Method, device and equipment for measuring amount of seafood in flowing water area

Abstract

The invention relates to the technical field of water area seafood quantity detection, in particular to a method, device and equipment for measuring the seafood quantity in a flowing water area. Generating a propeller smooth rotating speed parameter according to a preset fluid mechanics principle, a preset appearance parameter, the hull attitude sensing data and the water area sensing data; acquiring a navigation state according to the optimized navigation route; when the navigation state analysis result is that the ship body is located in the detection area, analyzing the optical imaging data, the sonar detection data and the water quality parameter data based on a preset multi-mode convolutional neural network model and a preset Poisson curved surface reconstruction algorithm; generating a seafood resource three-dimensional distribution map according to the first analysis result; analyzing the seafood resource three-dimensional distribution map to obtain a seafood amount and population density data set; the navigation route is optimized by dynamically adjusting the rotating speed of the propellers, energy consumption and vibration are reduced, the navigation efficiency and safety are improved, a seafood resource three-dimensional distribution map is constructed through multi-source data and an advanced algorithm, and the resource evaluation accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of seafood quantity detection technology in aquatic waters, specifically to a method, apparatus, and equipment for measuring seafood quantity in flowing water. Background Technology

[0002] Existing methods for coordinating aquatic seafood resource monitoring with vessel navigation have significant shortcomings. Traditional approaches to aquatic seafood resource surveys often employ static, single-method approaches or rely solely on optical or sonar detection. This results in poor data acquisition timeliness, limited spatial coverage, and difficulty adapting to dynamic vessel navigation scenarios. Furthermore, vessel navigation control and resource monitoring are disconnected, and navigation routes cannot adequately match resource survey requirements. In the data fusion and analysis phase, multi-source heterogeneous data such as optical imaging, sonar detection, and water quality parameters lack efficient integration and analysis methods. Simple overlaying or single-algorithm processing leads to significant errors in quantity estimation and population density analysis, making it impossible to accurately construct three-dimensional distribution maps. This hinders the provision of high-precision, dynamic data support for nearshore aquaculture assessment, fisheries resource surveys, and ecological restoration, thus restricting the scientific rigor and efficiency of aquatic resource development and ecological management. Summary of the Invention

[0003] To address the shortcomings of the prior art, this invention proposes a method for measuring the quantity of seafood in flowing water.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for measuring seafood quantity in flowing water includes: acquiring hull attitude sensing data and water area sensing parameter data; generating propeller smoothing speed parameters based on preset fluid dynamics principles, preset shape parameters, hull attitude sensing data, and water area sensing data; acquiring a navigation route and adjusting the navigation route according to the propeller smoothing speed parameters to obtain an optimized navigation route; acquiring the navigation state based on the optimized navigation route; analyzing the navigation state to obtain a navigation state analysis result; when the navigation state analysis result indicates that the hull is located in the detection area, acquiring optical imaging data, sonar detection data, and water quality parameter data; analyzing the optical imaging data, sonar detection data, and water quality parameter data based on a preset multimodal convolutional neural network model and a preset Poisson surface reconstruction algorithm to obtain a first analysis result; generating a three-dimensional distribution map of seafood resources based on the first analysis result; and analyzing the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density.

[0005] Furthermore, the step of generating smooth propeller speed parameters based on preset fluid dynamics principles, preset shape parameters, hull attitude sensing data, and water area sensing data includes: generating a drag coefficient based on fluid dynamics principles and shape parameters; generating an optimal counterweight based on a preset fluid dynamics model, drag coefficient, hull attitude sensing data, water area sensing data, and preset target anchoring coordinates; generating propeller speed parameters based on the optimal counterweight and a preset propeller thrust coefficient; and smoothing the propeller speed parameters based on a preset smoothing curve and a preset smoothing coefficient to obtain smooth propeller speed parameters.

[0006] Furthermore, the acquisition of optical imaging data, sonar detection data, and water quality parameter data includes: Acquire camera intrinsic parameters and calibration error pixels; obtain optical imaging data based on camera intrinsic parameters and calibration error pixels; obtain sonar detection data based on preset beamwidth and preset sampling frequency; obtain scattered light intensity based on preset scattering light principle; acquire current detection water depth and pH value sensor data; analyze scattered light intensity, pH value sensor data and current detection water depth to obtain second analysis results; generate water quality parameter data based on the second analysis results.

