A marine fishery resource dynamic monitoring management system based on multi-data fusion

Abstract

The application relates to the technical field of fishery management, and discloses a marine fishery resource dynamic monitoring management system based on multi-data fusion, which comprises the following modules: a data acquisition module, which is used for acquiring various data in a monitoring area in a preset acquisition mode through an unmanned ship carrying multiple sensors, wherein the various data comprises detection sound wave intensity, specific frequency signals, signal intensity, water temperature of each layer and chlorophyll concentration; a preprocessing module, which is used for carrying out clear denoising on the acquired various data, unifying the formats, and generating corresponding data labels according to the acquired various data; and a data analysis module, which is used for calculating the total number of spawning fish groups of a target fish group according to the acquired detection sound wave intensity and the water volume of the monitoring area. Comprehensive data is acquired through the unmanned ship carrying multiple sensors in combination with marked fish tracking, designated acquisition and random acquisition, and the total number of spawning fish groups and the complete migration route of the target fish group are calculated by the data analysis module.

Description

Technical Field

[0001] This invention relates to the field of fisheries management technology, specifically to a dynamic monitoring and management system for marine fisheries resources based on multi-data fusion. Background Technology

[0002] Dynamic monitoring of marine fishery resources utilizes technologies such as acoustic detection, remote sensing, and GIS to track and systematically analyze the quantity, distribution, growth, and environmental parameters of fishery resources in real time. Through long-term data accumulation, assessment models are built to predict resource change trends, providing a scientific basis for fishing quota setting, ecological restoration, and the supervision of closed fishing zones.

[0003] However, traditional monitoring relies heavily on fixed-point sampling and manual statistics. It also relies on fixed monitoring stations in the near sea and periodic sampling by research vessels. Fixed monitoring stations are limited by their deployment range and can only cover shallow sea areas close to the shore. In the open sea, due to the large waves and deep water, it is difficult to deploy fixed equipment and the maintenance cost is extremely high. Research vessels, on the other hand, have limited navigation range and high cost per voyage, making it impossible to achieve normalized and high-frequency coverage of the open sea. This naturally creates monitoring blind spots, resulting in data coverage with blind spots in the open sea and statistical lag, making it impossible to capture fish migration and spawning dynamics in real time.

[0004] To address this, the present invention proposes a dynamic monitoring and management system for marine fishery resources based on multi-data fusion, in order to overcome the shortcomings of existing technologies. Summary of the Invention

[0005] The purpose of this invention is to provide a dynamic monitoring and management system for marine fishery resources based on multi-data fusion.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A dynamic monitoring and management system for marine fishery resources based on multi-data fusion, characterized by comprising the following modules:

[0008] The data acquisition module is used to acquire various data within the monitoring area by using an unmanned vessel carrying multiple sensors in accordance with a preset acquisition mode. These data include sound wave intensity, specific frequency signal, signal strength, water temperature at each layer, and chlorophyll concentration.

[0009] The preprocessing module performs noise reduction and format standardization on the acquired data, and generates corresponding data labels based on the collected data.

[0010] The data analysis module calculates the total number of spawning fish in the target fish population based on the intensity of the detected sound waves and the volume of the water in the monitored area, and analyzes the migration routes of the target fish population based on the sound wave frequency, water temperature, and chlorophyll concentration.

[0011] The management decision-making module formulates fishing restriction plans in real time based on the total number of spawning fish and the migration routes of the target fish population.

[0012] Preferably, the plurality of sensors include an acoustic sensor, a temperature sensor, an echo detector, and a chlorophyll fluorometer, and the data acquisition module performs the following operations:

[0013] A random number of target fish are caught within the monitoring area, acoustic transmitters are implanted into the caught target fish, the target fish with implanted acoustic transmitters are released back into the monitoring area, and an unmanned boat is deployed into the monitoring area.

[0014] The preset acquisition modes include specified acquisition and random acquisition:

[0015] The monitoring area is divided into several grid units, each of which is numbered. A data collection sequence is formed according to the numbering order. Based on the data collection sequence, a cruise path is planned for the unmanned surface vessel (USV). When the USV cruises along the planned path, a designated data collection mode is activated.

[0016] The acoustic wave sensor continuously acquires the acoustic waves emitted by the target fish school's acoustic transmitter, and the echo detector scans within the standard frequency band. The acoustic wave intensity within the grid unit is collected at preset intervals. The water temperature and chlorophyll concentration data of the surface, middle and bottom layers within the grid unit are periodically collected by the temperature sensor and chlorophyll fluorometer.

[0017] After the unmanned surface vessel completes its cruise along the planned path and collects data from each designated cell, it returns to the starting grid cell and switches to random data collection mode.

[0018] An automatic data collection cycle is set, allowing the unmanned surface vessel (USV) to drift randomly with ocean currents. Random data collection is performed during the drifting process. When the USV drifts out of the monitoring area, the distance between the USV and the nearest grid cell is calculated, and a return path for the USV is planned to bring it back into the detection area.

[0019] Preferably, the preprocessing module's functions include:

[0020] The data from multiple data points acquired through specified and random sampling are denoised and the data format is standardized.

