Ship double-towing fishing behavior discrimination system and method based on track spatio-temporal characteristics

Abstract

The invention provides a ship double-towing fishing behavior discrimination system and method based on track spatio-temporal characteristics. The method comprises the following steps: acquiring track data of a target ship and unifying data formats; filtering static targets and small-range wandering targets in the track data; acquiring information of an alarm area, and filtering targets which are not in the alarm area in the track data; setting a parameter threshold according to the track data to judge a ship double-towing behavior; and storing the track data conforming to the double-towing behavior of the ship and the calculated feature data in a database. According to the method, whether the navigational speed and the navigational direction of the two ships are matched or not is directly judged by combining basic parameters such as the navigational speed and the navigational direction, spatial-temporal characteristic analysis is further introduced, the double-towing fishing behavior can be efficiently and simply recognized, factors such as duration and the relative position are considered, and the judgment accuracy and reliability are enhanced; the method does not depend on machine learning or model training, but carries out judgment based on a rule set based on experience, so that adaptability and high efficiency in different environments are ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular, relates to a system and method for distinguishing ship double-tow fishing behavior based on track temporal and spatial characteristics. Background Art

[0002] As a traditional fishing method, double tug fishing is widely used in offshore fisheries and some inland waters. In the process of double tug fishing, two fishing vessels work together through trawl nets and use specific trawl equipment to catch fish in the waters. This behavior usually involves two ships sailing in parallel at similar speeds and headings in the sea area, maintaining a consistent relative position for a long time. Due to its special operation method, double tug fishing has a high fishing efficiency, but it also has the risk of illegal operation, especially in no-fishing areas or fishery resource protection areas. Illegal double tug fishing is often difficult to detect and stop in time.

[0003] At present, the monitoring and early warning technology for double-tow fishing mainly relies on traditional manual inspections and basic track data analysis. However, with the continuous expansion of the operating area and the improvement of monitoring requirements, traditional methods face the problems of limited coverage and slow response speed when dealing with vast sea areas and complex navigation environments. In addition, since the track behavior of double-tow fishing vessels is very similar to that of ordinary ships, traditional methods are difficult to effectively distinguish and identify the collaborative operation behavior of the two ships, resulting in low recognition accuracy and efficiency.

[0004] Patent document "Method and device for identifying double-tow behavior of fishing vessels" (CN116776112A) discloses filtering real-time trajectory stream data, obtaining trajectory stream data to be analyzed after filtering, and processing the data set generated after data processing, inputting the trained target spatiotemporal graph convolutional network model, and obtaining the double-tow behavior identification result of fishing vessels output by the target spatiotemporal graph convolutional network model. However, it still relies on machine learning or model training, the judgment process is relatively complicated, and the applicability is low.

[0005] Therefore, a new double-trawling fishing behavior identification system and method is needed, which can monitor double-trawling fishing behavior in real time in key areas such as no-fishing zones, provide timely and reliable early warning information to fishery management departments, help protect fishery resources, and promote the maintenance of ecological balance. Summary of the invention

[0006] In view of the defects in the prior art, the object of the present invention is to provide a system and method for distinguishing ship double-tow fishing behavior based on the spatiotemporal characteristics of the track.

[0007] According to the present invention, a method for distinguishing double-tow fishing behavior of a ship based on the spatiotemporal characteristics of the track is provided, comprising:

[0008] Step S1: Acquire the track data of the target ship and unify the data format;

[0009] Step S2: filtering stationary targets and small-range wandering targets in the track data;

[0010] Step S3: Obtain information about the warning area and filter out targets in the track data that are not in the warning area;

[0011] Step S4: setting parameter thresholds according to track data to determine the double-tow behavior of the ship;

[0012] Step S5: Save the track data that conforms to the double-tow behavior of the ship and the calculated data to the database.

[0013] Preferably, in step S1, the trajectory data of the ship is obtained through radar, AIS, and Beidou equipment to form a list stored in chronological order, including the ship's unique identification, heading, speed, longitude and latitude coordinates, captain and time, and the original 2-second interval track points are reduced to a 30-second sampling interval.

[0014] In step S2, the combined weight of the target navigation length, the activity range and the center point deviation is calculated to filter the target.

[0015] The information of the warning area is the custom-set regional longitude and latitude data and parameter information of heading range, speed range, time period, ship type and captain size.

[0016] In step S4, the target ship is analyzed in real time, the track data of the two targets are traversed at the same time, the double towing behavior of the ship is judged in combination with the set empirical rules, and an alarm is issued in the alarm area;

[0017] The double towing behavior of ships is that two parallel ships in the warning area are sailing at a speed lower than the set threshold;

[0018] The parallel ships set the track points whose azimuth, distance, speed difference and heading difference meet the set threshold and whose time difference is within one minute as parallel points, accumulate the distance and duration of the continuous parallel point tracks, and the track data that meets the set parallel threshold.

[0019] In step S5, the distance, heading difference, speed difference and relative azimuth angle of the track points calculated by the track data determined as the double towing behavior of the ship are stored in the database and displayed in the alarm list of the front-end web page.

[0020] Preferably, in step S2, a static threshold td for determining whether the target is a static target is set to 0.02°, and the target track data is traversed to obtain four attributes of the maximum longitude maxLon, the minimum longitude minLon, the maximum latitude maxLat, and the minimum latitude minLat of the target activity longitude and latitude range, and the average longitude avgLon and the average latitude avgLat of the target.

