A method and system for assessing fishery resources in the entire water layer using fishery acoustic methods

Abstract

The present invention discloses a method and system for assessing fishery resources in the entire water layer using fishery acoustic methods. This method uses a multi-band sonar system to pre-detect each water layer of the target water body, obtain sonar echo data, analyze the sonar detection quality, and determine the optimal sonar frequency range. Water quality characteristics are detected based on this frequency, and receiving parameters are optimized to obtain fishery resource detection data and optimize data accuracy. Based on a deep learning algorithm, a fishery resource analysis model is constructed to assess the status of fishery resources in the target water body. The assessment results are integrated to generate a spatial distribution change map of fishery resources and predict resource change trends. The system includes modules such as sonar detection, data optimization, analysis and evaluation, and resource prediction, and can accurately assess and predict fishery resources in the entire water layer, improving the scientific nature and sustainability of fishery resource management.

Description

Technical Field

[0001] The present invention relates to the technical field of fishery resource detection, and in particular to a method and system for evaluating fishery resources in a whole water layer using a fishery acoustic method. Background Art

[0002] Fishery resources are a vital component of aquatic ecosystems. Their rational utilization and management are crucial for ecological conservation, sustainable fishery development, and the protection of aquatic resources. Traditional fishery resource assessment methods rely heavily on manual sampling and statistical analysis, which suffer from long data acquisition cycles, low accuracy, and limited coverage, making it difficult to provide accurate, real-time information on fishery resources. To overcome these challenges, the use of advanced acoustic technologies, particularly multi-band sonar systems, has become an increasingly effective tool for fishery resource assessment.

[0003] Existing acoustic detection methods often rely on a single frequency band or simple signal processing algorithms. While they can capture some water quality information, their accuracy and efficiency struggle to meet the demands of high-precision assessments in complex water environments. This is particularly true when assessing fishery resources in different water layers, where environmental factors such as water quality, temperature, and salinity significantly influence the propagation characteristics of sonar echoes. Existing methods are poorly adaptable to these complex factors, resulting in inaccurate detection results.

[0004] Furthermore, the application of deep learning technology to fishery resource analysis is still in its infancy. Existing research primarily focuses on traditional statistical methods or simple model construction. The application of deep learning analysis for multi-band sonar data remains immature, and there is a lack of effective solutions for predicting and dynamically assessing changes in the spatial distribution of fishery resources.

[0005] Therefore, there is an urgent need for a new full-water layer fishery resource assessment method that can comprehensively consider multi-band sonar detection, environmental factors, data optimization and intelligent analysis to improve the accuracy and real-time performance of fishery resource assessment, thereby providing more precise technical support for the scientific management of fishery resources. Summary of the Invention

[0006] In order to solve at least one of the above technical problems, the present invention proposes a method and system for full-water layer fishery resource assessment using fishery acoustic methods.

[0007] A first aspect of the present invention provides a method for assessing fishery resources in the entire water layer using fishery acoustic methods, comprising:

[0008] performing multi-frequency sonar pre-detection of each water layer of the target water body according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency, performing sonar detection quality analysis on the pre-detection sonar echo data, and determining an optimal sonar frequency range for sonar detection of each water layer of the target water body;

[0009] detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics;

[0010] acquiring fishery resource detection sonar data within a preset time period for each water layer of the target water body according to the received parameters, extracting waveform characteristics of the fishery resource detection sonar data, determining data accuracy of the fishery resource detection sonar data according to the waveform characteristics, and optimizing the fishery resource detection sonar data according to the data accuracy to obtain optimized detection sonar data;

[0011] Constructing a fishery resource analysis model based on a deep learning algorithm, importing the optimized detection sonar data into the fishery resource analysis model to perform fishery resource assessment on the target water body, and obtaining a fishery resource assessment result;

[0012] The fishery resource assessment results within a preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain the fishery resource change prediction result.

[0013] In this solution, the multi-frequency sonar pre-detection of each water layer of the target water body is performed according to the multi-band sonar system, and pre-detection sonar echo data at each frequency is obtained. The sonar detection quality analysis of the pre-detection sonar echo data is performed to determine the optimal sonar frequency range for sonar detection of each water layer of the target water body, specifically:

[0014] Obtaining frequency band information of the multi-band sonar system for sonar detection of each water layer, determining a frequency range for sonar detection of each water layer based on the frequency band information, dividing the frequency range according to a preset step size, and obtaining a pre-detection frequency set for each water layer;

[0015] performing multi-frequency sonar detection on each water layer according to the pre-detection frequency set of each water layer according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency in the pre-detection frequency set, and marking the pre-detection sonar echo data according to the acquired water layer to obtain water layer mark-pre-detection sonar echo data;

[0016] performing a time domain analysis on the water layer marker-pre-detection sonar echo data, calculating a root mean square value of the sonar echo signal in each time period, and evaluating the echo intensity of the water layer marker-pre-detection sonar echo data according to the root mean square value;

[0017] extracting frequency spectrum data of the water layer marker-pre-detection sonar echo data based on a fast Fourier transform algorithm, calculating a signal-to-noise ratio of the sonar echo data based on the frequency spectrum data, and determining the echo clarity of the water layer marker-pre-detection sonar echo data based on the signal-to-noise ratio;

[0018] Calculating an envelope curve of the water layer marker-pre-detection sonar echo data based on a signal envelope detection algorithm, and determining echo continuity of the water layer marker-pre-detection sonar echo data according to the envelope curve;

[0019] performing sonar detection quality analysis on the water layer marker-pre-detection sonar echo data according to the echo intensity, echo clarity, and echo continuity to obtain a quality analysis result;

[0020] The optimal sonar frequency range for sonar detection of each water layer in the target water body is determined based on the quality analysis results.