[0021] Acquire latitude and longitude coordinate change data of unmanned surface vessel during specified and random data collection processes. Based on the collection time of multiple data, locate the latitude and longitude coordinates of the unmanned surface vessel when the data is collected, and compare the coordinates of the unmanned surface vessel with the latitude and longitude coordinate range of each grid unit to identify the corresponding grid unit.

[0022] Generate corresponding data labels based on the specified and randomly collected data for each grid cell;

[0023] If the unmanned surface vessel (USV) collects random data in non-monitored areas after floating out of the monitoring area, then the grid cells with missing random data are obtained. Based on the distance between the latitude and longitude coordinates of the random data collected by the USV in non-monitored areas and the grid cells with missing random data, the estimated data of the grid cells with missing random data is calculated.

[0024] The estimated data is used as random data collected from the missing random data grid cells, and the corresponding data labels are generated by combining the specified collected data.

[0025] Preferably, the valuation data acquisition method includes:

[0026] Using the center point of the grid cell with missing random data as the reference point, calculate the straight-line distance between the latitude and longitude coordinates of each random data collection point in the non-monitoring area and the reference point;

[0027] The distance attenuation coefficient is obtained from historical data. The distance attenuation coefficient is used as the exponent of the straight-line distance. The reciprocal of the straight-line distance after being amplified by the distance attenuation coefficient is used as the weighting coefficient of the latitude and longitude coordinates when collecting random data in non-monitoring areas.

[0028] Water temperature estimates for missing randomly collected data grid cells are obtained based on water temperature data from non-monitored areas and their corresponding weighting coefficients.

[0029] The gradient correction coefficient is obtained based on the latitude and longitude coordinates when collecting chlorophyll concentration data in non-monitoring areas. The obtained chlorophyll concentration data in non-monitoring areas is multiplied by the product of the gradient correction coefficient and the straight-line distance, and then multiplied by the corresponding weight coefficient to obtain the estimated chlorophyll concentration data of missing random data grid cells.

[0030] Based on the natural attenuation of sound waves propagating in water, an attenuation coefficient is set. The sound wave intensity value of the non-monitored area is multiplied by the negative attenuation coefficient of the natural constant and the power of the straight-line distance, and then multiplied by the corresponding weighting coefficient to obtain the estimated sound wave intensity data of the missing random data grid cell.

[0031] Preferably, the data analysis module's functions include:

[0032] Based on the acoustic intensity data in the data labels of each grid unit, and combined with the acoustic feature library of the target fish group, the acoustic signals of the spawning fish group in the target fish group are screened out. The calibration coefficient is obtained by comparing historical fishing sampling data with acoustic data of the same period. The density of the spawning fish group per unit water volume is obtained by combining the calibration coefficient with the screened acoustic intensity data.

[0033] Calculate the total water volume within the monitoring area, multiply the spawning fish density per unit water volume by the total water volume, and obtain the total number of spawning fish in the target fish population.

[0034] Based on the acoustic transmitter signals of the target fish school in the data tags of each grid cell, the latitude and longitude coordinates of the signals at different time points are obtained to form the movement trajectory of the target fish school. The consistent movement path is selected from multiple movement trajectories as the basic reference for the migration route.

[0035] Based on the acquired sound wave frequency and intensity distribution data, heat maps of the target fish population density at different time points are generated. By comparing the direction of the center of gravity movement of the heat maps at adjacent time points, the overall movement trend of the population is determined.

[0036] Based on the water temperature, chlorophyll concentration data and fish distribution data of each grid unit, the correlation between fish movement and environmental factors was identified.

[0037] By combining the movement trajectory of the target fish population, the overall movement trend of the group, and the correlation patterns of environmental factors, a complete migration route can be obtained.

[0038] Preferably, the density of spawning fish per unit water volume is obtained by the following method:

[0039] The total scattering intensity of sound waves per unit volume of water is obtained based on historical data, and the average scattering intensity of sound waves per spawning fish in the target fish group is obtained.

[0040] The initial spawning fish density per unit volume is obtained by subtracting the scattering intensity of a single spawning fish from the total scattering intensity and then performing a logarithmic transformation.

[0041] By comparing historical data from the same period, the acoustic density of a certain area is calculated and compared with the density of the actual count of fixed-point fishing in the same area during the same period. The basic calibration coefficient is obtained by the ratio of the actual density to the acoustic density.

[0042] The density of spawning fish per unit water volume is obtained by combining the initial spawning fish density with the calibration coefficient.

[0043] Preferably, the overall movement trend of the group is obtained through the following methods:

[0044] Acquire the acoustic frequency and intensity data of the target fish group, form multiple continuous time series data groups at fixed time intervals, and obtain the target fish density value of each grid cell based on the acoustic intensity data in each time series data group.

[0045] Based on the fish density values ​​of each grid cell, gradient color levels are divided to generate a heat map of the target fish population density at each time point. The center point of each grid cell is defined, and the density centroid of the entire monitoring area is calculated using the density value of the grid cell as the weight. The density centroid of the heat map at each time point is calculated in sequence to obtain the centroid coordinate sequence.

[0046] On a latitude and longitude map, connect the coordinates of the center of gravity in chronological order to form a trajectory line of the center of gravity movement, and calculate the direction angle between two adjacent centers of gravity.

