Fishery resource evaluation method and system

Abstract

The invention relates to the technical field of data management, and discloses a fishery resource evaluation method and system, and the method comprises the steps: extracting multi-source fishery data and a ship trajectory set of a target drainage basin, and obtaining an original feature set after processing; decomposing the original feature set into orthogonal waveform components and waveform coefficients according to the spatial distribution of the target watershed; performing linear superposition on the basis vectors of the orthogonal waveform components based on the waveform coefficient to obtain a data coverage density field; sorting the density values in the data coverage density field, and dividing a density value interval of the target drainage basin; according to the density value interval, a field value in the data coverage density field corresponds to a reliability level, and a space reliability distribution map is obtained; mapping ship behavior characteristics in the ship trajectory set to a spatial reliability distribution map to obtain an evaluation sequence of the target drainage basin; the efficiency of multi-source data integration and feature extraction in the fishery resource evaluation process can be improved.

Description

Technical Field

[0001] This invention relates to the field of data management technology, and in particular to a method and system for assessing fishery resources. Background Technology

[0002] Traditional assessment methods often rely on single-type data or sampling information from local areas, lacking systematic integration and in-depth analysis of multi-source data from the target watershed. This fails to comprehensively capture the complex relationships between catch conditions, environmental conditions, and vessel operations, resulting in limited coverage and insufficient correlation of the assessment data. Furthermore, existing technologies, when processing spatially distributed data, fail to effectively establish a precise correspondence between data and geographic space, making it difficult to objectively reflect the true distribution characteristics of fishery resources in different regions. Consequently, the assessment results fail to reflect the spatial differences and overall patterns of fishery resources within the watershed, and cannot provide comprehensive data support for the scientific management of fishery resources.

[0003] Traditional assessment methods lack a scientifically sound quantitative system in the reliability analysis phase, relying heavily on subjective judgment or simple numerical statistics to determine data reliability. This fails to adequately consider key influencing factors such as data distribution characteristics and spatial consistency, resulting in insufficient accuracy and objectivity in reliability assessments. Furthermore, existing technologies lack standardized mapping processes and orderly sorting mechanisms when correlating vessel operation behavior with resource distribution characteristics. This prevents the formation of a clear, prioritized assessment sequence, significantly reducing the practicality of the assessment results and making it difficult to meet the needs of practical applications such as fisheries resource development planning and operational area optimization. This diminishes the guiding value and practicality of the assessment work. Summary of the Invention

[0004] This invention provides a method and system for assessing fishery resources, the main purpose of which is to solve the problem of low integration rate of multi-source data in fishery resource assessment.

[0005] To achieve the above objectives, the present invention provides a method for assessing fishery resources, comprising:

[0006] Extract multi-source fisheries data and vessel trajectory sets from the target watershed to obtain the original feature set of the target watershed;

[0007] Based on the spatial distribution of the target watershed, the original feature set is decomposed into orthogonal waveform components and waveform coefficients of the target watershed;

[0008] The data coverage density field of the target watershed is obtained by linearly superimposing the basis vectors of the orthogonal waveform components based on the waveform coefficients.

[0009] The density values ​​in the data coverage density field are sorted, and the density value interval of the target watershed is divided according to the sorting results.

[0010] Based on the density value range, the field values ​​in the data coverage density field are mapped to reliability levels to obtain a spatial reliability distribution map of the target watershed.

[0011] By mapping the ship behavior characteristics of the ship trajectory set to the spatial reliability distribution map, an evaluation sequence for the target watershed is obtained.

[0012] In a preferred embodiment, the step of extracting multi-source fisheries data and vessel trajectory sets from the target watershed to obtain the original feature set of the target watershed includes:

[0013] Collect fish catch records, environmental monitoring data, and vessel trajectory sets for the target watershed;

[0014] The fish catch record data, the environmental monitoring data, and the vessel trajectory set are spatiotemporally aligned to obtain preprocessed data for the target watershed;

[0015] Multidimensional features are extracted from the preprocessed data to obtain the original feature set of the target watershed.

[0016] In a preferred embodiment, the step of decomposing the original feature set into orthogonal waveform components and waveform coefficients of the target watershed based on the spatial distribution of the target watershed includes:

[0017] The original feature set is reconstructed into a two-dimensional feature matrix of the target watershed in the spatial dimension;

[0018] The wavelet packet transform is performed on the two-dimensional feature matrix to obtain the multi-band wavelet packet coefficients of the target watershed;

[0019] The dominant frequency band coefficients in the wavelet packet coefficients are used as the waveform coefficients of the target watershed, and the time-domain waveforms corresponding to the wavelet packet coefficients are used as the orthogonal waveform components of the target watershed.

[0020] In a preferred embodiment, reconstructing the original feature set into a two-dimensional feature matrix of the target watershed in the spatial dimension includes:

[0021] Establish a spatial index framework for the target watershed based on its latitude and longitude;

[0022] The multidimensional feature vectors in the original feature set are assigned to the grid cells in the spatial indexing frame;

[0023] Spatial interpolation is performed on the grid cells to obtain the two-dimensional feature matrix of the target watershed.

[0024] In a preferred embodiment, the step of linearly superimposing the basis vectors of the orthogonal waveform components based on the waveform coefficients to obtain the data coverage density field of the target watershed includes:

[0025] The effective components of the target watershed are selected based on the amplitude of the waveform coefficients;

[0026] The basis vectors are weighted based on the waveform coefficients corresponding to the effective components to obtain the density distribution surface of the target watershed;

[0027] The density distribution surface is normalized to obtain the data coverage density field of the target watershed.

[0028] In a preferred embodiment, the step of sorting the density values ​​in the data coverage density field and dividing the density value interval of the target watershed according to the sorting result includes:

[0029] The field values ​​of all spatial locations in the data coverage density field are sorted in ascending order to obtain the field value sequence of the target watershed.

[0030] The boundary point of the target watershed is determined based on the quantile position of the field value sequence;

[0031] The data coverage density field is divided using the boundary point as a threshold to obtain the density value range of the target watershed.

[0032] In a preferred embodiment, the step of mapping the field values ​​in the data coverage density field to reliability levels based on the density value interval to obtain the spatial reliability distribution map of the target watershed includes:

[0033] The initial reliability weight of the target watershed is assigned based on the distribution characteristics of the density value range;

[0034] The set of adjacent spatial coordinates of the target watershed is calculated based on the resolution of the target spatial location in the data coverage density field;

[0035] The proportion of coordinates in the spatial coordinate set that belong to the same density value interval as the target spatial location is used as a local consistency index of the target watershed.

[0036] By combining the initial reliability weights with the local consistency index, the overall reliability of the target watershed is obtained;

[0037] The overall reliability is assigned to the spatial location in the data coverage density field to obtain the spatial reliability distribution map of the target watershed.