[0021] In this solution, the target water body's various water layers are detected according to the preferred sonar frequency range to obtain water quality characteristics of each water layer of the target water body. The sonar propagation characteristics of the preferred sonar frequency range are determined according to the water quality characteristics. The receiving parameters of the multi-band sonar system for sonar echo data are determined according to the sonar propagation characteristics. Specifically,

[0022] Performing fishery resource detection on each water layer of the target water body according to the preferred sonar frequency range, and obtaining water quality characteristics of each water layer of the target water body based on a water quality sensor;

[0023] Acquire data on the effect of each water quality characteristic on sonar propagation of different frequencies based on an Internet search, construct a simulation model of the target water body according to the water quality characteristic, and construct a sonar signal propagation simulation model of the target water body according to the propagation effect data and the propagation effect data;

[0024] Acquiring sonar information from a multi-band sonar system performing sonar detection on each water layer of a target water body, the sonar information including sonar incident position and angle information, sonar frequency information, and sonar intensity information, and importing the sonar information into the sonar signal propagation simulation model to perform sonar signal propagation simulation;

[0025] Acquiring sonar propagation path information during the sonar signal propagation simulation process, and drawing a sonar propagation path schematic diagram based on the sonar propagation path information;

[0026] According to the sonar propagation path schematic diagram, the receiving angle of the multi-band sonar system for sonar echo data is determined; according to the sonar propagation path information, the water quality characteristics of the sonar propagation path are determined; according to the propagation impact data, the degree to which the water quality characteristics of the sonar propagation path weaken the sonar signal propagation intensity is judged; according to the weakening degree, the receiving gain of the multi-band sonar system for sonar echo data is determined, and the receiving parameters of the multi-band sonar system for sonar echo data are obtained.

[0027] In this solution, the fishery resource detection sonar data within a preset time period of each water layer of the target water body is obtained according to the receiving parameters, the waveform characteristics of the fishery resource detection sonar data are extracted, the data accuracy of the fishery resource detection sonar data is determined according to the waveform characteristics, and the fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data, specifically:

[0028] Acquire fishery resource detection sonar data within a preset time period of each water layer of the target water body according to the receiving parameters, introduce a wavelet transform algorithm, and perform multi-layer wavelet decomposition on the fishery resource detection sonar data based on the wavelet transform algorithm to obtain low-frequency approximate components and high-frequency detail components;

[0029] Performing local extreme value detection on the fishery resource detection sonar data based on the low-frequency approximate components and the high-frequency detail components, extracting amplitude characteristics of the fishery resource detection sonar data, and performing spectrum analysis on the high-frequency detail components to obtain frequency characteristics of the fishery resource detection sonar data, and integrating the amplitude characteristics with the frequency characteristics to obtain waveform characteristics of the fishery resource detection sonar data;

[0030] Dividing the fishery resource detection sonar data into sonar data segments according to a preset time length, and determining a waveform feature of each sonar data segment based on the waveform feature of the fishery resource detection sonar data;

[0031] Calculating the Euclidean distance between the waveform features of each sonar data segment, constructing a distance matrix based on the Euclidean distance, and evaluating the similarity between adjacent sonar data segments based on the distance matrix to obtain a similarity evaluation result;

[0032] Determining changes in similarity of sonar data segments in the fishery resource detection sonar data based on the similarity evaluation results, evaluating changes in data stability of the fishery resource detection sonar data based on the similarity changes, and determining data accuracy of the fishery resource detection sonar data based on the data stability changes;

[0033] The fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data.

[0034] In this solution, the fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data, specifically:

[0035] When the data accuracy is less than the preset data accuracy value, a regression analysis is performed on the data stability change of the fishery resource detection sonar data based on the least squares method, and an R-square value of the data stability change of the fishery resource detection sonar data is calculated;

[0036] When the R-squared value is greater than the preset value, the data accuracy impact item of the fishery resource detection sonar data is calibrated as the impact of multi-band sonar system performance degradation; when the R-squared value is not greater than the preset value, the data accuracy impact item of the fishery resource detection sonar data is calibrated as the impact of data collection vibration;

[0037] When the data accuracy influencing item is the performance degradation of the multi-band sonar system, obtaining the operating temperature change data of the multi-band sonar system within a preset time, jointly analyzing the operating temperature change data with the data stability change, determining the impact of the operating temperature change of the multi-band sonar system on the data stability change, and obtaining the temperature-stability impact data;

[0038] Determine a stability threshold for the multi-band sonar system to obtain fishery resource detection sonar data based on the preset data accuracy value, and determine an operating temperature range requirement for the multi-band sonar system based on the stability threshold and the temperature-stability impact data;

[0039] When the multi-band sonar system does not meet the operating temperature range requirement, cooling the multi-band sonar system and monitoring the cooling effect in real time; if the multi-band sonar system does not meet the operating temperature range requirement after the cooling operation, suspending the collection of fishery resource detection sonar data, and obtaining a first data optimization solution;

[0040] When the data accuracy is affected by vibration during data collection, the shock absorbing device of the multi-band sonar system is inspected and repaired to obtain a second data optimization solution.

[0041] In this solution, the fishery resource analysis model is constructed based on the deep learning algorithm, and the optimized detection sonar data is imported into the fishery resource analysis model to perform fishery resource assessment on the target water body, and the fishery resource assessment results are obtained, specifically:

[0042] Obtain historical fishery resource sonar detection data at different sonar frequencies and for different fishery resources using a multi-band sonar system;

[0043] Constructing a fishery resource analysis model based on a convolutional neural network, and performing data annotation on the historical fishery resource sonar detection data, wherein the data annotation includes fishery resource type annotation, fishery resource location annotation, and estimated biomass annotation;

[0044] The historical fishery resource sonar detection data after data annotation is imported into the resource analysis model for training operation, and the optimized detection sonar data is imported into the trained fishery resource analysis model to perform fishery resource assessment on each water layer in the target water body to obtain fishery resource assessment results, which include fishery resource type, fishery resource distribution location and fishery resource biomass assessment results.