[0047] If the directional consistency of at least three consecutive adjacent centroids is statistically analyzed, and the angular deviation of adjacent directions that is greater than the standard threshold is less than the preset standard angle, then the direction is determined to be the dominant movement direction in a given stage.

[0048] Calculate the straight-line distance between adjacent centroids to obtain the movement distance in each time period. Obtain the movement speed of the target fish group based on the movement distance and time interval. If the movement speed fluctuation is less than the preset speed standard value for at least three consecutive time periods and the dominant direction remains unchanged, the overall movement trend of the group is determined to be stable. If the speed fluctuation exceeds the preset speed standard value or the direction changes abruptly, the original data needs to be backtracked, abnormal grids are excluded, and the calculation is recalculated.

[0049] By combining the dominant direction of movement, average speed, and center of gravity trajectory, the overall movement trend of the group can be obtained.

[0050] Preferably, the work content of the management decision module includes:

[0051] Based on the total number of spawning fish in the target fish population and the stability of their migration routes, the target fish populations are classified into four levels: safe, attention, warning, and emergency. Corresponding fishing restrictions are then specified for each level.

[0052] When the safety level is set, the allowable catch standard is set according to the spawning volume, and fishing is prohibited only during the spawning period;

[0053] When the target level is raised, the total allowable catch is reduced, and the no-fishing zone along the migration route is expanded;

[0054] When the alert level is reached, the permitted catch limit is the minimum standard for spawning.

[0055] When the emergency level is reached, all commercial fishing will be banned, and ecological restoration will be carried out simultaneously.

[0056] The beneficial effects of this invention are:

[0057] 1. This invention utilizes an unmanned surface vessel equipped with multiple sensors to acquire comprehensive data through tagging fish tracking, designated collection, and random data collection. The data is then preprocessed by a module that removes noise, standardizes formats, generates tags, and performs spatial correlation estimation to complete the data. The data analysis module then accurately calculates the total number of spawning fish based on an acoustic feature library and historical calibration coefficients, and analyzes the complete migration route by combining the target fish trajectory with environmental factors. Finally, the management decision-making module provides tiered early warnings and dynamically formulates fishing plans. This invention overcomes the shortcomings of existing technologies, which rely heavily on fixed-point sampling and manual statistics, have blind spots in the open ocean, and suffer from statistical lag, making it impossible to capture the real-time dynamics of fish migration and spawning.

[0058] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0059] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This is a schematic diagram of the module framework of a dynamic monitoring and management system for marine fishery resources based on multi-data fusion according to the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Please see Figure 1 As shown, this invention is a dynamic monitoring and management system for marine fishery resources based on multi-data fusion, including a data acquisition module, a preprocessing module, a data analysis module, and a management decision module. The data acquisition module is mainly used to collect various data of target fish groups within the marine monitoring area. Specifically, the data includes sound wave intensity, specific frequency signal, signal strength, water temperature at each layer, and chlorophyll concentration. By catching a random number of target fish groups from the monitoring area, implanting acoustic transmitters into the caught target fish groups, releasing the target fish groups with implanted acoustic transmitters back into the monitoring area, and deploying an unmanned vessel into the monitoring area, the unmanned vessel is equipped with multiple sensors, including a sound wave sensor, a temperature sensor, an echo detector, and a chlorophyll fluorometer.

[0063] The monitoring area is divided into several grid units with a precision of 1km×1km. The latitude and longitude range of each grid is recorded and numbered by region + serial number, such as East-001. A continuous data collection sequence is formed based on the numbering. Then, a cruise path is planned for the unmanned vessel.

[0064] While the drone cruises along the planned path, a designated data acquisition mode is activated. The acoustic sensor continuously captures specific frequency signals and signal strengths from the acoustic emitters in the target fish school. The echo sounder collects the acoustic intensity of the grid cells every 2 minutes within the standard frequency band of 30kHz-200kHz. The temperature sensor and chlorophyll fluorometer synchronously collect chlorophyll concentration and water temperature at the surface, middle, and bottom layers of the grid cells every 5 minutes.

[0065] After the unmanned surface vessel completes the designated data collection for the entire area, it automatically returns to the starting grid cell and switches to random data collection mode. At this time, the sensor automatic data collection cycle is set, and the unmanned surface vessel floats naturally with the ocean currents in the monitoring area. During the floating process, it records its own latitude and longitude coordinates in real time and collects data periodically. If the unmanned surface vessel drifts out of the boundary of the monitoring area, it calculates the straight-line distance to the nearest grid cell and the return route, and controls itself to return to the monitoring area autonomously. During the return process, the sensor continues to collect data normally.

[0066] The preprocessing module is mainly used to format the collected data. For multiple data points such as sound wave intensity, specific frequency signal, signal strength, water temperature at each layer, and chlorophyll concentration acquired by specified and random sampling, it first performs noise reduction processing to eliminate invalid information such as equipment interference and environmental noise. Then, it converts heterogeneous data from different sensors, such as the decibel value of sound wave data and the Celsius value of water temperature, into a unified data format preset by the system. For example, the original output of the echo detector may be a string format with units, such as -62.5dB and -58dB. The system's preset format is a pure numerical value with one decibel, and the implicit unit is decibel. After conversion, it is uniformly -62.5 and -58.0. The temperature sensor may output numerical values ​​with different precisions or a signed format, such as 18.36℃, 20℃, and -1.2℃. The preset format is a pure numerical value with one decimal place. After conversion, it is uniformly 18.4, 20.0, and -1.2.