[0038] In a preferred embodiment, the comprehensive reliability calculation formula is:

[0039]

[0040] in, For the overall reliability, The initial reliability weights, It is a self-recognized constant. The entropy decay coefficient, The information entropy value. This refers to the local consistency index. The consistency contribution coefficient It is the hyperbolic tangent function. The scaling factor. This is the offset parameter.

[0041] In a preferred embodiment, mapping the ship behavior characteristics from the ship trajectory set to the spatial reliability distribution map to obtain the evaluation sequence for the target watershed includes:

[0042] The spatial coordinates of the ship's behavioral characteristics are correlated with the reliability levels of the same geographical locations in the spatial reliability distribution map to obtain the correlation level pairs of the target watershed;

[0043] The association level pairs are sorted in descending order to obtain the evaluation sequence of the target watershed.

[0044] To address the above problems, the present invention also provides a fishery resource assessment system, the system comprising:

[0045] The original feature set module extracts multi-source fisheries data and vessel trajectory sets for the target watershed to obtain the original feature set of the target watershed;

[0046] The component coefficient module decomposes the original feature set into orthogonal waveform components and waveform coefficients of the target watershed based on the spatial distribution of the target watershed;

[0047] The data coverage density field module linearly superimposes the basis vectors of the orthogonal waveform components based on the waveform coefficients to obtain the data coverage density field of the target watershed.

[0048] The density value interval module sorts the density values ​​in the data coverage density field and divides the density value interval of the target watershed according to the sorting results.

[0049] The spatial reliability distribution map module maps the field values ​​in the data coverage density field to reliability levels based on the density value interval, thereby obtaining the spatial reliability distribution map of the target watershed.

[0050] The evaluation sequence module maps the ship behavior characteristics in the ship trajectory set to the spatial reliability distribution map to obtain the evaluation sequence for the target watershed.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. This fisheries resource assessment technology systematically integrates multi-source fisheries data and vessel trajectory sets from the target watershed. Through spatiotemporal alignment and multi-dimensional feature extraction, it forms an original feature set. Then, based on spatial distribution decomposition of orthogonal waveform components and waveform coefficients, it constructs a data coverage density field and divides density value intervals using the waveform coefficients. The entire process revolves around multi-source data fusion and precise spatial mapping, comprehensively capturing the intrinsic correlation between catches, the environment, and vessel operations. This significantly improves the comprehensiveness and relevance of the assessment data, enabling the assessment results to accurately reflect the spatial distribution characteristics and overall patterns of fisheries resources within the watershed. Simultaneously, through a scientific and reliable quantification system, it integrates density value interval distribution characteristics and spatial consistency indicators to calculate comprehensive reliability, generating a spatial reliability distribution map. This makes the assessment of data reliability more objective and accurate, providing solid data support and scientific basis for fisheries resource assessment.

[0053] 2. This technology precisely correlates and sorts ship behavior characteristics with spatial reliability distribution maps in descending order, forming a clear and orderly assessment sequence. It intuitively presents the reliability priorities of fishery resources in different regions, significantly improving the practicality and guidance of the assessment results. The entire assessment process is progressive and logically rigorous, forming a complete closed loop from data collection, processing, and analysis to result output. This ensures the systematic and standardized nature of the assessment process while providing accurate and efficient decision-making references for practical applications such as the scientific development planning of fishery resources and the optimization of operating areas, effectively improving the operability and decision-making efficiency of fishery resource assessment. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating a fishery resource assessment method according to an embodiment of the present invention.

[0055] Figure 2 A functional block diagram of a fishery resource assessment system provided in an embodiment of the present invention;

[0056] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0057] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0058] This application provides a method for assessing fishery resources. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for assessing fishery resources can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0059] Reference Figure 1 The diagram shown is a flowchart illustrating a fishery resource assessment method according to an embodiment of the present invention. In this embodiment, the fishery resource assessment method includes:

[0060] In this embodiment of the invention, the step of extracting multi-source fisheries data and vessel trajectory sets from the target watershed to obtain the original feature set of the target watershed is specifically used for:

[0061] Collect fish catch records, environmental monitoring data, and vessel trajectory sets for the target watershed;

[0062] The fish catch record data, the environmental monitoring data, and the vessel trajectory set are spatiotemporally aligned to obtain preprocessed data for the target watershed;

[0063] Multidimensional features are extracted from the preprocessed data to obtain the original feature set of the target watershed.

[0064] Specifically, the catch recording equipment carried by legally registered fishing vessels within the target watershed is used to collect real-time catch data such as the type of catch, weight of a single type of catch, total catch, fishing time, and latitude and longitude range of the fishing operation for each fishing operation.

[0065] Specifically, using time and space as a unified benchmark, the collected fish catch record data, environmental monitoring data, and vessel trajectory sets are standardized. In the time dimension, a unified time interval is defined in hours, and data recorded by different devices within the same time interval are matched. For data in the fish catch record data where the fishing operation time, environmental monitoring data where the monitoring time, and vessel trajectory concentrated navigation and operation time are not in the same time interval, the data are adjusted to the corresponding interval according to the actual time.

[0066] Specifically, the system extracts multidimensional features from the preprocessed data that can reflect the status of fishery resources and related influencing factors. For the catch record data, it extracts features such as the percentage of single catch weight, the average total catch, and the richness of catch species within a single grid cell per hour.

[0067] Furthermore, by utilizing fixed environmental monitoring stations, mobile monitoring equipment, and satellite remote sensing monitoring systems deployed within the basin, environmental monitoring data such as water temperature, water quality pH value, dissolved oxygen content, plankton density, and water flow velocity are collected. Then, through the global positioning system and automatic identification system equipped on the ships, relevant information such as ship navigation trajectory, operation location, navigation speed, and turning frequency is continuously recorded.

[0068] Furthermore, in terms of spatial dimension, the grids divided by the latitude and longitude of the target watershed are used as the basic spatial units. The latitude and longitude range of fishing in the catch record data, the latitude and longitude coordinates of monitoring stations in the environmental monitoring data, and the latitude and longitude data of navigation and operation in the vessel trajectory set are all mapped to specific grid units. This ensures that the catch record data, environmental monitoring data, and vessel trajectory set in the same grid unit are completely corresponding in time and space, forming target watershed preprocessed data with a unified structure and spatiotemporal matching.

[0069] Furthermore, for environmental monitoring data, features such as hourly average water temperature, stable pH value, peak and trough dissolved oxygen content, and average plankton density are extracted within a single grid cell. For the ship trajectory set, features such as hourly ship dwell time, number of ships, frequency of ship operations, and average sailing speed are extracted within a single grid cell. All extracted features directly correspond to specific information in the preprocessed data without adding any extra irrelevant data, ultimately forming a multi-dimensional and comprehensive original feature set of the target watershed that includes features related to fish catch, environment, and ship behavior.