[0045] In this solution, the fishery resource assessment results within a preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain the fishery resource change prediction results, specifically:

[0046] Integrate the fishery resource assessment results within a preset time period to construct a fishery resource spatial distribution change map;

[0047] extracting fishery resource change information of each water layer in the target water body according to the fishery resource spatial change map, constructing a fishery resource change prediction model based on the LSTM algorithm, and importing the fishery resource change information into the fishery resource change prediction model for training;

[0048] The fishery resource change information within the current preset time period is obtained and imported into the fishery resource change prediction model, and the fishery resource change within the future preset time period is predicted to obtain the fishery resource change prediction result.

[0049] A second aspect of the present invention further provides a system for assessing fishery resources in a whole water layer using a fishery acoustic method. The system comprises: a memory and a processor. The memory includes a program for assessing fishery resources in a whole water layer using a fishery acoustic method. When the program for assessing fishery resources in a whole water layer using a fishery acoustic method is executed by the processor, the following steps are implemented:

[0050] performing multi-frequency sonar pre-detection of each water layer of the target water body according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency, performing sonar detection quality analysis on the pre-detection sonar echo data, and determining an optimal sonar frequency range for sonar detection of each water layer of the target water body;

[0051] detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics;

[0052] acquiring fishery resource detection sonar data within a preset time period for each water layer of the target water body according to the received parameters, extracting waveform characteristics of the fishery resource detection sonar data, determining data accuracy of the fishery resource detection sonar data according to the waveform characteristics, and optimizing the fishery resource detection sonar data according to the data accuracy to obtain optimized detection sonar data;

[0053] Constructing a fishery resource analysis model based on a deep learning algorithm, importing the optimized detection sonar data into the fishery resource analysis model to perform fishery resource assessment on the target water body, and obtaining a fishery resource assessment result;

[0054] The fishery resource assessment results within a preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain the fishery resource change prediction result.

[0055] The present invention discloses a method and system for assessing fishery resources in the entire water layer using fishery acoustic methods. This method uses a multi-band sonar system to pre-detect each water layer of the target water body, obtain sonar echo data, analyze the sonar detection quality, and determine the optimal sonar frequency range. Water quality characteristics are detected based on this frequency, and receiving parameters are optimized to obtain fishery resource detection data and optimize data accuracy. Based on a deep learning algorithm, a fishery resource analysis model is constructed to assess the status of fishery resources in the target water body. The assessment results are integrated to generate a spatial distribution change map of fishery resources and predict resource change trends. The system includes modules such as sonar detection, data optimization, analysis and evaluation, and resource prediction, and can accurately assess and predict fishery resources in the entire water layer, improving the scientific nature and sustainability of fishery resource management. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flow chart showing a method for assessing fishery resources in the entire water layer using fishery acoustic methods according to the present invention is shown;

[0057] Figure 2 A flow chart showing the results of fishery resource assessment obtained by the present invention is shown;

[0058] Figure 3 A flow chart showing the results of the fishery resource change prediction obtained by the present invention is shown;

[0059] Figure 4 A block diagram of a full-water layer fishery resource assessment system using fishery acoustic methods according to the present invention is shown. DETAILED DESCRIPTION

[0060] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0061] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0062] Figure 1 A flow chart of a method for assessing fishery resources in the entire water layer using fishery acoustic methods according to the present invention is shown.

[0063] like Figure 1 As shown, the first aspect of the present invention provides a method for assessing fishery resources in the entire water layer using a fishery acoustic method, comprising:

[0064] S102, performing multi-frequency sonar pre-detection of each water layer of the target water body using the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency, performing sonar detection quality analysis on the pre-detection sonar echo data, and determining an optimal sonar frequency range for sonar detection of each water layer of the target water body;

[0065] S104, detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics;

[0066] S106, obtaining fishery resource detection sonar data for each water layer of the target water body within a preset time period according to the received parameters, extracting waveform characteristics of the fishery resource detection sonar data, determining data accuracy of the fishery resource detection sonar data according to the waveform characteristics, and optimizing the fishery resource detection sonar data according to the data accuracy to obtain optimized detection sonar data;

[0067] S108, constructing a fishery resource analysis model based on a deep learning algorithm, importing the optimized detection sonar data into the fishery resource analysis model to perform fishery resource assessment on the target water body, and obtaining a fishery resource assessment result;

[0068] S110, integrating the fishery resource assessment results within a preset time period to construct a fishery resource spatial distribution change map, and predicting the fishery resource changes of the target water body based on the fishery resource spatial distribution change map to obtain a fishery resource change prediction result.

[0069] It should be noted that a multi-band sonar system performs multi-frequency sonar pre-scanning of each water layer in the target water body, acquiring sonar echo data at different frequencies. Sonar detection quality analysis accurately determines the optimal sonar frequency range for each water layer, thereby improving detection accuracy and efficiency. Further detection using this optimized sonar frequency range and analyzing sonar propagation characteristics in conjunction with water quality characteristics optimizes reception parameters based on water quality differences across different water layers, ensuring optimal signal reception quality. Waveform feature extraction and data accuracy analysis of sonar echo data, combined with wavelet transform algorithms, optimize the sonar data to eliminate noise and errors, improving data reliability and accuracy. A fishery resource analysis model based on deep learning algorithms intelligently analyzes the optimized detection data to accurately assess key information such as the type, distribution, and biomass of fishery resources in the water body. Furthermore, by integrating fishery resource assessment results and constructing a spatial distribution change map, the temporal and spatial trends of fishery resources can be dynamically reflected. Combined with a predictive model, scientific predictions can be made about future changes in fishery resources in the water, providing a precise basis for the rational utilization and sustainable management of fishery resources. The overall approach has significant technical advantages in improving assessment accuracy, optimizing the detection process, and achieving real-time monitoring and dynamic prediction.