[0007] Furthermore, the analysis of optical imaging data, sonar detection data, and water quality parameter data based on a preset multimodal convolutional neural network model and a preset Poisson surface reconstruction algorithm to obtain a first analysis result includes: acquiring historical sonar echo data; enhancing the sonar detection data based on the historical sonar echo data and preset Gaussian white noise data to obtain enhanced echo data; calculating the three-dimensional volume of the detection area based on the enhanced echo data, a preset trigonometric function, and a preset sound velocity value; and analyzing the three-dimensional volume of the detection area, water quality parameter data, and optical imaging data based on the multimodal convolutional neural network model and the Poisson surface reconstruction algorithm to obtain the first analysis result.

[0008] Furthermore, the sonar detection data is enhanced based on historical sonar echo data and preset Gaussian white noise data to obtain enhanced echo data. This includes: performing feature analysis on historical sonar echo data to obtain source-level features, target intensity features, and noise level features; constructing an echo intensity equation based on preset propagation loss values, source-level features, target intensity features, and noise level features; obtaining multi-beam echo data from the sonar detection data; and enhancing the multi-beam echo data based on the echo intensity equation and preset Gaussian white noise data to obtain enhanced echo data.

[0009] Further, the step of calculating the three-dimensional volume of the detection area based on the enhanced echo data, a preset trigonometric function, and a preset sound velocity value includes: obtaining the echo duration from the enhanced echo data; calculating the length of the detection area based on the echo duration and the sound velocity value; obtaining the maximum sonar beamwidth, the maximum vertical beam angle, and the sonar signal propagation distance from the enhanced echo data; calculating the horizontal width and vertical width of the detection area based on the trigonometric function, the maximum sonar beamwidth, the maximum vertical beam angle, and the sonar signal propagation distance; and calculating the three-dimensional volume of the detection area based on the length, horizontal width, and vertical width of the detection area.

[0010] Furthermore, the analysis of the three-dimensional volume of the detection area, water quality parameter data, and optical imaging data based on the multimodal convolutional neural network model and Poisson surface reconstruction algorithm to obtain a first analysis result includes: fusing optical imaging data, sonar detection data, and water quality parameter data based on the multimodal convolutional neural network model to obtain fused multi-source data; analyzing the three-dimensional volume of the detection area and optical imaging data according to a preset edge computing unit and a preset three-dimensional coordinate system to obtain a third analysis result; generating a rotation matrix and translation vector based on the third analysis result; and analyzing the fused multi-source data based on the Poisson surface reconstruction algorithm, rotation matrix, and translation vector to obtain the first analysis result.

[0011] Furthermore, the analysis of the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density includes: obtaining the number of sampling areas, the dataset of sampling area, and the set of single sampling times; calculating the relative abundance index based on the number of sampling areas, the dataset of sampling area, and the set of single sampling times; and analyzing the three-dimensional distribution map of seafood resources based on the relative abundance index to obtain a dataset of seafood quantity and population density.

[0012] Furthermore, a device for measuring the quantity of seafood in flowing water includes: a first data acquisition module for acquiring hull attitude sensing data and water area sensing parameter data; a parameter generation module for generating propeller smooth rotation speed parameters based on preset fluid dynamics principles, preset shape parameters, hull attitude sensing data, and water area sensing data; a route optimization module for acquiring a navigation route and adjusting the navigation route according to the propeller smooth rotation speed parameters to obtain an optimized navigation route; a navigation state acquisition module for acquiring the navigation state based on the optimized navigation route; and a first analysis module for analyzing the navigation state to obtain... The system comprises three modules: a navigation status analysis module and a second data acquisition module, used to acquire optical imaging data, sonar detection data, and water quality parameter data when the navigation status analysis result indicates that the ship is located in the detection area; a second analysis module, used to analyze the optical imaging data, sonar detection data, and water quality parameter data based on a preset multimodal convolutional neural network model and a preset Poisson surface reconstruction algorithm to obtain a first analysis result; a map generation module, used to generate a three-dimensional distribution map of seafood resources based on the first analysis result; and a third analysis module, used to analyze the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density.