[0070] In summary, collecting catch records, environmental monitoring data, and vessel trajectory sets from the target watershed can comprehensively gather core basic data reflecting the status of fishery resources in the target watershed. Catch records directly reflect the actual output of fishery resources, environmental monitoring data reflects the external conditions affecting the survival of fishery resources, and vessel trajectory sets reflect the spatial distribution and activity patterns of fishery operations. These three data points fully cover the key information dimensions required for fishery resource assessment, avoiding biased assessments due to data gaps. This lays a solid data foundation for subsequent preprocessing data generation and original feature set extraction, ensuring that the assessment work has comprehensive and authentic data source support.

[0071] In summary, spatiotemporal alignment of fish catch records, environmental monitoring data, and vessel trajectory sets yields preprocessed data. This process eliminates discrepancies between different data types in terms of time and space, transforming previously independently collected multi-source data into a unified spatiotemporal whole. Fish catch, environmental, and vessel-related data from the same time interval and spatial location are accurately correlated, preventing data association failures due to spatiotemporal misalignment. This ensures consistency and relevance of the preprocessed data, providing a well-structured and logically coherent data foundation for subsequent multidimensional feature extraction. It also guarantees that the feature extraction results accurately reflect the inherent relationships between the data.

[0072] In summary, extracting multidimensional features from preprocessed data to obtain the original feature set enables the mining of key information reflecting the core attributes of fishery resources from the integrated preprocessed data. This covers multiple dimensions, including catch-related yield and species characteristics, environmental characteristics such as water quality and temperature, and vessel-related operating location and frequency characteristics. It transforms scattered raw data into feature vectors with clear evaluation significance, forming a structured original feature set that can be used for subsequent decomposition operations. This provides direct data support for decomposing orthogonal waveform components and waveform coefficients based on spatial distribution, and promotes the evaluation process towards greater precision and systematization.

[0073] In this embodiment of the invention, when the original feature set is decomposed into orthogonal waveform components and waveform coefficients of the target watershed based on the spatial distribution of the target watershed, it is specifically used for:

[0074] The original feature set is reconstructed into a two-dimensional feature matrix of the target watershed in the spatial dimension;

[0075] The wavelet packet transform is performed on the two-dimensional feature matrix to obtain the multi-band wavelet packet coefficients of the target watershed;

[0076] The dominant frequency band coefficients in the wavelet packet coefficients are used as the waveform coefficients of the target watershed, and the time-domain waveforms corresponding to the wavelet packet coefficients are used as the orthogonal waveform components of the target watershed.

[0077] Specifically, the area is divided into regular rectangular grids at fixed geographical intervals. Each grid is an independent spatial unit and is given a unique identifier. At the same time, the latitude and longitude coordinates of the four corners of each grid unit are recorded to establish a spatial index framework covering the entire target watershed. Then, each multidimensional feature vector in the original feature set is extracted one by one. Based on the latitude and longitude information corresponding to the feature vector, the grid unit in the spatial index framework to which it belongs is determined, and the multidimensional feature vector is completely assigned to the corresponding grid unit.

[0078] Specifically, the constructed two-dimensional feature matrix is ​​used as the processing object, and the matrix is ​​decomposed layer by layer according to the preset number of decomposition layers. During the decomposition process, each layer will split the feature information of the previous layer into sub-band information of different frequency ranges. Each sub-band information corresponds to a set of feature data. These sub-band information cover all feature components from low frequency to high frequency in the original two-dimensional feature matrix.

[0079] Specifically, the obtained multi-band wavelet packet coefficients are sorted by energy value, and the coefficient corresponding to the band with the largest energy value is selected. This coefficient is then determined as the waveform coefficient of the target watershed, which can reflect the most important feature change patterns in the original feature set.

[0080] Furthermore, for some blank grid cells that have not been assigned multidimensional feature vectors, spatial interpolation is performed by filling the mean of the multidimensional feature vectors in adjacent grid cells, so that all grid cells have corresponding feature data, and finally a two-dimensional feature matrix of the target watershed is formed with grid cells as rows and multidimensional features as columns.

[0081] Furthermore, after the decomposition is completed, for each sub-band information, the energy value of the feature data within that sub-band is calculated to characterize the corresponding coefficient size. The energy values ​​corresponding to all sub-bands together constitute the multi-band wavelet packet coefficients of the target watershed. Each multi-band wavelet packet coefficient corresponds to a specific frequency range and feature components.

[0082] Furthermore, for each multi-band wavelet packet coefficient, we trace back its characteristic change pattern in the time dimension during the decomposition process. This characteristic change pattern in the time dimension is the time-domain waveform. Since there are no overlapping frequency components between the time-domain waveforms of different frequency bands, and they can complement each other to completely represent the original characteristic information, these time-domain waveforms are determined as the orthogonal waveform components of the target watershed.

[0083] In summary, reconstructing the original feature set into a two-dimensional feature matrix of the target watershed in the spatial dimension can organize the scattered multidimensional feature vectors according to the spatial distribution of the target watershed. Through operations such as establishing a spatial indexing framework, allocating feature vectors, and spatial interpolation, the feature data that was originally spatially unrelated can be transformed into a matrix form with clear rows and columns and clear spatial positioning. This allows the feature data to be accurately bound to the geographic space of the target watershed, eliminates the chaotic state of the data in the spatial dimension, and provides a data foundation with a well-structured structure and clear spatial information for subsequent wavelet packet transformation, ensuring that the subsequent processing can be carried out in an orderly manner based on the spatial dimension.

[0084] In summary, wavelet packet transform of a two-dimensional feature matrix yields multi-band wavelet packet coefficients, enabling in-depth mining of hidden feature information across different frequency ranges within the matrix. By decomposing the original feature data layer by layer into multiple sub-bands covering low to high frequencies, the coefficients corresponding to each sub-band fully preserve the feature details within that frequency range. This achieves multi-scale, comprehensive analysis of the original feature data, avoiding feature omissions caused by single-scale analysis. It provides rich and accurate frequency feature support for subsequent selection of dominant frequency band coefficients, determination of waveform coefficients, and orthogonal waveform components.

[0085] In summary, by using the dominant frequency band coefficients in the wavelet packet coefficients as waveform coefficients and the corresponding time-domain waveforms as orthogonal waveform components, the most core and representative feature components in the original feature set can be accurately extracted. The dominant frequency band coefficients, with their maximum energy values, can reflect the main variation patterns of the original feature data, while the corresponding time-domain waveforms have the characteristics of being independent of each other and able to completely represent the features. The clear division between the two makes the decomposition results of the original feature set more targeted and practical, eliminating irrelevant secondary feature interference while retaining key feature information, laying a precise and efficient foundation for the subsequent construction of a data coverage density field based on waveform coefficients.

[0086] In this embodiment of the invention, when the original feature set is reconstructed into a two-dimensional feature matrix of the target watershed in the spatial dimension, it is specifically used for:

[0087] Establish a spatial index framework for the target watershed based on its latitude and longitude;

[0088] The multidimensional feature vectors in the original feature set are assigned to the grid cells in the spatial indexing frame;

[0089] Spatial interpolation is performed on the grid cells to obtain the two-dimensional feature matrix of the target watershed.