[0070] According to an embodiment of the present invention, the multi-frequency sonar pre-detection of each water layer of the target water body is performed according to the multi-band sonar system, pre-detection sonar echo data at each frequency is obtained, and sonar detection quality analysis is performed on the pre-detection sonar echo data to determine the optimal sonar frequency range for sonar detection of each water layer of the target water body, specifically:

[0071] Obtaining frequency band information of the multi-band sonar system for sonar detection of each water layer, determining a frequency range for sonar detection of each water layer based on the frequency band information, dividing the frequency range according to a preset step size, and obtaining a pre-detection frequency set for each water layer;

[0072] performing multi-frequency sonar detection on each water layer according to the pre-detection frequency set of each water layer according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency in the pre-detection frequency set, and marking the pre-detection sonar echo data according to the acquired water layer to obtain water layer mark-pre-detection sonar echo data;

[0073] performing a time domain analysis on the water layer marker-pre-detection sonar echo data, calculating a root mean square value of the sonar echo signal in each time period, and evaluating the echo intensity of the water layer marker-pre-detection sonar echo data according to the root mean square value;

[0074] extracting frequency spectrum data of the water layer marker-pre-detection sonar echo data based on a fast Fourier transform algorithm, calculating a signal-to-noise ratio of the sonar echo data based on the frequency spectrum data, and determining the echo clarity of the water layer marker-pre-detection sonar echo data based on the signal-to-noise ratio;

[0075] Calculating an envelope curve of the water layer marker-pre-detection sonar echo data based on a signal envelope detection algorithm, and determining echo continuity of the water layer marker-pre-detection sonar echo data according to the envelope curve;

[0076] performing sonar detection quality analysis on the water layer marker-pre-detection sonar echo data according to the echo intensity, echo clarity, and echo continuity to obtain a quality analysis result;

[0077] The optimal sonar frequency range for sonar detection of each water layer in the target water body is determined based on the quality analysis results.

[0078] It should be noted that by comprehensively analyzing the three indicators of echo intensity, echo clarity, and echo continuity, the system can accurately assess the sonar detection quality of each water layer, thereby selecting the optimal sonar frequency range for different water layers. In water layers with strong echo intensity, high echo clarity, and echo continuity, higher-frequency sonar is preferred for detection. This can provide higher-resolution detection and is suitable for fishery resource detection with smaller, more distant targets or those requiring higher-precision positioning. In water layers with weaker echo intensity, lower echo clarity, and poorer echo continuity, the system will prefer lower-frequency sonar, which helps reduce signal attenuation and noise interference, and improve the stability and reliability of detection. By accurately selecting the frequency range, deviations in frequency selection can be avoided, improving the quality and accuracy of sonar detection data, and thus enhancing the accuracy and feasibility of fishery resource assessment. The frequency band information is the frequency band used by the multi-band sonar system for sonar detection of different water layers. This present invention proposes for the first time the use of a multi-band sonar system for dynamic monitoring of fishery resources in different water layers. High-frequency sonar is used to conduct detailed detection of surface resources, medium-frequency sonar is used to cover the resource distribution in the middle layer, and low-frequency sonar is used to detect deep water layers or bottom biological communities. This multi-band integrated use solution can comprehensively cover fishery resources at different depths of the water body, making up for the shortcomings of traditional single-band sonar. The water layer is divided into surface layer, middle layer and bottom layer. The pre-detection frequency set of each water layer, one set represents the pre-detection frequency set for a water layer, for example, the pre-detection frequency set of the high-frequency band only detects the surface water layer. The multi-frequency sonar pre-detection is to perform sonar detection on each water layer of the target water body in turn according to each pre-detection frequency in the pre-detection frequency set.

[0079] According to an embodiment of the present invention, detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining the sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining the receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics are specifically as follows:

[0080] Performing fishery resource detection on each water layer of the target water body according to the preferred sonar frequency range, and obtaining water quality characteristics of each water layer of the target water body based on a water quality sensor;

[0081] Acquire data on the effect of each water quality characteristic on sonar propagation of different frequencies based on an Internet search, construct a simulation model of the target water body according to the water quality characteristic, and construct a sonar signal propagation simulation model of the target water body according to the propagation effect data and the propagation effect data;

[0082] Acquiring sonar information from a multi-band sonar system performing sonar detection on each water layer of a target water body, the sonar information including sonar incident position and angle information, sonar frequency information, and sonar intensity information, and importing the sonar information into the sonar signal propagation simulation model to perform sonar signal propagation simulation;

[0083] Acquiring sonar propagation path information during the sonar signal propagation simulation process, and drawing a sonar propagation path schematic diagram based on the sonar propagation path information;

[0084] According to the sonar propagation path schematic diagram, the receiving angle of the multi-band sonar system for sonar echo data is determined; according to the sonar propagation path information, the water quality characteristics of the sonar propagation path are determined; according to the propagation impact data, the degree to which the water quality characteristics of the sonar propagation path weaken the sonar signal propagation intensity is judged; according to the weakening degree, the receiving gain of the multi-band sonar system for sonar echo data is determined, and the receiving parameters of the multi-band sonar system for sonar echo data are obtained.

[0085] It's important to note that water quality conditions (such as temperature, salinity, and turbidity) significantly affect the propagation characteristics of sonar signals. Traditional sonar systems often rely on fixed frequencies for detection, but they ignore the differences in propagation characteristics under varying water conditions. This can affect detection performance, particularly in different water layers, where varying water quality attenuates and interferes with sonar signals, reducing detection accuracy. Therefore, it's necessary to dynamically adjust the sonar system's operating parameters based on water quality characteristics to optimize the reception of sonar echo data. The system acquires water quality sensor data in real time, accurately reflecting changes in water quality within each water layer and dynamically adjusting the sonar frequency selection. By modeling the impact of different water quality characteristics on sonar signal propagation, the system can predict signal attenuation and propagation paths under different water conditions, providing accurate simulation data for sonar signal propagation. Combining sonar information from a multi-band sonar system (such as incident angle, frequency, and intensity), the system can accurately calculate the signal propagation path, create a schematic diagram of the propagation path, and use this diagram to determine the optimal reception angle and gain. By calculating the degree to which water quality characteristics along the sonar signal's propagation path attenuate the signal, the system can further optimize receiver gain, ensuring high-quality echo signals in varying water quality environments. This significantly improves detection accuracy across different water layers, reduces signal attenuation and noise interference, and ensures the stability and accuracy of detection data. Water quality characteristics include temperature, salinity, turbidity, dissolved oxygen content, and pH; propagation impact data includes data on the degree to which sonar propagation refraction affects sonar propagation and the degree to which sonar propagation intensity weakens; and sonar propagation path information includes both incident and reflected path information.