[0067] Subsequently, the latitude and longitude coordinate changes of the unmanned vessel were recorded under the two data acquisition modes. Based on the specific acquisition time of each data item, the precise latitude and longitude coordinates of the unmanned vessel at the time of data acquisition were located. Then, the coordinates were compared with the latitude and longitude range of each grid unit in the monitoring area to obtain the corresponding grid unit to which each data item belongs.

[0068] For each grid cell, its corresponding designated and randomly collected data are integrated to generate a unique data tag containing the grid number, collection time, data type, and numerical value. For example: East-005: sound wave intensity -61.3dB, specific frequency signal 38HKz, signal strength -48.5dB, surface water temperature 20.2℃, middle water temperature 18.7℃, bottom water temperature 16.5℃, chlorophyll concentration 4.8mg / m³ 3 ;

[0069] If the unmanned surface vessel floats out of the monitoring area during random data collection and collects data from the non-monitoring area, then the grid cells in the monitoring area that are missing random data are selected. Based on the latitude and longitude coordinates when the data from the non-monitoring area was collected, the estimated data of the missing grid cells is calculated and obtained. The estimated data is used as the random data collected for the missing grid cells, and then combined with the original designated data collected for the grid cells to generate complete data labels.

[0070] The estimated data for monitoring areas lacking random data collection is first calculated using the geometric center point of the missing data grid cell as the reference point. Latitude and longitude coordinates are converted: 1° latitude ≈ 111km, 1° longitude ≈ 111km × cos latitude. This accurately calculates the straight-line distance between the latitude and longitude coordinates of all valid random data collection points within the non-monitoring area and the reference point. Then, a distance attenuation coefficient obtained from historical data calibration is introduced. This coefficient is used as the exponent of the straight-line distance to calculate the weighting coefficient of the latitude and longitude coordinates of each random data collection point within the non-monitoring area, using the formula:

[0071]

[0072] in, For the first The initial weighting coefficients of latitude and longitude coordinates during the random data collection in each non-monitored area. No. The straight-line distance between the latitude and longitude coordinates and the reference point during the collection of valid random data. This is the distance attenuation coefficient; to ensure that the weighting coefficients of latitude and longitude coordinates during the random data collection in non-monitored areas are summed to one, for all... The actual weight coefficients are obtained by normalization: Ensure that all weight coefficients add up to one;

[0073] The distance attenuation coefficient is calibrated based on historical observation patterns. It converts the straight-line distance between the target point and the data collection point into a weighted proportion of the data's estimation of the target point. The closer the distance, the higher the reference value of the data and the greater the weight; conversely, the farther the distance, the lower the reference value and the smaller the weight. In marine fisheries monitoring scenarios, due to the relatively continuous water environment but interference from ocean currents, the distance attenuation coefficient is set between 1.5 and 2.5. The more complex the environment, the larger the coefficient and the faster the attenuation rate. For example, if a grid cell lacking water temperature data has a non-monitored area with water temperature data of 20℃ within 2km of it, then the distance attenuation coefficient is set to 2. In this case, the initial weighting coefficient... If the sum of the initial weighting coefficients of all other randomly collected data in the non-monitored area equals 0.9, then the actual weighting coefficient of the water temperature data in the non-monitored area is:

[0074] Non-monitored data points that are closer to the benchmark have a higher weight, and vice versa, ensuring that data with strong spatial correlation has a greater impact on valuation.

[0075] The estimation calculation methods vary depending on the characteristics of different types of parameters. Water temperature estimation is obtained directly by multiplying the water temperature data of non-monitoring areas with the corresponding weighting coefficient. Since the water temperature is relatively uniformly distributed in a small space, no additional correction is required.

[0076] Chlorophyll concentration estimation needs to consider the natural gradient difference from nearshore to offshore. First, based on the latitude and longitude coordinates of non-monitoring data points, corresponding gradient correction coefficients are matched from the historical environmental database. Then, the following formula is used:

[0077]

[0078] in, Estimate chlorophyll concentration. The concentration of chlorophyll in the non-monitoring area. These are gradient correction coefficients. No. The straight-line distance between the latitude and longitude coordinates and the reference point during the collection of valid random data;

[0079] The gradient correction coefficient is derived from historical monitoring data. Generally, chlorophyll concentration decreases from nearshore to offshore, resulting in a positive coefficient. Water temperature decreases with increasing latitude, resulting in a negative coefficient. The more pronounced the gradient, the larger the absolute value of the coefficient. For example, in a certain monitoring area, chlorophyll concentration shows a stable decreasing trend from nearshore to offshore. Historical data shows that for every 1 km away from the shoreline, the average chlorophyll concentration decreases by 0.03 mg / m³. 3 Therefore, the gradient correction factor is set to 0.03 mg;

[0080] For example: Chlorophyll concentration data is missing from randomly collected cells in the eastern -008 grid of the monitoring area. Chlorophyll concentration data exists at collection points A (2km away) and B (3km away) in non-monitoring areas. Therefore, the estimated chlorophyll concentration data is calculated using the chlorophyll concentration data from collection points A and B. The specific calculation process is as follows:

[0081] The straight-line distance between the center point of the East-008 grid and sampling point A is 2 km, and the chlorophyll concentration at sampling point A is 5.2 mg / m³. 3 The straight-line distance between the center point of the East-008 grid and sampling point B is 3km, and the chlorophyll concentration at sampling point B is 5.6mg / m³. 3 Then the weight coefficient of sampling point A Weighting coefficient of sampling point B The distance attenuation coefficient is set to 2, and the gradient correction coefficient is set to 0.03, according to the formula. ;

[0082] Thus, the estimated chlorophyll concentration data for the eastern-008 grid cell was obtained as follows: ;

[0083] The estimation of sound wave intensity needs to take into account the natural attenuation characteristics of sound waves propagating in water, introducing a preset attenuation coefficient, and using the formula:

[0084]

[0085] in, To estimate the intensity of the sound wave, Estimate the sound wave intensity in the non-monitoring area. It is a natural constant. It has natural decay characteristics. No. The straight-line distance between the latitude and longitude coordinates and the reference point during the collection of valid random data;

[0086] Natural attenuation refers to the inherent property of matter and energy propagating in a medium, where their intensity, concentration, and other physical quantities naturally decrease with distance or time due to interactions with the medium, such as absorption, scattering, and friction. When sound waves propagate in water, some energy is absorbed by water molecules and scattered by plankton or suspended particles, causing their intensity to gradually weaken with increasing propagation distance. The attenuation rate can be reflected by a preset attenuation coefficient, for example:

[0087] Assuming that the western-011 grid cell within the monitoring area lacks randomly collected acoustic intensity data, while there are two non-monitoring area collection points nearby: collection point C and collection point D, the acoustic intensity estimation calculation process for the monitoring area is as follows:

[0088] The straight-line distance between the center point (reference point) of the western-011 grid and sampling point C is 4km, and the original sound wave intensity at sampling point C is -52dB; the straight-line distance between the center point and sampling point D is 6km, and the original sound wave intensity at sampling point D is -58dB;

[0089] Through the formula: ;

[0090] ;

[0091] Through the formula: ;

[0092] Therefore, the estimated missing acoustic intensity of West-011 is approximately decibel.

[0093] The data analysis module is mainly used to obtain the total number of spawning fish in the target fish group and the complete migration route. Based on historical data, it obtains the total scattering intensity of sound waves per unit volume of water and the average scattering intensity of sound waves per spawning fish in the target fish group.

[0094] The initial spawning fish density per unit volume is obtained by subtracting the scattering intensity of a single spawning fish from the total scattering intensity and then performing a logarithmic transformation.

[0095] By comparing historical data from the same period, the acoustic density of a certain area is calculated and compared with the density of the actual count of fixed-point fishing in the same area during the same period. The basic calibration coefficient is obtained by the ratio of the actual density to the acoustic density.

[0096] The spawning fish density per unit water volume is obtained by combining the initial spawning fish density with the calibration coefficient, using the formula:

[0097]

[0098] in, This refers to the density of spawning fish per unit volume of water. The total scattering intensity of sound waves per unit volume of water. The average scattering intensity of sound waves by a single spawning fish in the target fish group. The reference value for sound wave scattering intensity is set based on the minimum effective scattering intensity in historical data. The acoustic density of a certain area during the same historical period is defined as the initial spawning fish density per unit volume in that area during the same historical period. This represents the actual density of fixed-point fishing in the area during the same historical period. The initial spawning density of the fish population. Basic calibration coefficients;

[0099] For example: Assuming the total scattering intensity of sound waves per unit volume of water in the historical data of the monitoring area... The average scattering intensity of sound waves by a single spawning fish in the target fish population is 80 per meter per spherical degree. The acoustic scattering intensity is set at 3 per meter per steradian, and the reference value is set at 1 per meter per steradian based on the minimum effective scattering intensity in historical data, representing the acoustic density of a certain area during the same historical period. The concentration was 6.2 tails per cubic meter.

[0100] The total scattering intensity of sound waves per unit volume of water in the monitoring area The average scattering intensity of sound waves by a single spawning fish in the target fish group is 95 per meter per spherical degree. The reference value for sound wave scattering intensity is set based on the minimum effective scattering intensity from historical data. Consistent with historical data;

[0101] The acoustic density for the same historical period is: ;

[0102] The basic calibration coefficient is: ;

[0103] The current initial spawning fish density in the monitored area is: ;

[0104] The spawning fish density per unit water volume is obtained by combining the initial spawning fish density with the calibration coefficient.

[0105] 6.30;

[0106] That is, the current density of spawning fish per unit volume of water in the monitored area is approximately 6.30 fish / m3;

[0107] Meanwhile, the total area is calculated based on the latitude and longitude range of the monitoring area, and the average water depth is obtained by averaging the water depth data collected by the unmanned vessel. Then, the total water volume of the monitoring area is calculated. Finally, the unit density is multiplied by the total water volume to obtain the total number of fish spawning in the target fish population.