[0021] Calculate the target spatial distribution range d1 = (maxLon + maxLat) - (minLon + minLat), the target track point compactness d2 = min(maxLon - avgLon, avgLon - minLon) + min(maxLat - avgLat, avgLat - maxLat), and the effective distance length of the target track

[0022] Combine d1, d2, and d3 to define the weight and calculate d1×0.1 + d2×0.3 + d3×0.6. If the calculation result is less than the static threshold td, it is considered a static target or a target wandering in a small range and is filtered. If the calculation result is greater than or equal to the static threshold td, no operation is performed.

[0023] Among them, the threshold pd = 0.002°;

[0024] n represents the length of the track;

[0025] i and j respectively represent the i-th and j-th track points;

[0026] span(i, j) represents the distance between points i and j;

[0027] lat i represents the latitude of the i-th track point;

[0028] lon i represents the longitude of the i-th track point;

[0029] lat j represents the latitude of the j-th track point;

[0030] lon j represents the longitude of the j-th track point.

[0031] Preferably, step S4 includes:

[0032] Step S4.1: Set the time difference Δt = |t1 - t2|. If Δt ≤ 60s, execute step S4.2. If Δt > 60s, adjust the matching index according to the time difference, and judge t1 and t2. If t1 ≥ t2, increment offset1 by one. If t1 < t2, increment offset2 by one;

[0033] Among them, t1 and t2 represent the time of the first and second targets respectively;

[0034] Δt is the time difference between the two targets;

[0035] offset1 and offset2 represent the offsets of the first and second target track points respectively.

[0036] Step S4.2: Calculate the distance to the target track point Heading Difference Speed ​​difference and relative azimuth angle adiff = atan2(sin(Δlon)·cos(lat1)·sin(lat2)-sin(lat1)·cos(lat2)·cos(Δlon));

[0037] Among them, Δlat = |lat1-lat2| and Δlon = |lon1-lon2I represent the latitude difference and longitude difference respectively;

[0038] Lat1 and lat2 represent the latitude of the first and second points;

[0039] lon1 and lon2 represent the longitudes of the first and second points;

[0040] R represents the radius of the earth;

[0041] Course1 and course2 indicate the courses of the first and second ships;

[0042] Δθ represents the heading difference.

[0043] Step S4.3: Determine the parallel conditions, including distance conditions, heading difference conditions, speed difference conditions, and relative azimuth difference conditions. If the parallel conditions are met, execute step S4.4. If the parallel conditions are not met, discard it.

[0044] Step S4.4: Calculate the parallel distance D parallel For each string of continuous parallel points of a target, the geographical distance is calculated. If the parallel distance exceeds the set parallel threshold, the two ships are considered to be in parallel and it is determined that there is a double towing behavior of the ships. If the parallel distance does not exceed the set parallel threshold, it is discarded.

[0045] Preferably, the distance condition is that the distance d between the two ships satisfies D min ≤d≤D max .

[0046] The heading difference condition is that the heading difference Δθ between the two ships is less than or equal to the maximum heading difference threshold: Δθ≤Δθmax .

[0047] The speed difference condition is that the speed difference between the two ships should be less than or equal to the maximum speed difference threshold: |speed1-speed2|≤Δ vmax .

[0048] The relative azimuth difference condition is that the difference of the relative azimuth adiff is within a preset range: adiff min ≤

[0049] adiff|≤adiff max .

[0050] Among them, D min , D max Respectively and represent the minimum and maximum distance;

[0051] Δθ max Indicates the maximum heading difference threshold;

[0052] speed1 and speed2 represent the speeds of the first and second ships respectively;

[0053] Δ vmax represents the maximum speed difference threshold;

[0054] adiff min 、adiff max They represent the minimum and maximum differences of relative azimuth respectively.

[0055] According to the present invention, a ship double-tow fishing behavior identification system based on track spatiotemporal characteristics includes:

[0056] Module M1: Obtain the track data of the target ship and unify the data format;

[0057] Module M2: Filter stationary targets and small-scale wandering targets in track data;

[0058] Module M3: Obtain the information of the warning area and filter the targets that are not in the warning area in the track data;

[0059] Module M4: Setting parameter thresholds based on track data to determine the double-tow behavior of the ship;

[0060] Module M5: Save the track data consistent with the double-tow behavior of the ship and the calculated data to the database.

[0061] Preferably, the module M1 obtains the trajectory data of the ship through radar, AIS, and Beidou equipment to form a list stored in chronological order, including the ship's unique identification, heading, speed, longitude and latitude coordinates, captain and time, and reduces the original 2-second interval track points to a 30-second sampling interval.

[0062] The module M2 calculates the combined weight of the target navigation length, activity range and center point deviation, and filters the target.

[0063] The information of the warning area is the custom-set regional longitude and latitude data and parameter information of heading range, speed range, time period, ship type and captain size.

[0064] The module M4 analyzes the target ship in real time, traverses the track data of the two targets at the same time, judges the double towing behavior of the ship in combination with the set empirical rules, and issues an alarm in the alarm area;

[0065] The double towing behavior of ships is that two parallel ships in the warning area are sailing at a speed lower than the set threshold;

[0066] The parallel ships set the track points whose azimuth, distance, speed difference and heading difference meet the set threshold and whose time difference is within one minute as parallel points, accumulate the distance and duration of the continuous parallel point tracks, and the track data that meets the set parallel threshold.