[0086] According to an embodiment of the present invention, the fishery resource detection sonar data within a preset time period of each water layer of the target water body is obtained according to the receiving parameters, the waveform characteristics of the fishery resource detection sonar data are extracted, the data accuracy of the fishery resource detection sonar data is determined according to the waveform characteristics, and the fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data, specifically:

[0087] Acquire fishery resource detection sonar data within a preset time period of each water layer of the target water body according to the receiving parameters, introduce a wavelet transform algorithm, and perform multi-layer wavelet decomposition on the fishery resource detection sonar data based on the wavelet transform algorithm to obtain low-frequency approximate components and high-frequency detail components;

[0088] Performing local extreme value detection on the fishery resource detection sonar data based on the low-frequency approximate components and the high-frequency detail components, extracting amplitude characteristics of the fishery resource detection sonar data, and performing spectrum analysis on the high-frequency detail components to obtain frequency characteristics of the fishery resource detection sonar data, and integrating the amplitude characteristics with the frequency characteristics to obtain waveform characteristics of the fishery resource detection sonar data;

[0089] Dividing the fishery resource detection sonar data into sonar data segments according to a preset time length, and determining a waveform feature of each sonar data segment based on the waveform feature of the fishery resource detection sonar data;

[0090] Calculating the Euclidean distance between the waveform features of each sonar data segment, constructing a distance matrix based on the Euclidean distance, and evaluating the similarity between adjacent sonar data segments based on the distance matrix to obtain a similarity evaluation result;

[0091] Determining changes in similarity of sonar data segments in the fishery resource detection sonar data based on the similarity evaluation results, evaluating changes in data stability of the fishery resource detection sonar data based on the similarity changes, and determining data accuracy of the fishery resource detection sonar data based on the data stability changes;

[0092] The fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data.

[0093] It should be noted that due to variations in the water environment and water layers, sonar signals may be subject to varying degrees of attenuation, distortion, and noise. Signal instability is particularly pronounced in complex water environments. By applying a wavelet transform algorithm to perform multi-level wavelet decomposition on fishery resource detection sonar data, the sonar echo signal can be effectively divided into low-frequency approximate components and high-frequency detail components. The low-frequency component primarily represents the overall signal trend, while the high-frequency component reflects detailed features and variation, particularly helping to detect potential noise and interference. The low- and high-frequency components derived from the wavelet transform, combined with local extrema detection methods, can effectively identify amplitude characteristics in sonar signals, further improving waveform recognition accuracy. Spectral analysis of the high-frequency component extracts the signal's frequency signature, providing more accurate feature data for subsequent signal processing. By extracting waveform features from each sonar data segment, calculating the Euclidean distance between adjacent segments, and constructing a distance matrix, we can quantitatively assess the similarity between data segments and further analyze the stability of the sonar data. This similarity assessment can be used to determine the changing trends of data segments, identify potential outliers, and assess data stability. According to the changes in stability, the data accuracy can be dynamically adjusted to ensure the reliability and consistency of the detection data.

[0094] According to an embodiment of the present invention, the fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data, specifically:

[0095] When the data accuracy is less than the preset data accuracy value, a regression analysis is performed on the data stability change of the fishery resource detection sonar data based on the least squares method, and an R-square value of the data stability change of the fishery resource detection sonar data is calculated;

[0096] When the R-squared value is greater than the preset value, the data accuracy impact item of the fishery resource detection sonar data is calibrated as the impact of multi-band sonar system performance degradation; when the R-squared value is not greater than the preset value, the data accuracy impact item of the fishery resource detection sonar data is calibrated as the impact of data collection vibration;

[0097] When the data accuracy influencing item is the performance degradation of the multi-band sonar system, obtaining the operating temperature change data of the multi-band sonar system within a preset time, jointly analyzing the operating temperature change data with the data stability change, determining the impact of the operating temperature change of the multi-band sonar system on the data stability change, and obtaining the temperature-stability impact data;

[0098] Determine a stability threshold for the multi-band sonar system to obtain fishery resource detection sonar data based on the preset data accuracy value, and determine an operating temperature range requirement for the multi-band sonar system based on the stability threshold and the temperature-stability impact data;

[0099] When the multi-band sonar system does not meet the operating temperature range requirement, cooling the multi-band sonar system and monitoring the cooling effect in real time; if the multi-band sonar system does not meet the operating temperature range requirement after the cooling operation, suspending the collection of fishery resource detection sonar data, and obtaining a first data optimization solution;

[0100] When the data accuracy is affected by vibration during data collection, the shock absorbing device of the multi-band sonar system is inspected and repaired to obtain a second data optimization solution.