[0090] Specifically, the latitude and longitude coordinates corresponding to the easternmost, westernmost, southernmost, and northernmost points of the watershed are determined, and the latitude and longitude range of the entire target watershed is divided into several rectangular grids of the same size at fixed geographical intervals, with each rectangular grid being an independent grid unit.

[0091] Specifically, each multidimensional feature vector in the original feature set is extracted one by one, and the latitude and longitude information attached to each multidimensional feature vector is examined. This latitude and longitude information accurately corresponds to the specific spatial location within the target watershed. Based on the latitude and longitude location, the latitude and longitude range of each grid cell in the spatial index frame is compared to determine the unique grid cell to which the multidimensional feature vector belongs.

[0092] Specifically, for blank grid cells in the spatial indexing frame that have not been assigned multidimensional feature vectors, spatial interpolation is used for spatial interpolation. First, the eight non-blank grid cells surrounding each blank grid cell are determined, and the feature values ​​of the multidimensional feature vectors stored in these eight adjacent grid cells are extracted.

[0093] Furthermore, a unique identification number is assigned to each grid cell, and the specific latitude and longitude coordinates of the four vertices of each grid cell are recorded to ensure that the spatial range of each grid cell is clear and non-overlapping. All grid cells completely cover the entire spatial area of ​​the target watershed without any omissions or exceeding the watershed range, thereby establishing a spatial index framework for the target watershed with a regular structure and clear spatial definition.

[0094] Furthermore, the multidimensional feature vector is completely stored in the feature data storage area of ​​the corresponding grid cell, ensuring that each multidimensional feature vector can be accurately assigned to the grid cell corresponding to its spatial location. This completes the precise matching and allocation of all multidimensional feature vectors with grid cells in the spatial index frame, so that each grid cell carries the corresponding multidimensional feature vector data within its covered spatial range.

[0095] Furthermore, the arithmetic mean of each feature value is calculated, and this arithmetic mean is used as the value of the corresponding feature in the blank grid cell. This process is repeated to fill in all feature values ​​in the blank grid cells, so that each grid cell in the spatial index frame has complete multidimensional feature data. Then, the multidimensional feature data of all grid cells are arranged in the order of rows and columns, using the grid cell identification number as the row index and the type of multidimensional feature as the column index, to form a two-dimensional feature matrix of the target watershed with clear rows and columns and complete data.

[0096] In summary, establishing a spatial indexing framework based on the latitude and longitude of the target watershed enables the construction of a well-organized spatial system covering the entire target watershed using geographic coordinates as a benchmark. By clearly defining grid units and assigning them unique identifiers, scattered feature data are given a unified spatial positioning benchmark, ensuring that each spatial location can find its corresponding affiliation within the framework. This eliminates the problem of data chaos in spatial dimensions and provides a clear and orderly spatial carrier for the subsequent allocation of multi-dimensional feature vectors, making the association between feature data and geographic space more accurate and laying a solid spatial foundation for the entire evaluation process.

[0097] In summary, assigning multidimensional feature vectors from the original feature set to grid cells in the spatial indexing framework enables precise binding of feature data to spatial locations. Each multidimensional feature vector is assigned to a corresponding grid cell based on its associated latitude and longitude information, avoiding mismatches between feature data and spatial locations. This makes the grid cell the smallest unit carrying all feature information within a specific spatial range, transforming the originally scattered multidimensional feature vectors into a dataset arranged in a regular spatial distribution. This provides clear structural and spatially correlated data support for subsequent spatial interpolation and the construction of two-dimensional feature matrices.

[0098] In summary, spatial interpolation of grid cells yields a two-dimensional feature matrix of the target watershed, filling the feature data gaps in blank grid cells. By employing the mean value of feature values ​​from adjacent non-blank grid cells, it ensures that all grid cells possess complete multidimensional feature data, avoiding deviations in evaluation results due to missing spatial location data. Furthermore, arranging the feature data of all grid cells in row and column order forms a structurally unified and data-complete two-dimensional feature matrix. This provides core data with standardized format and direct computation for subsequent wavelet packet transform and other processing steps, ensuring the smooth progress of the evaluation process and the accuracy of the evaluation results.

[0099] In this embodiment of the invention, when linearly superimposing the basis vectors of the orthogonal waveform components based on the waveform coefficients to obtain the data coverage density field of the target watershed, the specific method is as follows:

[0100] The effective components of the target watershed are selected based on the amplitude of the waveform coefficients;

[0101] The basis vectors are weighted based on the waveform coefficients corresponding to the effective components to obtain the density distribution surface of the target watershed;

[0102] The density distribution surface is normalized to obtain the data coverage density field of the target watershed.

[0103] Specifically, all waveform coefficients are sorted out, and the amplitude corresponding to each waveform coefficient is extracted one by one. All amplitudes are integrated into a complete amplitude set. By statistically analyzing the overall distribution of this set, a fixed screening criterion is determined.

[0104] Specifically, all waveform coefficients corresponding to the effective components obtained from the screening are extracted, and the basis vectors corresponding to the effective components in the orthogonal waveform components are extracted. The waveform coefficients of each effective component are used as the weight values ​​of the basis vectors corresponding to that component.

[0105] Specifically, first determine the numerical range of all spatial points in the density distribution surface, find the maximum and minimum values, and then subtract the minimum value from the value of each spatial point in the surface using the maximum and minimum values ​​as references to obtain the difference for each point. Then divide this difference by the difference between the maximum and minimum values. Through this series of fixed calculation operations, the values ​​of all spatial points in the density distribution surface are uniformly converted to a fixed numerical range.

[0106] Furthermore, the standard uses the middle level of the amplitude set as a benchmark, and explicitly stipulates that waveform coefficients with amplitudes higher than this benchmark are qualified coefficients. Subsequently, the orthogonal waveform components corresponding to these qualified waveform coefficients are defined as the effective components of the target watershed.

[0107] Furthermore, following the order of the dimensions of the basis vectors, the values ​​of each dimension of each basis vector are multiplied by the corresponding waveform coefficients to obtain the weighted dimension values ​​of each basis vector. Then, all the weighted basis vectors are superimposed point by point according to the correspondence of spatial positions, that is, the weighted basis vector values ​​corresponding to all effective components at each spatial position are summed to form a surface that can reflect the spatial characteristic distribution of the target watershed. This surface is the density distribution surface of the target watershed.

[0108] Furthermore, the converted values ​​retain the relative relationships of the original density distribution and have a unified comparability standard. The complete spatial density distribution formed after this processing is the data coverage density field of the target watershed. This density field can clearly and systematically present the data coverage of different spatial locations within the target watershed.