[0067] In the module M5, the distance, heading difference, speed difference and relative azimuth angle of the track points calculated by the track data determined as the double towing behavior of the ship are stored in the database and displayed in the alarm list of the front-end web page.

[0068] Preferably, in the module M2, a stationary threshold td=0.02° is set for judging whether the target is a stationary target, and the target track data is traversed to obtain the four attributes of the target activity longitude and latitude range, namely, the maximum longitude maxLon, the minimum longitude minLon, the maximum latitude maxLat, the minimum latitude minLat, and the average longitude avgLon and the average latitude avgLat of the target.

[0069] Calculate the target spatial distribution range d1 = (maxLon + maxLat) - (minLon + minLat), the target track point density d2 = min (maxLon - avgLon, avgLon - minLon) + min (maxLat - avgLat, avgLat - maxLat) and the effective distance length of the target track

[0070] The combination of d1, d2, and d3 defines the weight calculation d1×0.1+d2×0.3+d3×0.6. If the calculation result is less than the still threshold td, it is considered to be a still target or a small-range wandering target and is filtered. If the calculation result is greater than or equal to the still threshold td, no operation is performed.

[0071] Wherein, the threshold value pd = 0.002°;

[0072] n represents the length of the track;

[0073] i and j respectively represent the i-th and j-th track points;

[0074] span(i, j) represents the distance between points i and j;

[0075] lat i represents the latitude of the i-th track point;

[0076] lon i represents the longitude of the i-th track point;

[0077] lat j represents the latitude of the j-th track point;

[0078] lon j represents the longitude of the j-th track point.

[0079] Preferably, the module M4 includes:

[0080] Module M4.1: Set the time difference Δt = |t1 - t2|. If Δt ≤ 60s, then execute Module M4.2. If Δt > 60s, then adjust the matching index according to the time difference, judge t1 and t2. If t1 ≥ t2, then increment offset1 by one. If t1 < t2, then increment offset2 by one;

[0081] where, t1 and t2 respectively represent the times of the 1st and 2nd targets;

[0082] Δt is the time difference between the two targets;

[0083] offset1 and offset2 respectively represent the offsets of the 1st and 2nd target track points.

[0084] Module M4.2: Calculate the distance of the target track point Course difference Speed difference and relative azimuth angle adiff = atan2(sin(Δlon)·cos(lat1)·sin(lat2) - sin(lat1)·cos(lat2)·cos(Δlon));

[0085] where, Δlat = |lat1 - lat2|, Δlon = |lon1 - lon2| respectively represent the latitude difference and longitude difference;

[0086] lat1 and lat2 represent the latitudes of the 1st and 2nd points;

[0087] lon1 and lon2 represent the longitudes of the first and second points;

[0088] R represents the radius of the earth;

[0089] Course1 and course2 indicate the courses of the first and second ships;

[0090] Δθ represents the heading difference.

[0091] Module M4.3: Determine the parallel conditions, including distance conditions, heading difference conditions, speed difference conditions, and relative azimuth difference conditions. If the parallel conditions are met, execute module M4.4. If the parallel conditions are not met, discard it.

[0092] Module M4.4: Calculation of parallel distance D parallel For each string of continuous parallel points of a target, the geographical distance is calculated. If the parallel distance exceeds the set parallel threshold, the two ships are considered to be in parallel and it is determined that there is a double towing behavior of the ships. If the parallel distance does not exceed the set parallel threshold, it is discarded.

[0093] Preferably, the distance condition is that the distance d between the two ships satisfies D min ≤d≤D max .

[0094] The heading difference condition is that the heading difference Δθ between the two ships is less than or equal to the maximum heading difference threshold: Δθ≤Δθ max .

[0095] The speed difference condition is that the speed difference between the two ships should be less than or equal to the maximum speed difference threshold: |speed1-speed2|≤Δ vmax .

[0096] The relative azimuth difference condition is that the difference of the relative azimuth adiff is within a preset range: adiff min ≤|adiff|≤adiff max .

[0097] Among them, D min , D max Respectively and represent the minimum and maximum distance;

[0098] Δθ max Indicates the maximum heading difference threshold;

[0099] speed1 and speed2 represent the speeds of the first and second ships respectively;

[0100] Δ vmax represents the maximum speed difference threshold;

[0101] adiffmin 、adiff max They represent the minimum and maximum differences of relative azimuth respectively.

[0102] Compared with the prior art, the present invention has the following beneficial effects:

[0103] 1. The present invention combines basic parameters such as speed and heading, and further introduces spatiotemporal feature analysis, so that the determination method is concise and effective.

[0104] 2. The present invention can efficiently identify double-tow fishing behavior by directly judging whether the speeds and headings of the two ships match, and takes factors such as duration and relative position into consideration, thereby enhancing the accuracy and reliability of the judgment.

[0105] 3. The present invention does not rely on machine learning or model training, but makes judgments based on rules set by experience, ensuring adaptability and efficiency in different environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0107] Figure 1 This is a schematic diagram of the architecture of the ship double-tow fishing behavior identification system based on the spatiotemporal characteristics of the track;

[0108] Figure 2 It is a flow chart of the method for distinguishing the double-tow fishing behavior of ships based on the temporal and spatial characteristics of the track;

[0109] Figure 3 Schematic diagram of the double-tow fishing target of ships based on the temporal and spatial characteristics of the track. DETAILED DESCRIPTION

[0110] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0111] According to the present invention, a method for distinguishing double-tow fishing by a ship based on the spatiotemporal characteristics of the track is provided. The distinguishing process includes obtaining track data, determining the geographical information and related configuration data of the fishing area, filtering out ships in normal navigation, excluding irrelevant track data, analyzing and calculating the double-tow fishing characteristics, and finally judging the track characteristics according to preset rules. Figure 2 For example, the following steps are included:

[0112] Step S1: Obtain the track point data of radar, AIS and Beidou targets and unify the data format.