[0101] It should be noted that when using regression analysis (least squares regression) to analyze changes in the stability of fishery resource detection sonar data, the R-squared value is used to measure the model's goodness of fit, that is, the degree to which it explains data variation. Specifically, the R-squared value represents the proportion of the total variation explained by the model. Values ​​closer to 1 indicate a higher degree of explanation for the data variation; values ​​closer to 0 indicate a poor explanation for the data variation. In this method, the R-squared value assesses the relationship between changes in data stability and data accuracy. Therefore, a high R-squared value indicates that changes in data stability (such as gradual attenuation or enhancement) can be well explained by degradation in the performance of the multi-band sonar system. Multi-band sonar system performance degradation often manifests as a gradual loss of accuracy, resulting in data instability or fluctuation within certain time periods. This type of performance degradation typically occurs gradually, making the resulting changes in data stability predictable and exhibiting a high regression goodness of fit (i.e., a high R-squared value). A high R-squared value indicates a strong linear relationship between changes in data stability and these system performance degradation factors, thus confirming that data accuracy issues are caused by degradation in the multi-band sonar system. When the R-squared value is low, it means that the change in data stability cannot be well explained by the degradation of system performance, and there may be interference from other external factors. In this case, vibration may be a major influencing factor. Vibration will directly affect the accuracy of sonar equipment, resulting in large fluctuations in data, and the impact of vibration is usually short-term and irregular, making it difficult for the regression model to explain these changes. The impact of vibration is usually manifested as nonlinear and irregular fluctuations, which makes the fitting effect of the regression model poor and the R-squared value low. By adopting different optimization strategies in different situations, the efficiency and adaptability of the data optimization process are ensured. When the performance degradation of the multi-band sonar system affects the data accuracy, the system is optimized by temperature adjustment; and when vibration affects the data accuracy, it is optimized by repairing the shock absorber. This flexible optimization method can take the most effective measures for different influencing factors, thereby improving the quality of sonar data for fishery resource detection. The cooling operation includes starting the air cooling device.

[0102] Figure 2 A flow chart of obtaining fishery resource assessment results according to the present invention is shown.

[0103] According to an embodiment of the present invention, the fishery resource analysis model is constructed based on the deep learning algorithm, and the optimized detection sonar data is imported into the fishery resource analysis model to perform fishery resource assessment on the target water body, thereby obtaining a fishery resource assessment result, which is specifically:

[0104] S202, obtaining historical fishery resource sonar detection data of different fishery resources at different sonar frequencies of a multi-band sonar system;

[0105] S204, constructing a fishery resource analysis model based on a convolutional neural network, and performing data annotation on the historical fishery resource sonar detection data, wherein the data annotation includes annotation of fishery resource type, annotation of fishery resource location, and annotation of estimated biomass;

[0106] S206, importing the historical fishery resource sonar detection data after data annotation into the resource analysis model for training operation, importing the optimized detection sonar data into the trained fishery resource analysis model to conduct fishery resource assessment on each water layer in the target water body, and obtaining fishery resource assessment results, which include fishery resource type, fishery resource distribution location and fishery resource biomass assessment results.

[0107] It should be noted that historical data, including the type, location, and estimated biomass of fishery resources, is annotated using a convolutional neural network (CNN) to provide high-quality training data for the model. This process accurately labels the characteristics of different fishery resources, ensuring the model's high accuracy and robustness in identifying fishery resources in the target water body. The trained fishery resource analysis model can then be used to evaluate the optimized sonar data, providing detailed fishery resource assessment results for each water layer within the target water body. This improves the automation and accuracy of fishery resource assessment and reduces the need for manual intervention. Furthermore, the deep learning-based model continuously learns and optimizes, continuously improving its recognition accuracy and adaptability, enabling it to cope with changes in different water conditions and fishery resource types, thereby enhancing the scientific nature and reliability of fishery resource monitoring and assessment. Fishery resource types refer to the classification of fishery resources, including different species of aquatic animals, based on factors such as biological characteristics, habitat, and economic value in fishery resource assessments.

[0108] Figure 3 The flowchart of the present invention for obtaining the prediction results of fishery resources changes is shown.

[0109] According to an embodiment of the present invention, the fishery resource assessment results within a preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain a fishery resource change prediction result, which is specifically:

[0110] S302, integrating the fishery resource assessment results within a preset time period to construct a fishery resource spatial distribution change map;

[0111] S304, extracting fishery resource change information of each water layer in the target water body based on the fishery resource spatial change map, constructing a fishery resource change prediction model based on the LSTM algorithm, and importing the fishery resource change information into the fishery resource change prediction model for training;

[0112] S306, obtaining the fishery resource change information within the current preset time period and importing it into the fishery resource change prediction model, predicting the fishery resource change within the future preset time period, and obtaining the fishery resource change prediction result.

[0113] It's important to note that by integrating fishery resource assessment results from multiple time periods and constructing a map of the spatial distribution of fishery resources, we can clearly demonstrate the changing trends in the spatial distribution of fishery resources in each water layer of the target water body, providing a high-precision visualization of fishery resource changes and, in turn, providing a basis for resource management and decision-making. The long-term prediction capabilities of the LSTM algorithm can effectively process data with time series characteristics. By incorporating historical fishery resource change information into the LSTM prediction model, we can accurately predict future changes in fishery resources. By promptly capturing signals of resource changes, we can help managers take early warning measures before problems arise, avoiding resource depletion or ecological risks.

[0114] Figure 4 A block diagram of a full-water layer fishery resource assessment system using fishery acoustic methods according to the present invention is shown.

[0115] A second aspect of the present invention further provides a system 4 for assessing fishery resources in a whole water layer using a fishery acoustic method. The system comprises: a memory 41 and a processor 42. The memory includes a program for assessing fishery resources in a whole water layer using a fishery acoustic method. When the program is executed by the processor, the following steps are implemented:

[0116] performing multi-frequency sonar pre-detection of each water layer of the target water body according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency, performing sonar detection quality analysis on the pre-detection sonar echo data, and determining an optimal sonar frequency range for sonar detection of each water layer of the target water body;

[0117] detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics;

[0118] acquiring fishery resource detection sonar data within a preset time period for each water layer of the target water body according to the received parameters, extracting waveform characteristics of the fishery resource detection sonar data, determining data accuracy of the fishery resource detection sonar data according to the waveform characteristics, and optimizing the fishery resource detection sonar data according to the data accuracy to obtain optimized detection sonar data;

[0119] Constructing a fishery resource analysis model based on a deep learning algorithm, importing the optimized detection sonar data into the fishery resource analysis model to perform fishery resource assessment on the target water body, and obtaining a fishery resource assessment result;

[0120] The fishery resource assessment results within a preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain the fishery resource change prediction result.