[0109] In summary, by selecting effective components based on the amplitude of waveform coefficients, core components with significant features in the original feature set can be accurately identified. By clarifying the amplitude selection criteria, invalid components with low amplitudes and weak impact on fishery resource assessment can be eliminated, avoiding secondary information from interfering with subsequent calculations. This ensures that subsequent weighted calculations focus only on key components that truly reflect the characteristics of the target watershed data, ensuring that the construction of the density distribution surface is based on core effective information and improving the accuracy and relevance of the assessment results.

[0110] In summary, the density distribution surface obtained by weighting the basis vectors based on the waveform coefficients corresponding to the effective components can organically combine the characteristic intensity of the waveform coefficients with the spatial distribution characteristics of the basis vectors. The waveform coefficients, as weights, directly reflect the importance of the corresponding effective components, enabling the weighted basis vectors to highlight the influence of core features during the superposition process. The density distribution surface formed by superposition fully presents the spatial variation law of data coverage within the target watershed, clearly reflects the density differences in different regions, and provides an intuitive and comprehensive spatial density basis for subsequent normalization processing and density value interval division.

[0111] In summary, normalizing the density distribution surface to obtain the data-covered density field can uniformly transform the values ​​of different ranges in the density distribution surface into a fixed interval, eliminating the magnitude differences in density values ​​at different spatial locations. This provides a unified and comparable standard for density data across the entire target watershed, avoiding deviations in density value interval division caused by inconsistent numerical ranges. At the same time, the normalized density data retains the original spatial distribution relationships, resulting in a standardized and unified data-covered density field structure. This provides standardized data support for subsequent determination of reliability levels based on density values ​​and generation of spatial reliability distribution maps, ensuring the consistency and accuracy of the entire assessment process.

[0112] In this embodiment of the invention, the step of sorting the density values ​​in the data coverage density field and dividing the density value interval of the target watershed according to the sorting result is specifically used for:

[0113] The field values ​​of all spatial locations in the data coverage density field are sorted in ascending order to obtain the field value sequence of the target watershed.

[0114] The boundary point of the target watershed is determined based on the quantile position of the field value sequence;

[0115] The data coverage density field is divided using the boundary point as a threshold to obtain the density value range of the target watershed.

[0116] Specifically, the data coverage density field is traversed to all spatial locations, and the field value corresponding to each spatial location is extracted one by one. All these field values ​​are collected and organized to form a complete set of field values.

[0117] Specifically, the total length of the field value sequence is determined, that is, the total number of all field values. The number of dividing points required is determined according to the preset number of intervals. The specific position of each dividing point in the sequence is calculated according to the principle of equal division. For example, if three density value intervals need to be divided, two dividing points are determined. The first dividing point is located at one-quarter of the total length of the sequence, and the second dividing point is located at three-quarters of the total length of the sequence.

[0118] Specifically, all the determined dividing points are used as dividing thresholds. These thresholds are arranged in order of numerical size, with the smallest dividing point as the lower limit and the largest dividing point as the upper limit. Combined with the overall range of the field value of the data coverage density field, continuous and non-overlapping numerical intervals are divided.

[0119] Furthermore, the set of field values ​​is arranged sequentially in ascending order of numerical value, while maintaining the correspondence between each field value and its original spatial location during the arrangement process, ultimately forming an ordered sequence of field values ​​for the target watershed.

[0120] Furthermore, by counting, we start from the beginning of the field value sequence to find the field value at the corresponding position. These field values ​​are directly determined as the boundary points of the target watershed. Each boundary point corresponds to a specific value in the field value sequence, ensuring that the selection of the boundary points is accurate and has a clear basis for division.

[0121] Furthermore, field values ​​less than the minimum dividing point are assigned to the first density value interval, field values ​​between two adjacent dividing points are assigned to the corresponding middle density value interval, and field values ​​greater than the maximum dividing point are assigned to the last density value interval. Subsequently, the field values ​​at each spatial location in the data-covered density field are assigned to their corresponding numerical intervals, ultimately forming multiple density value intervals.

[0122] In summary, arranging the field values ​​at all spatial locations within the density field in ascending order yields the field value sequence for the target watershed. This process organizes the density field values ​​scattered across different spatial locations into an ordered set, clearly presenting the magnitude distribution patterns of all field values. This avoids subsequent classification biases caused by chaotic field values, ensuring that each field value can find a clear location within the sequence. It provides an ordered and complete numerical basis for determining the boundary points, ensuring that the selection of boundary points can comprehensively cover the distribution range of all field values ​​and guaranteeing the comprehensiveness of the density value interval division.

[0123] In summary, determining the boundary point of the target watershed based on the quantile position of the field value sequence allows for scientific division based on the natural distribution characteristics of the field values. The quantile position directly corresponds to a specific value in the field value sequence, ensuring that the boundary point can accurately divide different distribution intervals of the field values. This avoids the problem of unreasonable division caused by subjectively setting thresholds, making the boundary point conform to the distribution law of the data itself and accurately distinguish field values ​​of different density levels, thus providing an objective and accurate basis for the division of density value intervals.

[0124] In summary, dividing the data coverage density field into density value intervals for the target watershed using the boundary point as a threshold allows all field values ​​in the data coverage density field to be classified into clearly defined intervals. The boundaries of each interval are strictly defined by the boundary point, ensuring that the intervals do not overlap or omit any values. This enables density field values ​​at different spatial locations to find their corresponding intervals, forming a density value interval system with a clear structure and well-defined boundaries. This provides clearly categorized basic data for subsequent initial reliability weight allocation and comprehensive reliability calculation, promoting the assessment process towards greater precision and systematization.

[0125] In this embodiment of the invention, when the spatial reliability distribution map of the target watershed is obtained by corresponding the field values ​​in the data coverage density field to the reliability level according to the density value interval, it is specifically used for:

[0126] The initial reliability weight of the target watershed is assigned based on the distribution characteristics of the density value range;

[0127] The set of adjacent spatial coordinates of the target watershed is calculated based on the resolution of the target spatial location in the data coverage density field;

[0128] The proportion of coordinates in the spatial coordinate set that belong to the same density value interval as the target spatial location is used as a local consistency index of the target watershed.

[0129] By combining the initial reliability weights with the local consistency index, the overall reliability of the target watershed is obtained;

[0130] The overall reliability is assigned to the spatial location in the data coverage density field to obtain the spatial reliability distribution map of the target watershed.

[0131] Specifically, the distribution characteristics of all density value intervals are analyzed, with a focus on the number of spatial locations contained in each density value interval, the degree of concentration of field values ​​within the interval, and the spatial distribution range of the interval in the data coverage density field. Fixed weight allocation rules are set based on these characteristics.

[0132] Specifically, the resolution of the target spatial location in the data coverage density field is defined. This resolution represents the side length of the grid cell corresponding to the target spatial location. The grid extends in four directions: up, down, left, and right, with the coordinates of the target spatial location as the center and the length of the resolution as the step size.

[0133] Specifically, each coordinate in the adjacent spatial coordinate set is extracted one by one, the density value interval to which each coordinate belongs is determined, and it is compared with the density value interval to which the target spatial location belongs. The number of coordinates in the coordinate set that belong to the same density value interval as the target spatial location is counted.