[0113] The ship's trajectory data is obtained through radar, AIS, Beidou and other equipment, and the data is analyzed to determine whether the target ship is engaging in double trawling fishing.

[0114] Specifically, the fields required by the following algorithm are taken out. The ship track data includes the ship's unique identification, heading, speed, latitude and longitude coordinates, ship length, and time. The track data of each ship is a list stored in chronological order. The original track points with an interval of 2 seconds are reduced to track data with a sampling interval of 30 seconds to reduce the amount of calculation.

[0115] Step S2: Target filtering: filtering stationary targets and targets wandering in a small range to reduce the amount of calculation data.

[0116] Specifically, stationary targets and small-range wandering targets are filtered out. The track of stationary targets will cause the track points to shift due to the fluctuation of the sampling track, resulting in an increase in the target track activity range. Therefore, a combined weight analysis of the target navigation length, activity range and center point deviation is used to exclude such target tracks.

[0117] In more preferred embodiments, the specific calculation steps are:

[0118] Calculate the target spatial distribution range d1, the target track point density d2, and the target track effective distance length d3, and set the stationary threshold td = 0.02° (the geographical distance in China is approximately equal to 2000 meters) to determine whether the target is a stationary target. The units of d1, d2, and d3 are also longitude and latitude degrees, so as to omit the calculation amount of distance conversion.

[0119] Traverse the target track data to obtain the four attributes of the target's activity longitude and latitude range, namely, the maximum longitude maxLon, the minimum longitude minLon, the maximum latitude maxLat, the minimum latitude minLat, and the target's average longitude avgLon and average latitude avgLat.

[0120] Calculate the target spatial distribution range d1: Calculate the difference between the maximum and minimum latitude and longitude of the track to estimate the spatial distribution range of the entire track:

[0121] d1=(maxLon+maxLat)-(minLon+minLat)

[0122] Calculate the density d2 of target track points: measure the distribution of track points and evaluate the density of track point distribution. The d2 value of stationary targets and anchored targets is smaller, while the value of normal navigation is larger:

[0123] d2=min(maxLon-avgLon,avgLon-minLon)+min(maxLat-avgLat,avgLat

[0124] -maxLat)

[0125] Calculate the effective distance length d3 of the target track: set a threshold pd = 0.002°. Only when the distance between two points exceeds the preset pd will this distance be added to d3. The purpose is to prevent the stationary target from calculating an excessively long track length.

[0126]

[0127] Where n represents the length of the track, i and j represent the i-th and j-th track points respectively. span(i,j) represents the distance between two points. lat i Indicates the latitude of the i-th track point, lon i Indicates the longitude of the i-th track point, lat j Indicates the latitude of the jth track point, lon j Represents the longitude of the j-th track point.

[0128] Finally, d1, d2, and d3 are combined to define the weight calculation: if the calculation result is less than the static threshold td, it is considered to be a static target or a small range wandering target and filtered:

[0129] d1×0.1+d2×0.3+d3×0.6 <td。

[0130] Step S3: Acquire the coordinate data and related configuration information of the warning area, and filter out the targets that are not in the warning area.

[0131] Specifically, by obtaining customized regional longitude and latitude data, you can focus on filtering targets in specific waters and obtain relevant configuration information: heading range, speed range, time period, ship type, captain size and other parameters to further filter target data.

[0132] Step S4: Determination of the double-tow behavior of ships, including ships running in parallel, sailing at a low speed, and area limitation, and setting thresholds of parameters related to the parallel relative positions: angle, distance, and duration.

[0133] The characteristic of double tug fishing is that in a specific water area, at least two fishing vessels operate in parallel at a low speed, usually maintaining a nearly consistent heading and speed, and the track is characterized by long-term parallel low-speed movement. Unlike other fishing behaviors, double tug fishing relies heavily on the coordinated actions of the two vessels, and the two vessels usually maintain a consistent distance and relative speed for a long time.

[0134] By analyzing the track behavior characteristics and track data of the target ship in real time, and combining with the set empirical rules, it is determined whether there is a behavior that conforms to the characteristics of pair trawling fishing operations, and an alarm is issued in the fishing prohibited area. It does not rely on machine learning or model training, but makes judgments based on the rules set by experience, ensuring adaptability and efficiency in different environments.

[0135] Specifically, for the determination of the behavior of pair trawling fishing boats, based on empirical statistics, the main characteristics of the target are that two ships sail in parallel at a low speed and cast nets for fishing. Therefore, on this basis, an analysis of the target is carried out. The target speed can be set through the alarm configuration information. For the determination of ship parallelism, the tracks of the two targets are traversed simultaneously, and the relative positions are calculated for the track points with a time difference within one minute. The track points whose relative positions, including azimuth angle, distance, speed difference, and course difference, meet the set thresholds are determined as parallel points. Finally, the distances and durations of each continuous parallel track are accumulated, and the target track that meets the set threshold is determined as parallel, and the target that meets the above conditions is determined as a pair trawling fishing boat.