[0013] Furthermore, a flowing water seafood quantity measuring device includes: a memory and at least one processor, wherein the memory stores instructions; at least one processor invokes the instructions in the memory to cause the flowing water seafood quantity measuring device to perform the various steps of a flowing water seafood quantity measuring method as described above.

[0014] The beneficial effects of the present invention's method for measuring the quantity of seafood in flowing water are as follows: In terms of navigation control, the propeller speed is dynamically adjusted based on fluid dynamics principles, and the navigation route is optimized in conjunction with the speed parameters. This effectively reduces navigation resistance and energy consumption, minimizes vibration problems caused by improper speed, and improves the navigation efficiency and safety of ships in complex water environments, avoiding energy waste and time loss due to unreasonable route planning. Regarding resource monitoring, multi-source data such as optical imaging, sonar detection, and water quality parameters are used. Multimodal convolutional neural networks and Poisson surface reconstruction algorithms are employed to deeply mine data features, constructing a three-dimensional distribution map of seafood resources. This improves the accuracy of quantity estimation and population density analysis, providing a reliable basis for fishery resource assessment and aquaculture planning. Furthermore, this solution organically integrates navigation control and resource monitoring, forming a closed-loop optimization mechanism: navigation optimization ensures the efficient implementation of resource monitoring, while monitoring data in turn feeds back into navigation strategy adjustments, achieving synergistic effects between data collection and ship operation. This powerfully promotes the scientific and efficient development and management of aquatic resources. Attached Figure Description

[0015] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a first flowchart of a method for measuring the quantity of seafood in flowing water, provided by an embodiment of the present invention. Figure 2 This is a second flowchart of a method for measuring the quantity of seafood in flowing water, provided by an embodiment of the present invention. Figure 3 This is a third flowchart of a method for measuring the quantity of seafood in flowing water, provided by an embodiment of the present invention. Figure 4 This is a fourth flowchart of a method for measuring the quantity of seafood in flowing water, provided by an embodiment of the present invention. Figure 5 A fifth flowchart of a method for measuring the quantity of seafood in flowing water provided in an embodiment of the present invention; Figure 6 A sixth flowchart of a method for measuring the quantity of seafood in flowing water provided in an embodiment of the present invention; Figure 7 The seventh flowchart of a method for measuring the quantity of seafood in flowing water provided in an embodiment of the present invention; Figure 8 The eighth flowchart of a method for measuring the quantity of seafood in flowing water provided in an embodiment of the present invention; Figure 9 This is a schematic diagram of a seafood quantity measuring device in flowing water provided in an embodiment of the present invention; Figure 10 This is a schematic diagram of a seafood quantity measuring device in flowing water, provided as an embodiment of the present invention. Detailed Implementation