[0134] Specifically, the initial reliability weight and local consistency index corresponding to the target spatial location are first obtained. The values ​​of the initial reliability weight and the local consistency index are then combined and calculated. The values ​​of the initial reliability weight are first standardized so that they are incorporated into the calculation process at a fixed ratio.

[0135] Specifically, the comprehensive reliability value corresponding to each target spatial location is directly assigned to that target spatial location in the data coverage density field, ensuring that each spatial location has a unique corresponding comprehensive reliability value.

[0136] Furthermore, the more spatial locations a density value interval contains, the more concentrated the field values ​​are, and the wider the spatial distribution, the larger the assigned weight value. According to this rule, each density value interval is assigned a corresponding fixed value as the initial reliability weight of the target watershed, ensuring that each density value interval has a clear and unique initial reliability weight, and that the weight value can accurately reflect the distribution characteristics of the interval.

[0137] Furthermore, the coordinates of the eight grid cells adjacent to the target spatial location are determined, including the coordinates of the target spatial location itself. These coordinates are collected and organized into a complete set, which is the set of adjacent spatial coordinates of the target watershed.

[0138] Furthermore, the ratio of this number to the total number of coordinates in the adjacent spatial coordinate set is calculated. This ratio is the local consistency index of the target watershed. The value of the local consistency index directly reflects the degree of consistency between the target spatial location and its surrounding spatial locations in terms of density value range.

[0139] Furthermore, the values ​​of the local consistency index are adjusted by the same proportion, and then the two adjusted values ​​are added together to obtain a combined value that comprehensively reflects the initial reliability weight and the local consistency index. This combined value is the comprehensive reliability of the target watershed.

[0140] Furthermore, after the assignment is completed, based on the spatial framework of the data coverage density field, the comprehensive reliability values ​​of all spatial locations are presented in a visual manner. Different colors or gray levels are used to distinguish the magnitude of different comprehensive reliability values, forming a spatial reliability distribution map of the target watershed that clearly shows the reliability distribution of each spatial location within the target watershed.

[0141] In summary, allocating initial reliability weights based on the distribution characteristics of density value intervals allows for a precise match between the initial reliability weights and the actual situation of the density value intervals. The distribution characteristics of the density value intervals, such as the number of spatial locations, the degree of concentration of field values, and the spatial distribution range, directly determine the weight magnitude. This provides clear and realistic basis for the initial reliability weights, avoids weight distortion caused by subjective and arbitrary allocation, and provides basic parameters that fit the actual data characteristics of the target watershed for subsequent comprehensive reliability calculations. This ensures that the starting point for comprehensive reliability calculations is scientific and reasonable.

[0142] In summary, calculating the neighboring spatial coordinate set based on the resolution of the target spatial location in the data coverage density field can accurately pinpoint the relevant spatial coordinates around the target spatial location. The resolution clarifies the side length of the grid cell, and the neighboring spatial coordinate set obtained based on this can completely cover the surrounding associated area of ​​the target spatial location. This ensures that the collected neighboring spatial coordinates have a direct spatial correlation with the target spatial location, providing comprehensive and accurate coordinate data support for the subsequent calculation of local consistency indicators and avoiding calculation deviations caused by improper selection of surrounding coordinates.

[0143] In summary, using the proportion of coordinates in the spatial coordinate set that belong to the same density value interval as the target spatial location as a local consistency index can intuitively reflect the degree of consistency between the target spatial location and the surrounding spatial locations in terms of density distribution. The larger the proportion, the more the density characteristics of the target spatial location match the surroundings. The index value can quantify this spatial consistency, injecting spatial correlation dimension reference factors into the calculation of comprehensive reliability, so that comprehensive reliability not only reflects the characteristics of a single interval, but also reflects the coordination of spatial distribution.

[0144] In summary, the integrated reliability obtained by combining the initial reliability weights and the local consistency index can integrate information from both the inherent characteristics of the density value interval and the spatial distribution consistency. It retains the basic reliability attributes of the density value interval reflected by the initial reliability weights and incorporates the spatial correlation reliability attributes reflected by the local consistency index. This gives the integrated reliability a more comprehensive set of dimensions. Compared with a single-dimensional reliability assessment, the results are more accurate and comprehensive, and can more realistically reflect the data reliability of the target spatial location.

[0145] In summary, assigning comprehensive reliability values ​​to spatial locations within the data coverage density field yields a spatial reliability distribution map. This transforms abstract comprehensive reliability values ​​into intuitive spatial distribution visualizations, with each spatial location corresponding to a comprehensive reliability value. The distribution map clearly presents the reliability differences between different regions within the target watershed, enabling fisheries resource assessment personnel to quickly locate areas with high reliability. This provides an intuitive and easy-to-understand spatial reference for subsequent vessel behavior characteristic mapping and assessment sequence generation, thereby improving the efficiency and relevance of fisheries resource assessment work.

[0146] In this embodiment of the invention, the formula for calculating the overall reliability is specifically used for:

[0147]

[0148] in, For the overall reliability, The initial reliability weights, It is a self-recognized constant. The entropy decay coefficient, The information entropy value. This refers to the local consistency index. The consistency contribution coefficient It is the hyperbolic tangent function. The scaling factor. This is the offset parameter.

[0149] Specifically, the initial reliability weight is derived from the distribution characteristics of density value intervals. This is achieved by statistically analyzing the number of spatial locations within each density value interval, the concentration of field values ​​within the interval, and the spatial distribution range of the interval within the data coverage density field. A fixed value is assigned to each density value interval according to fixed rules; this value is the initial reliability weight. The information entropy value is obtained by calculating the dispersion of field values ​​within each density value interval. First, the frequency of occurrence of all field values ​​within the interval is statistically analyzed. Then, a value reflecting the degree of disorder in the field value distribution is calculated based on the frequency; this value is the information entropy value. The local consistency index is derived from the analysis of the neighboring spatial coordinate set. The number of coordinates in the neighboring spatial coordinate set that belong to the same density value interval as the target spatial location is statistically analyzed. The ratio of this number to the total number of coordinates in the neighboring spatial coordinate set is calculated; this ratio is the local consistency index. Natural constants are fixed mathematical constants. The entropy decay coefficient, consistency contribution coefficient, scale coefficient, and offset parameter are all fixed values ​​determined based on the actual needs of fishery resource assessment in the target watershed and historical assessment data. These values ​​are used to adjust the influence of each part in the formula.

[0150] Furthermore, the significance of this formula lies in integrating the initial reliability weights and the local consistency index, while also incorporating the correction effect of the information entropy value on the initial reliability weights and the nonlinear adjustment of the local consistency index by the hyperbolic tangent function. This allows for the calculation of a comprehensive reliability that fully reflects the reliability of the target's spatial location. The comprehensive reliability not only reflects the basic reliability brought about by the distribution characteristics of the density value interval itself, but also considers the impact of the consistency between the target's spatial location and the surrounding spatial locations in terms of density value interval assignment on reliability. It also corrects the rationality of the initial reliability weights through the information entropy value, ultimately obtaining a value that accurately characterizes the reliability of the target's spatial location.