[0136] In more preferred examples, the specific implementation process is as follows:

[0137] In order to compare the parallel states of two ships, first, the time differences of the track points of the two ships need to be similar. The set threshold is 1 minute, that is, only when the time difference Δt = |t1 - t2| of the two ships does not exceed one minute, these two points are considered to be matched. If the time difference is greater than 60 seconds, that is, Δt > 60s, the matching index is adjusted according to the time difference. Compare t1 and t2. If t1 ≥ t2, then offset1 is incremented by one. If t1 < t2, then offset2 is incremented by one.

[0138] Among them, t1 and t2 are the times of the two targets respectively, Δt is the time difference between the two targets, and offset1 and offset2 represent the offsets of the first and second target track points.

[0139] Calculate the distance, course difference, speed difference, and relative azimuth angle between the track points of the two targets. The distance calculation uses the Haversine formula:

[0140]

[0141] Among them, lon1, lon2, lat1, and lat2 are the longitudes and latitudes of the two points, R is the radius of the earth, in meters, and Δlat and Δlon are the latitude difference and longitude difference respectively.

[0142] The course difference is calculated and the problem of crossing zero is processed:

[0143]

[0144] Among them, course1 and course2 represent the courses of the first and second ships, and Δθ is the difference in the courses of the two ships.

[0145] Relative bearing angle calculation, calculate the direction from the first ship to the second ship. This feature can be used to observe whether the relative bearing of the two ships is stable:

[0146] adiff=atan2(sin(Δlon)·cos|lat1)·sin(lat2)-sin(lat1)·cos(lat2)·cos(Δlon))

[0147] The parallel judgment conditions include distance condition, heading difference condition, speed difference condition, and relative azimuth difference condition. Distance condition: The distance d between the two ships should satisfy:

[0148] D min ≤d≤D max

[0149] Heading difference condition: The heading difference between the two ships should be less than or equal to the maximum heading difference threshold:

[0150] Δθ≤Δθ max

[0151] Speed ​​difference condition: The speed difference between the two ships should be less than or equal to the maximum speed difference threshold:

[0152] |speed1-speed2|≤Δ vmax

[0153] Relative azimuth difference condition: The relative azimuth difference should be within the preset range to ensure that the target parallel is always on the same side and does not cross parallel:

[0154] adiff min ≤|adiff|≤adiff max

[0155] Among them, D min , D max are the minimum and maximum distances, Δθ max is the maximum heading difference, Δ vmax is the maximum speed difference, adiff min 、adiff max is the minimum and maximum difference of relative azimuth.

[0156] Under the above conditions, calculate the parallel distance D between the two ship tracks. parallel ,For each string of continuous parallel points of a target, the geographical distance between them is calculated. If the parallel distance exceeds the set threshold, the two ships are considered to be in a parallel state, and further it is determined that there may be double trawling fishing behavior.

[0157] Step S5: Save the double-drag target track data and feature data into a database.

[0158] The analysis results will be stored for subsequent regulatory use to monitor and prevent double-trawling fishing, especially in closed areas or important fishery resource protection areas.

[0159] Specifically, for fishing vessels that are judged to be double-trawling, their calculated feature data are stored in the database as training data for the thresholds of the subsequent optimization algorithm, and are displayed in the alarm list of the front-end web page, so that the track drawing display of such targets can be queried, helping relevant law enforcement personnel to better analyze the purpose of the ship.

[0160] The present invention also provides a ship double-tow fishing identification system based on the track temporal and spatial characteristics. The ship double-tow fishing identification system based on the track temporal and spatial characteristics can be realized by executing the process steps of the ship double-tow fishing identification method based on the track temporal and spatial characteristics, that is, technical personnel in this field can understand the ship double-tow fishing identification method based on the track temporal and spatial characteristics as the preferred implementation of the ship double-tow fishing identification system based on the track temporal and spatial characteristics.

[0161] According to the present invention, a ship double-tow fishing behavior identification system based on track spatiotemporal characteristics includes:

[0162] Module M1: Obtain the track data of the target ship and unify the data format;

[0163] Module M2: Filter stationary targets and small-scale wandering targets in track data;

[0164] Module M3: Obtain the information of the warning area and filter the targets that are not in the warning area in the track data;

[0165] Module M4: Setting parameter thresholds based on track data to determine the double-tow behavior of the ship;

[0166] Module M5: Save the track data consistent with the double-tow behavior of the ship and the calculated data to the database.

[0167] In more preferred examples, the module M1 obtains the ship's trajectory data through radar, AIS, and Beidou equipment to form a list stored in chronological order, including the ship's unique identification, heading, speed, longitude and latitude coordinates, captain and time, and reduces the original 2-second interval track points to a 30-second sampling interval.

[0168] The module M2 calculates the combined weight of the target navigation length, activity range and center point deviation, and filters the target.

[0169] The information of the warning area is the custom-set regional longitude and latitude data and parameter information of heading range, speed range, time period, ship type and captain size.

[0170] The module M4 analyzes the target ship in real time, traverses the track data of the two targets at the same time, judges the double-tow behavior of the ship in combination with the set empirical rules, and issues an alarm in the alarm area;

[0171] The double towing behavior of ships is that two parallel ships in the warning area are sailing at a speed lower than the set threshold;

[0172] The parallel ships set the track points whose azimuth, distance, speed difference and heading difference meet the set threshold and whose time difference is within one minute as parallel points, accumulate the distance and duration of the continuous parallel point tracks, and the track data that meets the set parallel threshold.