[0016] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 An embodiment of a method for measuring the quantity of seafood in flowing water according to the present invention includes: 101. Acquire ship attitude sensor data and water area sensor parameter data; In this embodiment, attitude sensors (such as gyroscopes and accelerometers) installed on the ship are used to acquire real-time ship attitude sensing data such as pitch angle, roll angle, and yaw angle, and water area sensing parameters such as water flow speed and direction are acquired using water area sensors (such as current sensors, tide gauges, and wave sensors) deployed on the ship. 102. Generate smooth propeller speed parameters based on preset fluid dynamics principles, preset shape parameters, hull attitude sensing data, and water area sensing data. In this embodiment, the propeller speed is dynamically adjusted by combining environmental factors with the ship's own parameters through the principles of fluid mechanics. This ensures stable navigation of the ship, reduces energy consumption, and avoids ship vibration or low navigation efficiency caused by unreasonable speed. 103. Obtain the navigation route and adjust it according to the propeller smooth speed parameters to obtain an optimized navigation route; In this embodiment, the coordination between the power system and the navigation path is considered, enabling the ship to travel along a better path in complex water environments, thereby improving navigation efficiency and safety. 104. Obtain navigation status based on the optimized navigation route; 105. Analyze the navigation status to obtain navigation status analysis results; 106. When the navigation status analysis result indicates that the hull is located in the detection area, acquire optical imaging data, sonar detection data, and water quality parameter data. 107. Based on a preset multimodal convolutional neural network model and a preset Poisson surface reconstruction algorithm, optical imaging data, sonar detection data and water quality parameter data are analyzed to obtain the first analysis result; 108. Generate a three-dimensional distribution map of seafood resources based on the results of the first analysis; 109. Analyze the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density; In this embodiment, the multimodal convolutional neural network can fully extract features from different data, while the Poisson surface reconstruction algorithm is used to construct a three-dimensional spatial structure. The combination of the two can accurately analyze the distribution of seafood resources in the detection area, and finally generate a three-dimensional distribution map of seafood resources. Furthermore, the quantity and population density datasets are obtained through analysis, providing reliable data support for fishery resource assessment and aquaculture planning. In this embodiment, regarding navigation control, the propeller speed is dynamically adjusted based on fluid dynamics principles, and the navigation route is optimized in conjunction with the speed parameters. This effectively reduces navigation resistance and energy consumption, minimizes vibration problems caused by improper speed, and improves the ship's navigation efficiency and safety in complex water environments, avoiding energy waste and time loss due to unreasonable route planning. In terms of resource monitoring, multi-source data such as optical imaging, sonar detection, and water quality parameters are used. Multimodal convolutional neural networks and Poisson surface reconstruction algorithms are employed to deeply mine data features, constructing a three-dimensional distribution map of seafood resources. This improves the accuracy of quantity estimation and population density analysis, providing a reliable basis for fishery resource assessment and aquaculture planning. Furthermore, this solution organically integrates navigation control and resource monitoring, forming a closed-loop optimization mechanism: navigation optimization ensures the efficient implementation of resource monitoring, while monitoring data in turn feeds back into navigation strategy adjustments, achieving synergistic effects between data collection and ship operation. This powerfully promotes the scientific and efficient development and management of aquatic resources.

[0019] Please see Figure 2 A second embodiment of a method for measuring the quantity of seafood in flowing water, as described in this invention, includes: 201. Generate the drag coefficient based on fluid mechanics principles and shape parameters; In this embodiment, the drag coefficient is calculated using fluid dynamics principles and hull shape parameters (such as hull type and draft). This drag coefficient reflects the magnitude of the resistance experienced by the ship when it is sailing in water and is the basis for subsequent dynamic calculations. 202. Generate the optimal counterweight based on the preset fluid dynamics model, drag coefficient, hull attitude sensor data, water area sensor data, and preset target anchoring coordinates; In this embodiment, by adjusting the balance between the ship's center of gravity and hydrodynamics according to the optimal ballast, attitude fluctuations caused by wind, waves, or uneven load are reduced, thereby improving navigation stability. 203. Generate propeller speed parameters based on the optimal counterweight and the preset propeller thrust coefficient; In this embodiment, the propeller thrust coefficient (reflecting propeller performance) couples the ship's ballast with the ship's power system to calculate the required propeller speed, ensuring that the ship generates sufficient thrust to overcome resistance and reach the target position. 204. Smooth the propeller speed parameters according to the preset smoothing curve and preset smoothing coefficient to obtain the smoothed propeller speed parameters. In this embodiment, a smooth curve (such as an S-curve) and a smoothing coefficient are used to filter the propeller speed parameters to avoid mechanical shock or increased energy consumption caused by sudden changes, and to reduce hull vibration caused by speed fluctuations. In this embodiment, the drag coefficient is calculated based on fluid mechanics and hull parameters, quantifying navigation resistance into an operable parameter index, thus laying a data foundation for ship power control. Secondly, by integrating the drag coefficient, hull attitude data, and water area sensor data (water flow velocity, wave level), an optimal ballast scheme is dynamically generated, effectively balancing the ship's center of gravity and hydrodynamics, significantly reducing attitude fluctuations caused by wind, waves, and load changes, and ensuring stable operation of the measurement equipment. By optimizing propeller speed parameters, the ship is ensured to navigate to the target detection area with optimal efficiency. Simultaneously, the mechanical shock and energy waste caused by sudden speed changes during navigation are reduced, and the interference of hull vibration on the accuracy of quantity and population density measurements in the detection area is minimized. This scheme achieves deep synergy between ship navigation control and quantity measurement, improving the stability and accuracy of the measurement process, and providing reliable support for fishery resource surveys and ecological assessments in flowing water areas.