[0108] The acoustic density of a certain area during the same historical period is the initial spawning fish density per unit volume in that area during the same historical period. The initial spawning fish density is a name for the density calculated using only acoustic data at present, while the acoustic density of the same historical period is a name for the density calculated using only acoustic data at the same historical period. Both are calculated based on acoustic data, use logarithmic transformation, and have the same uncalibrated preliminary values. Therefore, the acoustic density of the same historical period is essentially the initial spawning fish density of that area during that period. It is by comparing this historical initial density with the historical actual density that the basic calibration coefficient for correcting the current initial density can be calculated.

[0109] To determine the migration routes of target fish groups, we first extract specific frequency signals from the acoustic transmitters of the target fish groups from the data tags, record the latitude and longitude coordinates corresponding to the signals at different collection times, and form the movement trajectory of a single marked fish. Then, we perform overlap statistics on the trajectories of multiple marked fish, and filter out common trajectories with the same direction and path as the basic reference for the migration routes.

[0110] Acquire the acoustic frequency and intensity data of the target fish group, form multiple continuous time series data groups at fixed time intervals, and obtain the target fish density value of each grid cell based on the acoustic intensity data in each time series data group.

[0111] Based on the fish density values ​​of each grid cell, gradient color levels are defined to generate a heatmap of the target fish population density at each time point, and the center point of each grid cell is defined.

[0112] The geometric center point of a grid cell is the average of the grid's latitude and longitude range. For example, if the grid's longitude range is 120°E to 120.1°E and its latitude range is 30°N to 30.1°N, then the center point is 120.05°E and 30.05°N.

[0113] The density centroid of the entire monitoring area is calculated using the density value of the grid cells as weights. The formula logic is as follows:

[0114] The centroid longitude is the product of the longitude of all grid center points and the grid density value, divided by the total grid density value.

[0115] The centroid latitude is the product of the latitude of all grid center points and the grid density value, divided by the total grid density value.

[0116] The density centroid of the heatmap at each time point is calculated sequentially to obtain the centroid coordinate sequence, such as... The center of gravity of the heatmap corresponding to the first time point, The center of gravity of the heatmap corresponding to the second time point;

[0117] On a latitude and longitude map, connect the coordinates of the center of gravity in chronological order to form a trajectory line of the center of gravity movement, and calculate the direction angle between two adjacent centers of gravity.

[0118] If the directional consistency of at least three consecutive adjacent centroids is statistically analyzed, and the angular deviation of adjacent directions that is greater than the standard threshold is less than the preset standard angle, then the direction is determined to be the dominant movement direction in a given stage.

[0119] Calculate the straight-line distance between adjacent centroids to obtain the movement distance in each time period. Obtain the movement speed of the target fish group based on the movement distance and time interval. If the movement speed fluctuation is less than the preset speed standard value for at least three consecutive time periods and the dominant direction remains unchanged, the overall movement trend of the group is determined to be stable. If the speed fluctuation exceeds the preset speed standard value or the direction changes abruptly, the original data needs to be backtracked, abnormal grids are excluded, and the calculation is recalculated.

[0120] By combining the dominant direction of movement, average speed, and center of gravity trajectory, the overall movement trend of the group can be obtained.

[0121] By correlating the stratified water temperature and chlorophyll concentration data of each grid cell with the fish distribution data, the matching patterns between fish movement and environmental factors were identified. For example, the migration direction of fish was consistent with the movement direction in the suitable water temperature zone of 18-20℃, and the feeding stage was concentrated when the chlorophyll concentration was greater than or equal to 5 mg / m³. 3 The area;

[0122] Ultimately, by combining the common movement trajectories of the tagged fish, the overall movement trend of the group, and the correlation patterns of environmental factors, the complete migration route of the target fish group was obtained.

[0123] The management decision-making module primarily categorizes target fish populations into four levels—safe, watchful, warning, and emergency—based on the total number of spawning fish and the stability of their migration routes. Corresponding fishing restrictions are then assigned according to each level.

[0124] When the safety level is set, the allowable catch standard is set according to the spawning rate. For example, 40% of the spawning rate is set as the allowable catch standard, and fishing is only prohibited during the spawning period.

[0125] When the target level is raised, the total allowable catch is reduced, for example, to 20% of the spawning capacity, and the no-fishing zone along the migration route is expanded;

[0126] When the alert level is reached, the permitted catch limit is the minimum standard for spawning.

[0127] When the emergency level is reached, all commercial fishing will be banned, and ecological restoration will be carried out simultaneously.

[0128] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in the claims, they should all fall within the protection scope of the present invention.