[0173] In the module M5, the distance, heading difference, speed difference and relative azimuth angle of the track points calculated by the track data determined as the double towing behavior of the ship are stored in the database and displayed in the alarm list of the front-end web page.

[0174] In more preferred examples, in the module M2, a stationary threshold td=0.02° is set for judging whether the target is a stationary target, and the target track data is traversed to obtain the four attributes of the target's activity longitude and latitude range, namely, the maximum longitude maxLon, the minimum longitude minLon, the maximum latitude maxLat, and the minimum latitude minLat, and the target's average longitude avgLon and average latitude avgLat.

[0175] Calculate the target spatial distribution range d1 = (maxLon + maxLat) - (minLon + minLat), the target track point density d2 = min (maxLon - avgLon, avgLon - minLon) + min (maxLat - avgLat, avgLat - maxLat) and the effective distance length of the target track

[0176] The weight calculation of d1×0.1+d2×0.3+d3×0.6 is defined by combining d1, d2, and d3. If the calculated result is less than the stationary threshold td, it is considered to be a stationary target or a small-range wandering target and is filtered. If the calculated result is greater than or equal to the stationary threshold td, no operation is performed.

[0177] Wherein, the threshold value pd = 0.002°;

[0178] n represents the length of the track;

[0179] i and j represent the i-th and j-th track points respectively;

[0180] span(i, j) represents the distance between points i and j;

[0181] lat i represents the latitude of the i-th track point;

[0182] lon i represents the longitude of the i-th track point;

[0183] lat j represents the latitude of the j-th track point;

[0184] lon j represents the longitude of the j-th track point.

[0185] In more preferred examples, the module M4 includes:

[0186] Module M4.1: Set the time difference Δt = |t1 - t2|. If Δt ≤ 60s, then execute module M4.2. If Δt > 60s, then adjust the matching index according to the time difference, and judge t1 and t2. If t1 ≥ t2, then increment offset1 by one. If t1 < t2, then increment offset2 by one;

[0187] where t1 and t2 respectively represent the times of the 1st and 2nd targets;

[0188] Δt is the time difference between the two targets;

[0189] offset1 and offset2 respectively represent the offsets of the 1st and 2nd target track points.

[0190] Module M4.2: Calculate the distance of the target track point Course difference Speed difference and relative azimuth angle adiff = atan2(sin(Δlon)·cos(lat1)·sin(lat2) - sin(lat1)·cos(lat2)·cos(Δlon));

[0191] where Δlat = |lat1 - lat2|, Δlon = |lon1 - lon2| respectively represent the latitude difference and longitude difference;

[0192] lat1 and lat2 represent the latitudes of the 1st and 2nd points;

[0193] lon1 and lon2 represent the longitudes of the 1st and 2nd points;

[0194] R represents the radius of the earth;

[0195] Course1 and course2 indicate the courses of the first and second ships;

[0196] Δθ represents the heading difference.

[0197] Module M4.3: Determine the parallel conditions, including distance conditions, heading difference conditions, speed difference conditions, and relative azimuth difference conditions. If the parallel conditions are met, execute module M4.4. If the parallel conditions are not met, discard it.

[0198] Module M4.4: Calculation of parallel distance D parallel For each string of continuous parallel points of a target, the geographical distance is calculated. If the parallel distance exceeds the set parallel threshold, the two ships are considered to be in parallel and it is determined that there is a double towing behavior of the ships. If the parallel distance does not exceed the set parallel threshold, it is discarded.

[0199] In more preferred embodiments, the distance condition is that the distance d between the two ships satisfies D min ≤d≤D max .

[0200] The heading difference condition is that the heading difference Δθ between the two ships is less than or equal to the maximum heading difference threshold: Δθ≤Δθ max .

[0201] The speed difference condition is that the speed difference between the two ships should be less than or equal to the maximum speed difference threshold: |speed1-speed2|≤Δ vmax .

[0202] The relative azimuth difference condition is that the difference of the relative azimuth adiff is within a preset range: adiff min ≤

[0203] adiff|≤adiff max .

[0204] Among them, D min , D max Respectively and represent the minimum and maximum distance;

[0205] Δθ max Indicates the maximum heading difference threshold;

[0206] speed1 and speed2 represent the speeds of the first and second ships respectively;

[0207] Δ vmax represents the maximum speed difference threshold;

[0208] adiff min 、adiff max They represent the minimum and maximum differences of relative azimuth respectively.

[0209] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0210] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for distinguishing double-tow fishing by ship based on the temporal and spatial characteristics of track, characterized in that: Including: Step S1: Obtain the track data of the target ship and unify the data format; Step S2: Filter stationary targets and small-range wandering targets in the track data; Step S3: Obtain the information of the warning area and filter the targets not in the warning area in the track data; Step S4: Set parameter thresholds according to the track data to judge the double-towing behavior of the ship; Step S5: Save the track data conforming to the double-towing behavior of the ship and the calculated data to the database.