[0020] Please see Figure 3 A third embodiment of a method for measuring the quantity of seafood in flowing water, as described in this invention, includes: 301. Obtain camera intrinsic parameters and calibration error pixels; In this embodiment, by calibrating the camera's internal parameters (such as focal length and principal point coordinates) and calibration errors (distortion coefficients), the geometric deviation of optical imaging is corrected to ensure the accuracy of the acquired image data; 302. Obtain optical imaging data based on camera intrinsic parameters and calibration error pixels; In this embodiment, based on the calibrated parameters, the original image is processed by distortion correction, perspective transformation and other methods to obtain high-quality optical imaging data that can be used for target recognition; 303. Acquire sonar detection data based on the preset beamwidth and preset sampling frequency; In this embodiment, based on the beamwidth (which determines the detection angle range) and sampling frequency (which controls the number of data acquisitions), the sonar equipment emits sound waves and receives reflected signals to generate sonar detection data containing spatial information such as underwater topography and object distribution. 304. Obtain the intensity of scattered light based on the preset principle of scattered light; 305. Acquire sensor data on the current depth and pH value of the detected water area; 306. Analyze the scattered light intensity, pH value sensor data, and current depth of the detected water area to obtain the second analysis result; 307. Generate water quality parameter data based on the results of the second analysis; In this embodiment, based on the principle of light scattering (such as the positive correlation between turbidity and the intensity of scattered light), the degree of light scattering by suspended particles in the water is measured by an optical sensor to quantify the turbidity of the water. The depth of the current water area (affected by environmental factors such as water pressure and light) and pH value sensor data are combined to perform cross-analysis of depth data and output core water quality parameters such as dissolved oxygen, turbidity, and pH value to achieve a comprehensive assessment of the water environment. In this embodiment, by calibrating camera intrinsic parameters and adjusting error pixels, geometric deviations in optical imaging are effectively corrected, generating high-quality optical data and providing a clear image foundation for fish target identification. Sonar detection data is acquired using preset beamwidth and sampling frequency to achieve high-precision spatial mapping of underwater topography and object distribution. Water turbidity is quantified based on the principle of scattered light, and cross-analysis is performed by integrating water depth and pH data to output core water quality parameters such as dissolved oxygen and turbidity, enabling a comprehensive assessment of the physical and chemical environment of the water body. The synergistic complementarity of multi-source data ensures the accuracy of optical and sonar data in measurement and provides comprehensive and scientific data support for fishery resource surveys, water pollution monitoring, and ecological restoration by correlating water quality parameters with ecological environmental factors.