[0121] The present invention discloses a method and system for assessing fishery resources in the entire water layer using fishery acoustic methods. This method uses a multi-band sonar system to pre-detect each water layer of the target water body, obtain sonar echo data, analyze the sonar detection quality, and determine the optimal sonar frequency range. Water quality characteristics are detected based on this frequency, and receiving parameters are optimized to obtain fishery resource detection data and optimize data accuracy. Based on a deep learning algorithm, a fishery resource analysis model is constructed to assess the status of fishery resources in the target water body. The assessment results are integrated to generate a spatial distribution change map of fishery resources and predict resource change trends. The system includes modules such as sonar detection, data optimization, analysis and evaluation, and resource prediction, and can accurately assess and predict fishery resources in the entire water layer, improving the scientific nature and sustainability of fishery resource management.

[0122] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0123] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0124] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0125] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for assessing fishery resources in the whole water layer using fishery acoustic methods, characterized in that: The following steps are involved: Performing multi-frequency sonar pre-detection of each water layer of the target water body according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency, performing sonar detection quality analysis on the pre-detection sonar echo data, and determining an optimal sonar frequency range for sonar detection of each water layer of the target water body; detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics; acquiring fishery resource detection sonar data within a preset time period for each water layer of the target water body according to the received parameters, extracting waveform characteristics of the fishery resource detection sonar data, determining data accuracy of the fishery resource detection sonar data according to the waveform characteristics, and optimizing the fishery resource detection sonar data according to the data accuracy to obtain optimized detection sonar data; Constructing a fishery resource analysis model based on a deep learning algorithm, importing the optimized detection sonar data into the fishery resource analysis model to perform fishery resource assessment on the target water body, and obtaining a fishery resource assessment result; The fishery resource assessment results within a preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain the fishery resource change prediction result.

2. The method for assessing fishery resources in the whole water layer using fishery acoustics according to claim 1, characterized in that: The multi-frequency sonar pre-detection of each water layer of the target water body is performed according to the multi-band sonar system, pre-detection sonar echo data at each frequency is obtained, and sonar detection quality analysis is performed on the pre-detection sonar echo data to determine the optimal sonar frequency range for sonar detection of each water layer of the target water body, specifically: Obtaining frequency band information of the multi-band sonar system for sonar detection of each water layer, determining a frequency range for sonar detection of each water layer based on the frequency band information, dividing the frequency range according to a preset step size, and obtaining a pre-detection frequency set for each water layer; performing multi-frequency sonar detection on each water layer according to the pre-detection frequency set of each water layer according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency in the pre-detection frequency set, and marking the pre-detection sonar echo data according to the acquired water layer to obtain water layer mark-pre-detection sonar echo data; performing a time domain analysis on the water layer marker-pre-detection sonar echo data, calculating a root mean square value of the sonar echo signal in each time period, and evaluating the echo intensity of the water layer marker-pre-detection sonar echo data according to the root mean square value; extracting frequency spectrum data of the water layer marker-pre-detection sonar echo data based on a fast Fourier transform algorithm, calculating a signal-to-noise ratio of the sonar echo data based on the frequency spectrum data, and determining the echo clarity of the water layer marker-pre-detection sonar echo data based on the signal-to-noise ratio; Calculating an envelope curve of the water layer marker-pre-detection sonar echo data based on a signal envelope detection algorithm, and determining echo continuity of the water layer marker-pre-detection sonar echo data according to the envelope curve; performing sonar detection quality analysis on the water layer marker-pre-detection sonar echo data according to the echo intensity, echo clarity, and echo continuity to obtain a quality analysis result; The optimal sonar frequency range for sonar detection of each water layer in the target water body is determined based on the quality analysis results.

3. The method for assessing fishery resources in the whole water layer using fishery acoustics according to claim 1, characterized in that: The method comprises detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics, specifically: Performing fishery resource detection on each water layer of the target water body according to the preferred sonar frequency range, and obtaining water quality characteristics of each water layer of the target water body based on a water quality sensor; Acquire data on the effect of each water quality characteristic on sonar propagation of different frequencies based on an Internet search, construct a simulation model of the target water body based on the water quality characteristic, and construct a sonar signal propagation simulation model of the target water body based on the propagation effect data and the simulation model; Acquiring sonar information from a multi-band sonar system performing sonar detection on each water layer of a target water body, the sonar information including sonar incident position and angle information, sonar frequency information, and sonar intensity information, and importing the sonar information into the sonar signal propagation simulation model to perform sonar signal propagation simulation; Acquiring sonar propagation path information during the sonar signal propagation simulation process, and drawing a sonar propagation path schematic diagram based on the sonar propagation path information; According to the sonar propagation path schematic diagram, the receiving angle of the multi-band sonar system for sonar echo data is determined; according to the sonar propagation path information, the water quality characteristics of the sonar propagation path are determined; according to the propagation impact data, the degree to which the water quality characteristics of the sonar propagation path weaken the sonar signal propagation intensity is judged; according to the weakening degree, the receiving gain of the multi-band sonar system for sonar echo data is determined, and the receiving parameters of the multi-band sonar system for sonar echo data are obtained.

4. The method for assessing fishery resources in the whole water layer using fishery acoustics according to claim 1, wherein: The method comprises: obtaining fishery resource detection sonar data within a preset time period of each water layer of the target water body according to the receiving parameters, extracting waveform characteristics of the fishery resource detection sonar data, determining data accuracy of the fishery resource detection sonar data according to the waveform characteristics, and optimizing the fishery resource detection sonar data according to the data accuracy to obtain optimized detection sonar data. Specifically, Acquire fishery resource detection sonar data within a preset time period of each water layer of the target water body according to the receiving parameters, introduce a wavelet transform algorithm, and perform multi-layer wavelet decomposition on the fishery resource detection sonar data based on the wavelet transform algorithm to obtain low-frequency approximate components and high-frequency detail components; Performing local extreme value detection on the fishery resource detection sonar data based on the low-frequency approximate components and the high-frequency detail components, extracting amplitude characteristics of the fishery resource detection sonar data, and performing spectrum analysis on the high-frequency detail components to obtain frequency characteristics of the fishery resource detection sonar data, and integrating the amplitude characteristics with the frequency characteristics to obtain waveform characteristics of the fishery resource detection sonar data; Dividing the fishery resource detection sonar data into sonar data segments according to a preset time length, and determining a waveform feature of each sonar data segment based on the waveform feature of the fishery resource detection sonar data; Calculating the Euclidean distance between the waveform features of each sonar data segment, constructing a distance matrix based on the Euclidean distance, and evaluating the similarity between adjacent sonar data segments based on the distance matrix to obtain a similarity evaluation result; Determining changes in similarity of sonar data segments in the fishery resource detection sonar data based on the similarity evaluation results, evaluating changes in data stability of the fishery resource detection sonar data based on the similarity changes, and determining data accuracy of the fishery resource detection sonar data based on the data stability changes; The fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data.