2. The method for distinguishing double-tow fishing by ship based on the temporal and spatial characteristics of track according to claim 1, characterized in that: In the said step S1, the track data of the ship is obtained through radar, AIS, and Beidou devices, and a list stored in chronological order is formed, including the unique identification of the ship, course, speed, longitude and latitude coordinates, ship length, and time. The original track points with a 2-second interval are downsampled to a 30-second sampling interval; In the said step S2, the combined weights of the target navigation length, activity range, and deviation degree from the center point are calculated to filter the targets; The information of the warning area is the parameter information of the custom-set area longitude and latitude data, course range, speed range, time period, ship type, and ship length; In the said step S4, the target ship is analyzed in real time, and the track data of two targets is traversed simultaneously. The double-towing behavior of the ship is judged in combination with the set empirical rules, and an alarm is issued in the warning area; The double-towing behavior of the ship is that two parallel ships in the warning area are both sailing at a speed lower than the set threshold; The parallel ships are the track points with the azimuth angle, distance, speed difference, and course difference conforming to the set threshold and the time difference within one minute. The distance and duration of the continuous parallel point tracks are accumulated, and the track data conforming to the set parallel threshold; In the said step S5, the distance, course difference, speed difference, and relative azimuth angle of the track points calculated from the track data determined to be the double-towing behavior of the ship are stored in the database and displayed in the warning list on the front-end web page.

3. The method for distinguishing double-tow fishing by ship based on the temporal and spatial characteristics of track according to claim 1, characterized in that: In the said step S2, the stationary threshold td for judging whether the target is a stationary target is set to 0.02°. The four attributes of the maximum longitude maxLon, minimum longitude minLon, maximum latitude maxLat, and minimum latitude minLat of the target activity longitude and latitude range and the average longitude avgLon and average latitude avgLat of the target are obtained by traversing the target track data; Calculate the target spatial distribution range d1 = (maxLon + maxLat) - (minLon + minLat), the target track point density d2 = min (maxLon - avgLon, avgLon - minLon) + min (maxLat - avgLat, avgLat - maxLat) and the effective distance length of the target track The combination of d1, d2, and d3 is defined as the weight to calculate d1×0.1 + d2×0.3 + d3×0.

6. If the calculation result is less than the stationary threshold td, it is considered a stationary target or a small-range wandering target and is filtered. If the calculation result is greater than or equal to the stationary threshold td, no operation is performed; Among them, the threshold pd = 0.002°; n represents the length of the track; i and j respectively represent the i-th and j-th track points; span(i,j) represents the distance between points i and j; lat i represents the latitude of the i-th track point; lon i represents the longitude of the i-th track point; lat j represents the latitude of the jth track point; lon j Represents the longitude of the j-th track point.

4. The method for distinguishing double-tow fishing by ship based on the temporal and spatial characteristics of track according to claim 1, characterized in that: The said step S4 includes: Step S4.1: Set the time difference Δt = |t1 - t2|. If Δt ≤ 60s, execute step S4.

2. If Δt > 60s, adjust the matching index according to the time difference, judge t1 and t2. If t1 ≥ t2, increment offset1 by one. If t1 < t2, increment offset2 by one; Among them, t1 and t2 represent the time of the first and second targets respectively; Δt is the time difference between the two targets; offset1 and offset2 represent the offsets of the first and second target track points respectively; Step S4.2: Calculate the distance to the target track point Heading Difference Speed ​​difference and relative azimuth angle adiff = atan2(sin(Δlon)·cos(lat1)·sin(lat2)-sin(lat1)·cos(lat2)·cos(Δlon)); Among them, Δlat = |lat1-lat2| and Δlon = |lon1-lon2| represent the latitude difference and longitude difference respectively; Lat1 and lat2 represent the latitude of the first and second points; lon1 and lon2 represent the longitudes of the first and second points; R represents the radius of the earth; Course1 and course2 indicate the courses of the first and second ships; Δθ represents the heading difference; Step S4.3: Determine the parallel conditions, including distance conditions, heading difference conditions, speed difference conditions, and relative azimuth difference conditions. If the parallel conditions are met, execute step S4.

4. If the parallel conditions are not met, discard the step. Step S4.4: Calculate the parallel distance D parallel For each string of continuous parallel points of a target, the geographical distance is calculated. If the parallel distance exceeds the set parallel threshold, the two ships are considered to be in parallel and it is determined that there is a double towing behavior of the ships. If the parallel distance does not exceed the set parallel threshold, it is discarded.

5. The method for distinguishing double-tow fishing by ship based on the temporal and spatial characteristics of track according to claim 4, characterized in that: The distance condition is that the distance d between the two ships satisfies D min ≤d≤D max ; The heading difference condition is that the heading difference Δθ between the two ships is less than or equal to the maximum heading difference threshold: Δθ≤Δθ max ; The speed difference condition is that the speed difference between the two ships should be less than or equal to the maximum speed difference threshold: |speed1-speed2|≤Δ vmax ; The relative azimuth difference condition is that the difference of the relative azimuth adiff is within a preset range: adiff min ≤|adiff|≤adiff max ; Among them, D min , D max Respectively and represent the minimum and maximum distance; Δθ max Indicates the maximum heading difference threshold; speed1 and speed2 represent the speeds of the first and second ships respectively; Δ vmax represents the maximum speed difference threshold; adiff min 、adiff max They represent the minimum and maximum differences of relative azimuth respectively.

6. A ship double-tow fishing identification system based on track temporal and spatial characteristics, characterized in that: include: Module M1: Obtain the track data of the target ship and unify the data format; Module M2: Filter stationary targets and small-scale wandering targets in track data; Module M3: Obtain the information of the warning area and filter the targets that are not in the warning area in the track data; Module M4: Setting parameter thresholds based on track data to determine the double-tow behavior of the ship; Module M5: Save the track data consistent with the double-tow behavior of the ship and the calculated data to the database.