[0021] Please see Figure 4 The fourth embodiment of a method for measuring the quantity of seafood in flowing water according to the present invention includes: 401. Obtain historical sonar echo data; 402. Enhance the sonar detection data based on historical sonar echo data and preset Gaussian white noise data to obtain enhanced echo data; In this embodiment, historical sonar echo data and Gaussian white noise data (simulating random noise) are mixed to enhance the sonar detection data, simulating the complex noise scene in the real water environment, making the sonar detection data more robust and effectively reducing the false detection rate (such as distinguishing fish echoes from rock and debris echoes). 403. The three-dimensional volume of the detection area is calculated based on the enhanced echo data, the preset trigonometric function, and the preset sound velocity value; In this embodiment, by combining enhanced echo data with 3D modeling, the dynamic spatial distribution of other seafood populations (such as vertical migration paths) can be obtained by capturing fish schools in flowing water, providing a reliable spatial reference for quantity calculation; 404. Based on a multimodal convolutional neural network model and a Poisson surface reconstruction algorithm, the three-dimensional volume, water quality parameter data, and optical imaging data of the detection area are analyzed to obtain the first analysis result; In this embodiment, a three-dimensional distribution model of seafood resources can be generated by using a multimodal convolutional neural network model to extract visual features (such as fish shape and quantity) from optical imaging data and spatial features (such as fish distribution density) from sonar volume. This model is then fitted with discrete point cloud data (sonar detection points and optical identification points) into a continuous surface, enabling dynamic tracking of seafood resources (such as fish migration path analysis). In this embodiment, historical sonar echo data is mixed with Gaussian white noise to enhance real-time sonar detection data, simulating complex noise scenarios in real waters. This effectively suppresses environmental interference (such as water flow and ship vibration clutter), improves the signal-to-noise ratio of sonar data, reduces the false detection rate of fish echoes and rocks / debris, and enhances the model's ability to detect weak targets. By combining enhanced echo data, trigonometric functions, and sound velocity values ​​to construct a three-dimensional volume model of the detection area, the dynamic spatial distribution of seafood populations such as fish schools can be captured, providing a reliable spatial benchmark for quantification. By fusing optical imaging visual features (fish shape and quantity) with sonar volume spatial features (distribution density) through a multimodal convolutional neural network, and combining the Poisson surface reconstruction algorithm to fit discrete point clouds, a three-dimensional distribution model of seafood resources is obtained. This enables dynamic information tracking such as fish migration paths, and simultaneously correlates water quality parameters to analyze the impact of the environment on quantity, providing multi-dimensional scientific data support for fishery resource management, ecological protection, and water restoration.

[0022] Please see Figure 5 The fifth embodiment of a method for measuring the quantity of seafood in flowing water according to the present invention includes: 501. Perform feature analysis on historical sonar echo data to obtain source-level features, target intensity features, and noise level features; In this embodiment, the source-level features are used to reflect the intensity of the sonar emitted signal and determine the initial energy of the sound wave propagating in the water; the target intensity features are used to quantify the ability of a target object (such as fish or reefs) to reflect sound waves and to distinguish different target types; the noise level features are used to statistically analyze the intensity distribution of environmental noise (water flow, ship vibration, background clutter) and identify interference signals; the above three types of features are extracted from historical sonar echo data through feature analysis methods such as spectrum analysis and time-domain statistics, providing a data foundation for subsequent modeling. 502. Based on the preset propagation loss value, sound source level characteristics, target intensity characteristics, and noise level characteristics, the echo intensity equation is constructed. In this embodiment, the expression for the echo intensity equation is as follows: In the formula, To propagate the loss value, The source level coefficient, For the target intensity coefficient, Noise level factor; The echo coefficient is to be solved; 503. Obtain multi-beam echo data from sonar detection data; 504. Enhance the multi-beam echo data according to the echo intensity equation and the preset Gaussian white noise data to obtain enhanced echo data; In this embodiment, the enhanced processing of multi-beam echo data fully utilizes the advantages of sonar multi-beam detection to achieve three-dimensional positioning and identification of underwater targets, which is especially suitable for tracking dynamic targets (such as fish migration) in flowing water. In this embodiment, by extracting three types of features—sound source level, target intensity, and noise level—the energy, target reflection, and environmental interference information of the sonar echo are deconstructed, laying a data foundation for subsequent modeling. Based on the feature-constructed echo intensity equation, the key physical quantity relationships in the sound wave propagation process are scientifically quantified, making the echo signal analysis more consistent with the actual acoustic environment. By combining Gaussian white noise data to enhance the multi-beam echo, the advantages of sonar multi-beam detection are fully utilized to achieve three-dimensional positioning and dynamic tracking of underwater targets, effectively improving the ability to identify dynamic targets such as fish schools in flowing water, and providing data support for fishery resource monitoring, underwater topographic mapping, and other applications.