Claims

1. A dynamic monitoring and management system for marine fishery resources based on multi-data fusion, characterized in that, Includes the following modules: The data acquisition module is used to acquire various data within the monitoring area by using an unmanned vessel carrying multiple sensors in accordance with a preset acquisition mode. The various data include sound wave intensity, frequency signal, signal strength, water temperature at each layer, and chlorophyll concentration. The preprocessing module performs noise reduction and format standardization on the acquired data, and generates corresponding data labels based on the collected data. The data analysis module calculates the total number of spawning fish in the target fish population based on the intensity of the detected sound waves and the volume of the water in the monitored area, and analyzes the migration routes of the target fish population based on the sound wave frequency, water temperature, and chlorophyll concentration. The method for obtaining the total number of spawning fish in the target fish population is as follows: The total scattering intensity of sound waves per unit volume of water is obtained based on historical data, and the average scattering intensity of sound waves per spawning fish in the target fish group is obtained. The initial spawning fish density per unit volume is obtained by subtracting the scattering intensity of a single spawning fish from the total scattering intensity and then performing a logarithmic transformation. By comparing historical data from the same period, the acoustic density of a certain area is calculated and compared with the density of the actual count of fixed-point fishing in the same area during the same period. The basic calibration coefficient is obtained by the ratio of the actual density to the acoustic density. The spawning fish density per unit water volume is obtained by combining the initial spawning fish density with the calibration coefficient, using the formula: in, This refers to the density of spawning fish per unit volume of water. The total scattering intensity of sound waves per unit volume of water. The average scattering intensity of sound waves by a single spawning fish in the target fish group. The reference value for sound wave scattering intensity is set based on the minimum effective scattering intensity in historical data. The acoustic density of a certain area during the same historical period is defined as the initial spawning fish density per unit volume in that area during the same historical period. This represents the actual density of fixed-point fishing in the area during the same historical period. The initial spawning density of the fish population. Basic calibration coefficients; Calculate the total water volume within the monitoring area, multiply the spawning fish density per unit water volume by the total water volume, and obtain the total number of spawning fish in the target fish population. The method for obtaining the migration route of the target fish school is as follows: The frequency signal of the acoustic transmitter of the target fish group is extracted from the data tag. The latitude and longitude coordinates of the signal at different collection time points are recorded to form the movement trajectory of a single marked fish. The overlap of multiple marked fish trajectories is statistically analyzed to filter out common trajectories with the same direction and path, which serve as the basic reference for the migration route. Acquire the acoustic frequency and intensity data of the target fish group, form multiple continuous time series data groups at fixed time intervals, and obtain the target fish density value of each grid cell based on the acoustic intensity data in each time series data group. Based on the fish density value of each grid cell, gradient color levels are divided to generate a heat map of the target fish population density at each time point, and the center point of each grid cell is defined. The geometric center point of the grid cell is taken as the average value of the grid's latitude and longitude range; The density centroid of the entire monitoring area is calculated using the density value of the grid cells as weights. The formula logic is as follows: The centroid longitude is the product of the longitudes of all grid center points and the grid density values, divided by the total grid density values. The centroid latitude is the product of the latitude of all grid center points and the grid density value, divided by the total grid density value. Calculate the density centroid of the heatmap at each time point in sequence to obtain the centroid coordinate sequence; On a latitude and longitude map, connect the coordinates of the center of gravity in chronological order to form a trajectory line of the center of gravity movement, and calculate the direction angle between two adjacent centers of gravity. If the directional consistency of at least three consecutive adjacent centroids is statistically analyzed, and the angular deviation of adjacent directions that is greater than the standard threshold is less than the preset standard angle, then the direction is determined to be the dominant movement direction in a phase. Calculate the straight-line distance between adjacent centroids to obtain the movement distance in each time period. Obtain the movement speed of the target fish group based on the movement distance and time interval. If the movement speed fluctuation is less than the preset speed standard value for at least three consecutive time periods and the dominant direction remains unchanged, the overall movement trend of the group is determined to be stable. If the speed fluctuation exceeds the preset speed standard value or the direction changes abruptly, the original data needs to be backtracked, abnormal grids are excluded, and the calculation is recalculated. By combining the dominant direction of movement, average speed, and center of gravity trajectory, the overall movement trend of the group can be obtained. By correlating the stratified water temperature and chlorophyll concentration data of each grid unit with the fish distribution data, the matching pattern between fish movement and environmental factors is identified. The complete migration route of the target fish group is obtained by comprehensively marking the common movement trajectory of the fish, the overall movement trend of the group, and the correlation pattern of environmental factors. The management decision-making module formulates fishing restriction plans in real time based on the total number of spawning fish and the migration routes of the target fish population.

2. The marine fishery resource dynamic monitoring and management system based on multi-data fusion according to claim 1, characterized in that, The multiple sensors include an acoustic sensor, a temperature sensor, an echo detector, and a chlorophyll fluorometer. The data acquisition module's functions include: A random number of target fish are caught within the monitoring area, acoustic transmitters are implanted into the caught target fish, the target fish with implanted acoustic transmitters are released back into the monitoring area, and an unmanned boat is deployed into the monitoring area. The preset acquisition modes include specified acquisition and random acquisition: The monitoring area is divided into several grid units, each of which is numbered. A data collection sequence is formed according to the numbering order. Based on the data collection sequence, a cruise path is planned for the unmanned surface vessel (USV). When the USV cruises along the planned path, a designated data collection mode is activated. The acoustic wave sensor continuously acquires the acoustic waves emitted by the target fish school's acoustic transmitter, and the echo detector scans within the standard frequency band. The acoustic wave intensity within the grid unit is collected at preset intervals. The water temperature and chlorophyll concentration data of the surface, middle and bottom layers within the grid unit are periodically collected by the temperature sensor and chlorophyll fluorometer. After the unmanned surface vessel completes its cruise along the planned path and collects data from each designated cell, it returns to the starting grid cell and switches to random data collection mode. An automatic data collection cycle is set so that the unmanned surface vessel (USV) drifts randomly with the ocean currents and collects data randomly during the drifting process. When the USV drifts out of the monitoring area, the distance between the USV and the nearest grid cell is calculated, and a return path for the USV is planned so that the USV returns to the detection area.