[0151] In summary, as the initial reliability weight increases, assuming the information entropy remains constant, the value of the part of the initial reliability weight related to the natural constant in the formula increases, and the overall reliability increases accordingly. Conversely, as the information entropy increases, the value of the part of the initial reliability weight related to the natural constant decreases, and the overall reliability decreases accordingly. As the local consistency index increases, the value adjusted by the hyperbolic tangent function increases, and the overall reliability increases accordingly. When the entropy decay coefficient increases, the weakening effect of the information entropy on the initial reliability weight strengthens, and the overall reliability decreases. When the consistency contribution coefficient increases, the contribution of the local consistency index to the overall reliability strengthens, and the overall reliability increases. When the scaling coefficient increases, the influence of the local consistency index on the hyperbolic tangent function result strengthens, and when the local consistency index exceeds a certain value, the increase in overall reliability becomes more significant. When the offset parameter increases, the result of the hyperbolic tangent function shifts overall, and assuming the local consistency index remains constant, the overall reliability changes accordingly, with the specific trend consistent with the adjustment direction of the offset parameter.

[0152] In this embodiment of the invention, the step of mapping the ship behavior characteristics of the ship trajectory set to the spatial reliability distribution map to obtain the evaluation sequence of the target watershed is specifically used for:

[0153] The spatial coordinates of the ship's behavioral characteristics are correlated with the reliability levels of the same geographical locations in the spatial reliability distribution map to obtain the correlation level pairs of the target watershed;

[0154] The association level pairs are sorted in descending order to obtain the evaluation sequence of the target watershed.

[0155] Specifically, each ship behavior feature in the ship trajectory set is extracted one by one, and the specific spatial coordinates corresponding to each feature are determined. These coordinates accurately correspond to geographical locations within the target watershed. At the same time, the spatial reliability distribution map of the target watershed is opened. This map clearly marks the reliability level of all geographical locations within the watershed. The spatial coordinates of the extracted ship behavior features are matched one by one with the geographical locations in the spatial reliability distribution map.

[0156] Specifically, all generated association level pairs are collected, the specific value of the reliability level in each association level pair is determined, and all association level pairs are arranged in descending order based on the size of the reliability level value.

[0157] Furthermore, the geographical location that completely overlaps with the spatial coordinates on the distribution map is found, the reliability level corresponding to the geographical location is read, and the ship behavior characteristics and the matched reliability level are combined into a set of correspondences. Each set of correspondences contains complete ship behavior characteristic information and a unique reliability level. All the combined correspondences together constitute the association level pair of the target watershed.

[0158] Furthermore, during the arrangement process, the correspondence between the internal ship behavior characteristics and reliability levels of each association level remains unchanged. For association level pairs with the same reliability level value, a secondary sorting is performed according to the order of longitude from largest to smallest in the target watershed, based on the spatial coordinates corresponding to the ship behavior characteristics. If the longitudes are also the same, the sorting is performed according to the order of latitude from largest to smallest, ensuring that all association level pairs can obtain a unique position after sorting. The ordered set of association level pairs formed after sorting is the assessment sequence of the target watershed. This sequence clearly presents the order of reliability levels corresponding to different ship behavior characteristics, providing a direct reference for fishery resource assessment.

[0159] In summary, by linking the spatial coordinates of vessel behavior characteristics with the reliability levels of the same geographical location on the spatial reliability distribution map to form correlation level pairs, a precise correspondence between vessel operation behavior and watershed spatial reliability can be established. This allows for a direct link between vessel operations at different locations within the target watershed and the reliability of fishery resource data at those locations, preventing a disconnect between vessel behavior characteristics and spatial reliability information. It ensures that each vessel behavior characteristic can find corresponding reliability support, providing accurate matching data for subsequent assessments. Furthermore, this correlation process strictly relies on geographical consistency, eliminating irrelevant interference factors and improving the accuracy and relevance of data correlation. This ensures that the correlation level pairs truly reflect the intrinsic connection between vessel operations and spatial reliability.

[0160] In summary, sorting the correlation levels in descending order to obtain the assessment sequence clearly presents the reliability priority of different vessel operation locations within the target watershed. This allows fisheries resource assessors to quickly identify the behavioral characteristics and corresponding spatial locations of vessels with high reliability levels, and to clarify areas with better fisheries resource data quality. This provides an intuitive and orderly reference for the rational development of fisheries resources and the planning of operational areas. At the same time, the sorted assessment sequence integrates the core information of vessel behavior and spatial reliability, simplifies the assessment decision-making process, avoids the problem of low assessment efficiency caused by disorganized data, improves the systematicness and practicality of fisheries resource assessment, and helps to form scientific and efficient assessment results.

[0161] Compared with the prior art, the present invention has the following beneficial effects:

[0162] 1. This fisheries resource assessment technology systematically integrates multi-source fisheries data and vessel trajectory sets from the target watershed. Through spatiotemporal alignment and multi-dimensional feature extraction, it forms an original feature set. Then, based on spatial distribution decomposition of orthogonal waveform components and waveform coefficients, it constructs a data coverage density field and divides density value intervals using the waveform coefficients. The entire process revolves around multi-source data fusion and precise spatial mapping, comprehensively capturing the intrinsic correlation between catches, the environment, and vessel operations. This significantly improves the comprehensiveness and relevance of the assessment data, enabling the assessment results to accurately reflect the spatial distribution characteristics and overall patterns of fisheries resources within the watershed. Simultaneously, through a scientific and reliable quantification system, it integrates density value interval distribution characteristics and spatial consistency indicators to calculate comprehensive reliability, generating a spatial reliability distribution map. This makes the assessment of data reliability more objective and accurate, providing solid data support and scientific basis for fisheries resource assessment.

[0163] 2. This technology precisely correlates and sorts ship behavior characteristics with spatial reliability distribution maps in descending order, forming a clear and orderly assessment sequence. It intuitively presents the reliability priorities of fishery resources in different regions, significantly improving the practicality and guidance of the assessment results. The entire assessment process is progressive and logically rigorous, forming a complete closed loop from data collection, processing, and analysis to result output. This ensures the systematic and standardized nature of the assessment process while providing accurate and efficient decision-making references for practical applications such as the scientific development planning of fishery resources and the optimization of operating areas, effectively improving the operability and decision-making efficiency of fishery resource assessment.

[0164] like Figure 2 The diagram shown is a functional block diagram of a fishery resource assessment system provided in an embodiment of the present invention.