[0023] Please see Figure 6 The sixth embodiment of a method for measuring the quantity of seafood in flowing water according to the present invention includes: 601. Obtain the echo duration from the enhanced echo data; 602. The length of the detection area is calculated based on the echo duration and sound velocity. 603. Obtain the maximum sonar beamwidth, maximum vertical beam angle, and sonar signal propagation distance from the enhanced echo data; 604. The horizontal width and vertical width of the detection area are calculated based on trigonometric functions, maximum sonar beamwidth, maximum vertical beamwidth, and sonar signal propagation distance. In this embodiment, based on trigonometric functions (such as sine and cosine functions), combined with the half-angle of the maximum beamwidth (θ / 2) and the sonar signal propagation distance (d), the horizontal width of the detection area is calculated using the formula: 2 × d × sin(θ / 2), considering the horizontal coverage area formed by beam spread; combined with the half-angle of the maximum vertical beamwidth (θ / 2), the horizontal coverage area of ​​the detection area is calculated using the formula: 2 × d × sin(θ / 2). The horizontal width of the detection area is calculated using the formula: 2 × d × sin( ) Calculate, taking into account the vertical coverage area formed by beam spread; 605. The three-dimensional volume of the detection area is calculated based on the length, horizontal width, and vertical width of the detection area. In this embodiment, the three-dimensional volume of the detection area is calculated by measuring the length, horizontal width, and vertical width of the detection area, thus restoring the scale of the underwater detection space. This solution is suitable for dynamic volume monitoring of targets such as fish gathering areas and underwater obstacles in flowing waters. It can also provide spatial data support for fields such as fishery resource assessment, underwater engineering planning, and ecological environment research, significantly improving the scientific nature and efficiency of marine exploration and management.

[0024] Please see Figure 7 The seventh embodiment of a method for measuring the quantity of seafood in flowing water according to the present invention includes: 701. Based on a multimodal convolutional neural network model, optical imaging data, sonar detection data, and water quality parameter data are fused to obtain fused multi-source data; In this embodiment, a multimodal convolutional neural network (CNN) is used to extract features from different modal data (optical imaging data (visual features of seafood), sonar detection data (underwater spatial structure, population distribution density in sonar data) and water quality parameter data (environmental physicochemical indicators)) through parallel branches. Then, cross-modal information complementarity is achieved through a fusion layer to output fused multi-source data and enhance data representation capabilities. 702. Analyze the three-dimensional volume and optical imaging data of the detection area based on the preset edge calculation unit and the preset three-dimensional coordinate system to obtain the third analysis result; 703. Generate the rotation matrix and translation vector based on the third analysis result; In this embodiment, the target position in the optical image is mapped to three-dimensional space through coordinate system transformation, and the coordinates of the target in the real environment are calculated. At the same time, the spatial topological relationship in the three-dimensional volume data is analyzed, and a third analysis result (such as target spatial position and attitude parameters) is output. Based on the third analysis result, the rotation matrix (describing the target attitude change) and the translation vector (describing the target spatial displacement) are calculated through geometric transformation (such as least squares method and ICP registration algorithm), so as to achieve the alignment of different modal data in a unified coordinate system. 704. Based on the Poisson surface reconstruction algorithm, rotation matrix and translation vector, the fused multi-source data are analyzed to obtain the first analysis result; In this embodiment, the Poisson surface reconstruction algorithm is used to fit the discrete point cloud (optical identification points, sonar detection points) in the fused multi-source data into a continuous surface, and spatial calibration is performed by combining the rotation matrix and translation vector to generate a high-precision three-dimensional distribution model of seafood resources, and output the first analysis results (such as the three-dimensional distribution of seafood population and the estimated quantity). In this embodiment, a multimodal convolutional neural network is used to extract features from optical, sonar, and water quality data in parallel, breaking the limitations of single data sources. A fusion layer enables complementary information from visual features, spatial structure, and environmental parameters, enhancing the data's ability to represent seafood resources. Using edge computing units and a three-dimensional coordinate system, optical targets are mapped to real space. Combined with three-dimensional volume analysis of the detection area, rotation matrices and translation vectors are generated, aligning the spatial coordinates of multi-source data. Finally, based on the Poisson surface reconstruction algorithm, discrete point clouds are fitted into a high-precision three-dimensional distribution model, intuitively presenting the distribution and quantity estimates of seafood populations. This solution effectively integrates environment and data, providing scientific quantitative basis for fishery resource assessment and ecological restoration planning, and promoting the upgrade of water monitoring from single-indicator analysis to multi-dimensional intelligent decision-making.