3. The marine fishery resource dynamic monitoring and management system based on multi-data fusion according to claim 1, characterized in that, The preprocessing module's functions include: The data from multiple data points acquired through specified and random sampling are denoised and the data format is standardized. Acquire latitude and longitude coordinate change data of unmanned surface vessel during specified and random data collection processes. Based on the collection time of multiple data, locate the latitude and longitude coordinates of the unmanned surface vessel when the data is collected, and compare the coordinates of the unmanned surface vessel with the latitude and longitude coordinate range of each grid unit to identify the corresponding grid unit. Generate corresponding data labels based on the specified and randomly collected data for each grid cell; If the unmanned surface vessel (USV) collects random data in non-monitored areas after floating out of the monitoring area, then the grid cells with missing random data are obtained. Based on the distance between the latitude and longitude coordinates of the random data collected by the USV in non-monitored areas and the grid cells with missing random data, the estimated data of the grid cells with missing random data is calculated. The estimated data is used as random data collected from the missing random data grid cells, and the corresponding data labels are generated by combining the specified collected data.

4. The marine fishery resource dynamic monitoring and management system based on multi-data fusion according to claim 3, characterized in that, The valuation data acquisition methods include: Using the center point of the grid cell with missing random data as the reference point, calculate the straight-line distance between the latitude and longitude coordinates of each random data collection point in the non-monitoring area and the reference point; The distance attenuation coefficient is obtained from historical data. The distance attenuation coefficient is used as the exponent of the straight-line distance. The reciprocal of the straight-line distance after being amplified by the distance attenuation coefficient is used as the weighting coefficient of the latitude and longitude coordinates when collecting random data in non-monitoring areas. Based on the water temperature data of non-monitored areas and the corresponding weighting coefficients, obtain the estimated water temperature data of the missing randomly collected data grid cells; The gradient correction coefficient is obtained based on the latitude and longitude coordinates of the chlorophyll concentration data collected in the non-monitoring area. The obtained chlorophyll concentration data of the non-monitoring area is multiplied by the product of the gradient correction coefficient and the straight-line distance, and then multiplied by the corresponding weighting coefficient to obtain the estimated chlorophyll concentration data for the missing random data grid cells. This is then calculated using the formula: in, Estimate chlorophyll concentration. The concentration of chlorophyll in the non-monitoring area. These are gradient correction coefficients. No. The straight-line distance between the latitude and longitude coordinates and the reference point during the collection of valid random data; Based on the natural attenuation of sound waves propagating in water, an attenuation coefficient is set. The sound wave intensity value of the non-monitored area is multiplied by the negative attenuation coefficient of the natural constant and the power of the straight-line distance, and then multiplied by the corresponding weighting coefficient to obtain the estimated sound wave intensity data of the missing random data grid cell.

5. A dynamic monitoring and management system for marine fishery resources based on multi-data fusion as described in claim 1, characterized in that, The overall movement trend of the group was obtained through the following methods: Acquire the acoustic frequency and intensity data of the target fish group, form multiple continuous time series data groups at fixed time intervals, and obtain the target fish density value of each grid cell based on the acoustic intensity data in each time series data group. Based on the fish density values ​​of each grid cell, gradient color levels are divided to generate a heat map of the target fish population density at each time node. The center point of each grid cell is defined, and the density centroid of the entire monitoring area is calculated with the density value of the grid cell as the weight. The density centroid of the heat map at each time node is calculated in sequence to obtain the centroid coordinate sequence. On a latitude and longitude map, connect the coordinates of the center of gravity in chronological order to form a trajectory line of the center of gravity movement, and calculate the direction angle between two adjacent centers of gravity. If the directional consistency of at least three consecutive adjacent centroids is statistically analyzed, and the angular deviation of adjacent directions that is greater than the standard threshold is less than the preset standard angle, then the direction is determined to be the dominant movement direction in a certain stage. Calculate the straight-line distance between adjacent centers of gravity, obtain the movement distance in each time period, and obtain the movement speed of the target fish group based on the movement distance and time interval; if the movement speed fluctuation is less than the preset speed standard value for at least three consecutive time periods, and the dominant direction remains unchanged, then the overall movement trend of the group is determined to be stable. If the speed fluctuation exceeds the preset speed standard value or the direction changes abruptly, the original data needs to be traced back, abnormal grids are excluded, and the calculation is repeated. By combining the dominant direction of movement, average speed, and center of gravity trajectory, the overall movement trend of the group can be obtained.

6. The marine fishery resource dynamic monitoring and management system based on multi-data fusion according to claim 1, characterized in that, The work content of the management decision-making module includes: Based on the total number of spawning fish in the target fish population and the stability of their migration routes, the target fish populations are classified into four levels: safe, attention, warning, and emergency. Corresponding fishing restrictions are then specified for each level. When the safety level is set, the allowable catch standard is set according to the spawning volume, and fishing is prohibited only during the spawning period; When the target level is raised, the total allowable catch is reduced, and the fishing ban area is expanded to include migration routes; When the alert level is reached, the permitted catch limit is the minimum standard for spawning. When the emergency level is reached, all commercial fishing will be banned, and ecological restoration will be carried out simultaneously.

Images