the DBSCAN according to claim 1, wherein the step (1) comprises the following steps: counting the frequency of ship points at each navigational speed according to the navigational speed of the fixed barbed wire fishing ship, finding out a frequency peak value, obtaining valley values at two sides of the peak value as the minimum value and the maximum value of a navigational speed threshold value of the fixed barbed wire net collection, and screening out the points which are possibly in the net collection state by utilizing the navigational speed threshold value.
3. The method for extracting the net times of the VMS fixed gill-net fishing boat based on DBSCAN according to claim 2, wherein the obtaining of the inter-net time interval threshold value using the return time in step (2) is specifically: and calculating the time interval of the adjacent ship positions in the points possibly in the network receiving state, counting the frequency of the occurrence of each time interval, taking logarithm of the time interval frequency, and finding out the minimum value when the change in the distribution diagram tends to be stable as the threshold value of the time interval between networks.
4. The method for extracting the net times of the VMS fixed net stabbing fishing boat based on the DBSCAN in the step (2), wherein the radius range in the clustering parameters of the DBSCAN algorithm is the product of the time resolution and the speed threshold value in the net receiving state of the fixed net stabbing fishing boat, and the number of designated points in the radius is 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811405919.3A CN109389186B (en) | 2018-11-23 | 2018-11-23 | VMS fixed gill net fishing boat net number extraction method based on DBSCAN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811405919.3A CN109389186B (en) | 2018-11-23 | 2018-11-23 | VMS fixed gill net fishing boat net number extraction method based on DBSCAN |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109389186A CN109389186A (en) | 2019-02-26 |
| CN109389186B true CN109389186B (en) | 2022-03-18 |
Family
ID=65429470
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811405919.3A Active CN109389186B (en) | 2018-11-23 | 2018-11-23 | VMS fixed gill net fishing boat net number extraction method based on DBSCAN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109389186B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110047331A (en) * | 2019-05-22 | 2019-07-23 | 镇江市地方海事局 | A kind of distance exam method between ship and object based on AIS |
| CN111199103B (en) * | 2019-12-30 | 2023-07-18 | 中国船舶重工集团公司第七一一研究所 | Full-flow automatic calculation navigational speed optimization method and system for full-electric propulsion ship |
| CN112434465B (en) * | 2020-11-19 | 2021-11-09 | 江苏省海洋水产研究所 | Method for extracting effective net length of shrimp net based on ship position data |
| CN113987838A (en) * | 2021-11-19 | 2022-01-28 | 上海海洋大学 | Method for extracting light net laying times based on ship position data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678902A (en) * | 2013-12-05 | 2014-03-26 | 中国水产科学研究院东海水产研究所 | Method for extracting trawl fishing state from Beidou ship position data |
| CN105404648A (en) * | 2015-10-29 | 2016-03-16 | 东北大学 | Density and closeness clustering based user moving behavior determination method |
| CN108427824A (en) * | 2018-01-29 | 2018-08-21 | 中国水产科学研究院东海水产研究所 | A method of utilizing accommodation data extraction gill net net length and angle |
-
2018
- 2018-11-23 CN CN201811405919.3A patent/CN109389186B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678902A (en) * | 2013-12-05 | 2014-03-26 | 中国水产科学研究院东海水产研究所 | Method for extracting trawl fishing state from Beidou ship position data |
| CN105404648A (en) * | 2015-10-29 | 2016-03-16 | 东北大学 | Density and closeness clustering based user moving behavior determination method |
| CN108427824A (en) * | 2018-01-29 | 2018-08-21 | 中国水产科学研究院东海水产研究所 | A method of utilizing accommodation data extraction gill net net length and angle |
Non-Patent Citations (1)
| Title |
|---|
| "基于DBSCAN的VMS定置刺网渔船网次提取方法研究";原作辉等;《2018年中国水产学会学术年会论文摘要集》;20181115;摘要 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109389186A (en) | 2019-02-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109389186B (en) | VMS fixed gill net fishing boat net number extraction method based on DBSCAN | |
| CN108427824B (en) | Method for extracting length and angle of gill net by using ship position data | |
| CN108415975B (en) | BDCH-DBSCAN-based taxi passenger carrying hot spot identification method | |
| CN114444819B (en) | Fishery resource prediction method and device, storage medium and electronic equipment | |
| CN112434465B (en) | Method for extracting effective net length of shrimp net based on ship position data | |
| KR102401337B1 (en) | Method for Identification of Fishing Gear Type using Trajectory Data | |
| CN103837891A (en) | High-precision pickup method of first arrival of microseism | |
| CN109919514B (en) | Transformer area line loss abnormity detection method and device based on line loss marker post threshold value | |
| CN112784180A (en) | Method for extracting catching strength spatial information of tuna seine fishing boat | |
| CN109817023A (en) | A kind of novel sea cable waters AIS object detection method | |
| CN112926468B (en) | Tidal flat elevation automatic extraction method | |
| CN114936332A (en) | Method and device for analyzing ship traffic flow in water area of wind power plant and storage medium | |
| CN113052235A (en) | Method and device for detecting abnormal stay of ship | |
| Zimmermann | Calculation of untrawlable areas within the boundaries of a bottom trawl survey | |
| CN114139608A (en) | Fishing boat behavior identification method based on multi-step clustering of AIS (automatic identification System) track data | |
| CN113093119A (en) | Time-frequency constant false alarm high-frequency radar target detection method and system | |
| CN104484826A (en) | Method for extracting trawl fishing effort temporal and spatial characteristics based on Beidou vessel monitoring system data | |
| Chen et al. | Mapping the fishing intensity in the coastal waters off Guangdong province, China through AIS data | |
| CN104463433B (en) | A kind of method based on fishing boat accommodation data extraction gill net fishing state | |
| CN116562467B (en) | Marine fish target species identification and distribution prediction method and system | |
| CN115952410A (en) | Landslide hazard detection system based on deep learning | |
| CN115309948A (en) | Fishing boat operation method based on automatic motion trail identification | |
| CN113360544A (en) | Short-time catching Knoop force distribution prediction method based on catching time sequence relation | |
| CN114819278A (en) | Method for improving prediction precision of east Pacific squid fishery based on middle-layer hydrological features and ship position state | |
| Pei et al. | Spatial distribution of fishing intensity of canvas stow net fishing vessels in the East China Sea and the Yellow Sea |
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 |