5. The method for assessing fishery resources in the whole water layer using fishery acoustics according to claim 4, characterized in that: The fishery resource detection sonar data is optimized according to the data accuracy to obtain optimized detection sonar data, specifically: When the data accuracy is less than the preset data accuracy value, a regression analysis is performed on the data stability change of the fishery resource detection sonar data based on the least squares method, and an R-square value of the data stability change of the fishery resource detection sonar data is calculated; When the R-squared value is greater than the preset value, the data accuracy impact item of the fishery resource detection sonar data is calibrated as the impact of multi-band sonar system performance degradation; when the R-squared value is not greater than the preset value, the data accuracy impact item of the fishery resource detection sonar data is calibrated as the impact of data collection vibration; When the data accuracy influencing item is the performance degradation of the multi-band sonar system, obtaining the operating temperature change data of the multi-band sonar system within a preset time, jointly analyzing the operating temperature change data with the data stability change, determining the impact of the operating temperature change of the multi-band sonar system on the data stability change, and obtaining the temperature-stability impact data; Determine a stability threshold for the multi-band sonar system to obtain fishery resource detection sonar data based on the preset data accuracy value, and determine an operating temperature range requirement for the multi-band sonar system based on the stability threshold and the temperature-stability impact data; When the multi-band sonar system does not meet the operating temperature range requirement, cooling the multi-band sonar system and monitoring the cooling effect in real time; if the multi-band sonar system does not meet the operating temperature range requirement after the cooling operation, suspending the collection of fishery resource detection sonar data, and obtaining a first data optimization solution; When the data accuracy is affected by vibration during data collection, the shock absorbing device of the multi-band sonar system is inspected and repaired to obtain a second data optimization solution.

6. The method for assessing fishery resources in the whole water layer using fishery acoustics according to claim 1, characterized in that: The fishery resource analysis model is constructed based on the deep learning algorithm, and the optimized detection sonar data is imported into the fishery resource analysis model to perform fishery resource assessment on the target water body to obtain the fishery resource assessment result, which is specifically: Obtain historical fishery resource sonar detection data at different sonar frequencies and for different fishery resources using a multi-band sonar system; Constructing a fishery resource analysis model based on a convolutional neural network, and performing data annotation on the historical fishery resource sonar detection data, wherein the data annotation includes fishery resource type annotation, fishery resource location annotation, and estimated biomass annotation; The historical fishery resource sonar detection data after data annotation is imported into the resource analysis model for training operation, and the optimized detection sonar data is imported into the trained fishery resource analysis model to perform fishery resource assessment on each water layer in the target water body to obtain fishery resource assessment results, which include fishery resource type, fishery resource distribution location and fishery resource biomass assessment results.

7. The method for assessing fishery resources in the whole water layer using fishery acoustics according to claim 1, characterized in that: The fishery resource assessment results within the preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain the fishery resource change prediction result, specifically: Integrate the fishery resource assessment results within a preset time period to construct a fishery resource spatial distribution change map; extracting fishery resource change information of each water layer in the target water body according to the fishery resource spatial change map, constructing a fishery resource change prediction model based on the LSTM algorithm, and importing the fishery resource change information into the fishery resource change prediction model for training; The fishery resource change information within the current preset time period is obtained and imported into the fishery resource change prediction model, and the fishery resource change within the future preset time period is predicted to obtain the fishery resource change prediction result.

8. A system for assessing fishery resources in the entire water layer using fishery acoustic methods, characterized in that: The system for assessing fishery resources in a whole water layer using a fishery acoustic method includes a storage device and a processor. The storage device includes a program for assessing fishery resources in a whole water layer using a fishery acoustic method. When the program for assessing fishery resources in a whole water layer using a fishery acoustic method is executed by the processor, the following steps are implemented: Performing multi-frequency sonar pre-detection of each water layer of the target water body according to the multi-band sonar system, obtaining pre-detection sonar echo data at each frequency, performing sonar detection quality analysis on the pre-detection sonar echo data, and determining an optimal sonar frequency range for sonar detection of each water layer of the target water body; detecting each water layer of the target water body according to the preferred sonar frequency range, obtaining water quality characteristics of each water layer of the target water body, determining sonar propagation characteristics of the preferred sonar frequency range according to the water quality characteristics, and determining receiving parameters of the multi-band sonar system for sonar echo data according to the sonar propagation characteristics; acquiring fishery resource detection sonar data within a preset time period for each water layer of the target water body according to the received parameters, extracting waveform characteristics of the fishery resource detection sonar data, determining data accuracy of the fishery resource detection sonar data according to the waveform characteristics, and optimizing the fishery resource detection sonar data according to the data accuracy to obtain optimized detection sonar data; Constructing a fishery resource analysis model based on a deep learning algorithm, importing the optimized detection sonar data into the fishery resource analysis model to perform fishery resource assessment on the target water body, and obtaining a fishery resource assessment result; The fishery resource assessment results within a preset time period are integrated to construct a fishery resource spatial distribution change map, and the fishery resource changes of the target water body are predicted based on the fishery resource spatial distribution change map to obtain the fishery resource change prediction result.

Images