7. The ship double-tow fishing identification system based on track temporal and spatial characteristics according to claim 6 is characterized in that: The module M1 obtains the ship's track data through radar, AIS, and Beidou equipment to form a list stored in chronological order, including the ship's unique identification, heading, speed, longitude and latitude coordinates, length, and time, and reduces the original 2-second interval track points to a 30-second sampling interval; The module M2 calculates the combined weight of the target navigation length, the activity range and the center point deviation, and filters the target; The information of the warning area is the custom-set regional latitude and longitude data and parameter information of heading range, speed range, time period, ship type and captain size; The module M4 analyzes the target ship in real time, traverses the track data of the two targets at the same time, judges the double-tow behavior of the ship in combination with the set empirical rules, and issues an alarm in the alarm area; The double towing behavior of ships is that two parallel ships in the warning area are sailing at a speed lower than the set threshold; The parallel ships are track data that meets the set parallel threshold by setting the track points whose azimuth, distance, speed difference and heading difference meet the set threshold and whose time difference is within one minute. The distance and duration of the continuous parallel point tracks are accumulated, and the track data meets the set parallel threshold. In the module M5, the distance, heading difference, speed difference and relative azimuth angle of the track points calculated by the track data determined as the double towing behavior of the ship are stored in the database and displayed in the alarm list of the front-end web page.

8. The ship double-tow fishing identification system based on track temporal and spatial characteristics according to claim 6 is characterized in that: In the module M2, a static threshold td for determining whether the target is a static target is set to 0.02°. Traverse the target track data to obtain four attributes of the maximum longitude maxLon, minimum longitude minLon, maximum latitude maxLat, and minimum latitude minLat of the target activity longitude and latitude range, and the average longitude avgLon and average latitude avgLat of the target. Calculate the target spatial distribution range d1 = (maxLon + maxLat) - (minLon + minLat), the target track point density d2 = min (maxLon - avgLon, avgLon - minLon) + min (maxLat - avgLat, avgLat - maxLat) and the effective distance length of the target track Define the weights by combining d1, d2, and d3 as d1×0.1 + d2×0.3 + d3×0.

6. If the calculation result is less than the static threshold td, it is considered a static target or a target with small - range wandering and is filtered. If the calculation result is greater than or equal to the static threshold td, no operation is performed. Among them, the threshold pd = 0.002°. n represents the length of the track. i and j respectively represent the i - th and j - th track points. span(i, j) represents the distance between points i and j. lat i represents the latitude of the i-th track point; lon i represents the longitude of the i-th track point; lat j represents the latitude of the jth track point; lon j Represents the longitude of the j-th track point.

9. The ship double-tow fishing identification system based on track temporal and spatial characteristics according to claim 6, characterized in that: The module M4 includes: Module M4.1: Set the time difference Δt = |t1 - t2|. If Δt ≤ 60s, execute module M4.

2. If Δt > 60s, adjust the matching index according to the time difference, and judge t1 and t2. If t1 ≥ t2, increment offset1 by one. If t1 < t2, increment offset2 by one. Among them, t1 and t2 respectively represent the times of the first and second targets. Δt is the time difference between two targets. offset1 and offset2 respectively represent the offsets of the track points of the first and second targets. Module M4.2: Calculate the distance to the target track point Heading Difference Speed ​​difference and relative azimuth angle adiff = atan2(sin(Δlon)·cos(lat1)·sin(lat2)-sin(lat1)·cos(lat2)·cos(Δlon)); Among them, Δlat = |lat1 - lat2| and Δlon = |lon1 - lon2| respectively represent the latitude difference and longitude difference. lat1 and lat2 represent the latitudes of the first and second points. lon1 and lon2 represent the longitudes of the first and second points. R represents the radius of the earth. course1 and course2 represent the headings of the first and second ships. Δθ represents the heading difference. Module M4.3: Judge the parallel conditions, including distance conditions, heading difference conditions, speed difference conditions, and relative azimuth difference conditions. If the parallel conditions are met, execute module M4.

4. If the parallel conditions are not met, discard. Module M4.4: Calculation of parallel distance D parallel For each string of continuous parallel points of a target, the geographical distance is calculated. If the parallel distance exceeds the set parallel threshold, the two ships are considered to be in parallel and it is determined that there is a double towing behavior of the ships. If the parallel distance does not exceed the set parallel threshold, it is discarded.

10. The ship double-tow fishing identification system based on track temporal and spatial characteristics according to claim 9, characterized in that: The distance condition is that the distance d between the two ships satisfies D min ≤d≤D max ; The heading difference condition is that the heading difference Δθ between the two ships is less than or equal to the maximum heading difference threshold: Δθ≤Δθ max ; The speed difference condition is that the speed difference between the two ships should be less than or equal to the maximum speed difference threshold: |speed1-speed2|≤Δ vmax ; The relative azimuth difference condition is that the difference of the relative azimuth adiff is within a preset range: adiff min ≤adiff|≤adiff max ; Among them, D min , D max Respectively and represent the minimum and maximum distance; Δθ max Indicates the maximum heading difference threshold; speed 1 and speed2 respectively represent the speeds of the first and second ships. Δ vmax represents the maximum speed difference threshold; adiff min 、adiff max They represent the minimum and maximum differences of relative azimuth respectively.

Images