[0165] The fishery resource assessment system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the fishery resource assessment system 100 may include an original feature set module 101, a component coefficient module 102, a data coverage density field module 103, a density value interval module 104, a spatial reliability distribution map module 105, and an assessment sequence module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0166] In this embodiment, the functions of each module / unit are as follows:

[0167] The original feature set module extracts multi-source fisheries data and vessel trajectory sets for the target watershed to obtain the original feature set of the target watershed;

[0168] The component coefficient module decomposes the original feature set into orthogonal waveform components and waveform coefficients of the target watershed based on the spatial distribution of the target watershed;

[0169] The data coverage density field module linearly superimposes the basis vectors of the orthogonal waveform components based on the waveform coefficients to obtain the data coverage density field of the target watershed.

[0170] The density value interval module sorts the density values ​​in the data coverage density field and divides the density value interval of the target watershed according to the sorting results.

[0171] The spatial reliability distribution map module maps the field values ​​in the data coverage density field to reliability levels based on the density value interval, thereby obtaining the spatial reliability distribution map of the target watershed.

[0172] The evaluation sequence module maps the ship behavior characteristics in the ship trajectory set to the spatial reliability distribution map to obtain the evaluation sequence for the target watershed.

[0173] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0174] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0175] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0176] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0177] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for assessing fishery resources, characterized in that, The method includes: Extract multi-source fisheries data and vessel trajectory sets from the target watershed to obtain the original feature set of the target watershed; Based on the spatial distribution of the target watershed, the original feature set is decomposed into orthogonal waveform components and waveform coefficients of the target watershed; The data coverage density field of the target watershed is obtained by linearly superimposing the basis vectors of the orthogonal waveform components based on the waveform coefficients. The density values ​​in the data coverage density field are sorted, and the density value interval of the target watershed is divided according to the sorting results. Based on the density value range, the field values ​​in the data coverage density field are mapped to reliability levels to obtain a spatial reliability distribution map of the target watershed. By mapping the ship behavior characteristics of the ship trajectory set to the spatial reliability distribution map, an evaluation sequence for the target watershed is obtained.

2. The fishery resource assessment method as described in claim 1, characterized in that, The extraction of multi-source fisheries data and vessel trajectory sets from the target watershed yields the original feature set of the target watershed, including: Collect fish catch records, environmental monitoring data, and vessel trajectory sets for the target watershed; The fish catch record data, the environmental monitoring data, and the vessel trajectory set are spatiotemporally aligned to obtain preprocessed data for the target watershed; Multidimensional features are extracted from the preprocessed data to obtain the original feature set of the target watershed.

3. The fishery resource assessment method as described in claim 1, characterized in that, The step of decomposing the original feature set into orthogonal waveform components and waveform coefficients of the target watershed based on the spatial distribution of the target watershed includes: The original feature set is reconstructed into a two-dimensional feature matrix of the target watershed in the spatial dimension; The wavelet packet transform is performed on the two-dimensional feature matrix to obtain the multi-band wavelet packet coefficients of the target watershed; The dominant frequency band coefficients in the wavelet packet coefficients are used as the waveform coefficients of the target watershed, and the time-domain waveforms corresponding to the wavelet packet coefficients are used as the orthogonal waveform components of the target watershed.

4. The fishery resource assessment method as described in claim 3, characterized in that, The step of reconstructing the original feature set into a two-dimensional feature matrix of the target watershed in the spatial dimension includes: Establish a spatial index framework for the target watershed based on its latitude and longitude; The multidimensional feature vectors in the original feature set are assigned to the grid cells in the spatial indexing frame; Spatial interpolation is performed on the grid cells to obtain the two-dimensional feature matrix of the target watershed.

5. The fishery resource assessment method as described in claim 1, characterized in that, The step of linearly superimposing the basis vectors of the orthogonal waveform components based on the waveform coefficients to obtain the data coverage density field of the target watershed includes: The effective components of the target watershed are selected based on the amplitude of the waveform coefficients; The basis vectors are weighted based on the waveform coefficients corresponding to the effective components to obtain the density distribution surface of the target watershed; The density distribution surface is normalized to obtain the data coverage density field of the target watershed.

6. The fishery resource assessment method as described in claim 1, characterized in that, The step of sorting the density values ​​in the data coverage density field and dividing the density value interval of the target watershed according to the sorting result includes: The field values ​​of all spatial locations in the data coverage density field are sorted in ascending order to obtain the field value sequence of the target watershed. The boundary point of the target watershed is determined based on the quantile position of the field value sequence; The data coverage density field is divided using the boundary point as a threshold to obtain the density value range of the target watershed.

7. The fishery resource assessment method as described in claim 1, characterized in that, The step of mapping the field values ​​in the data coverage density field to reliability levels based on the density value interval to obtain the spatial reliability distribution map of the target watershed includes: The initial reliability weight of the target watershed is assigned based on the distribution characteristics of the density value range; The set of adjacent spatial coordinates of the target watershed is calculated based on the resolution of the target spatial location in the data coverage density field; The proportion of coordinates in the spatial coordinate set that belong to the same density value interval as the target spatial location is used as a local consistency index of the target watershed. By combining the initial reliability weights with the local consistency index, the overall reliability of the target watershed is obtained; The overall reliability is assigned to the spatial location in the data coverage density field to obtain the spatial reliability distribution map of the target watershed.

8. The fishery resource assessment method as described in claim 7, characterized in that, The formula for calculating the overall reliability is as follows: in, For the overall reliability, The initial reliability weights, It is a self-recognized constant. The entropy decay coefficient, The information entropy value. This refers to the local consistency index. The consistency contribution coefficient It is the hyperbolic tangent function. The scaling factor. This is the offset parameter.

9. A method for assessing fishery resources as described in claim 1, characterized in that, The step of mapping the ship behavior characteristics from the ship trajectory set to the spatial reliability distribution map to obtain the evaluation sequence for the target watershed includes: The spatial coordinates of the ship's behavioral characteristics are correlated with the reliability levels of the same geographical locations in the spatial reliability distribution map to obtain the correlation level pairs of the target watershed; The association level pairs are sorted in descending order to obtain the evaluation sequence of the target watershed.

10. A fishery resource assessment system for implementing the fishery resource assessment method according to any one of claims 1-9, characterized in that, The system includes: The original feature set module extracts multi-source fisheries data and vessel trajectory sets for the target watershed to obtain the original feature set of the target watershed; The component coefficient module decomposes the original feature set into orthogonal waveform components and waveform coefficients of the target watershed based on the spatial distribution of the target watershed; The data coverage density field module linearly superimposes the basis vectors of the orthogonal waveform components based on the waveform coefficients to obtain the data coverage density field of the target watershed. The density value interval module sorts the density values ​​in the data coverage density field and divides the density value interval of the target watershed according to the sorting results. The spatial reliability distribution map module maps the field values ​​in the data coverage density field to reliability levels based on the density value interval, thereby obtaining the spatial reliability distribution map of the target watershed. The evaluation sequence module maps the ship behavior characteristics in the ship trajectory set to the spatial reliability distribution map to obtain the evaluation sequence for the target watershed.

Images