[0025] Please see Figure 8 The eighth embodiment of a method for measuring the quantity of seafood in flowing water according to the present invention includes: 801. Obtain the number of sampling regions, the dataset of sampling areas, and the set of single sampling times; In this embodiment, multiple sampling areas can be obtained by dividing the detection area into water areas, and the area size of each sampling area and the time taken to complete data collection for each sampling area are recorded. 802. The relative abundance index is calculated based on the number of sampling regions, the sampling area dataset, and the set of single sampling times. In this embodiment, the relative abundance index takes into account the sampling range, area and time cost, and reflects the resource sampling efficiency per unit area per unit time. The higher the index, the greater the relative resource density. 803. Analyze the three-dimensional distribution map of seafood resources based on the relative abundance index to obtain a dataset of seafood quantity and population density; In this embodiment, the target region in the map is weighted by the relative abundance index, and the population density (unit: individuals / square meter) is further calculated to finally generate a dataset of total seafood quantity and population density. In this embodiment, by acquiring data from multiple sampling areas, the system records the area and sampling time to construct a comprehensive basic dataset. An innovative relative abundance index is introduced, incorporating sampling range, area, and time cost into a unified calculation framework. This effectively reflects the resource sampling efficiency per unit time and space, providing a standardized indicator for resource density assessment. Using this relative abundance index as weight, a weighted analysis of the three-dimensional distribution map of seafood resources is performed to calculate the quantity in each region. Combined with the sampling area, population density is calculated, achieving a quantitative assessment from local to overall levels. This solution is applicable to different aquatic environments, dynamically tracks changes in resource distribution, and provides reliable data support for fisheries planning, aquaculture density optimization, and the formulation of ecological protection strategies, promoting the scientific and refined development of aquatic resource management.

[0026] The above describes a method for measuring the quantity of seafood in flowing water according to an embodiment of the present invention. The following describes a device for measuring the quantity of seafood in flowing water according to an embodiment of the present invention. Please refer to [link / reference]. Figure 9 One embodiment of the present invention provides a device for measuring the quantity of seafood in flowing water, comprising: The first data acquisition module 1 is used to acquire ship attitude sensing data and water area sensing parameter data; The parameter generation module 2 is used to generate smooth propeller speed parameters based on preset fluid dynamics principles, preset shape parameters, hull attitude sensing data and water area sensing data. The route optimization module 3 is used to obtain the flight route and adjust the flight route according to the propeller smooth speed parameters to obtain an optimized flight route. Navigation status acquisition module 4 is used to acquire navigation status based on the optimized navigation route; The first analysis module 5 is used to analyze the navigation status in order to obtain the navigation status analysis results; The second data acquisition module 6 is used to acquire optical imaging data, sonar detection data and water quality parameter data when the navigation status analysis result indicates that the ship is located in the detection area. The second analysis module 7 is used to analyze optical imaging data, sonar detection data and water quality parameter data based on a preset multimodal convolutional neural network model and a preset Poisson surface reconstruction algorithm to obtain the first analysis result; Map generation module 8 is used to generate a three-dimensional distribution map of seafood resources based on the first analysis results; The third analysis module 9 is used to analyze the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density. In this embodiment, regarding navigation control, the propeller speed is dynamically adjusted based on fluid dynamics principles, and the navigation route is optimized in conjunction with the speed parameters. This effectively reduces navigation resistance and energy consumption, minimizes vibration problems caused by improper speed, and improves the ship's navigation efficiency and safety in complex water environments, avoiding energy waste and time loss due to unreasonable route planning. In terms of resource monitoring, multi-source data such as optical imaging, sonar detection, and water quality parameters are used. Multimodal convolutional neural networks and Poisson surface reconstruction algorithms are employed to deeply mine data features, constructing a three-dimensional distribution map of seafood resources. This improves the accuracy of quantity estimation and population density analysis, providing a reliable basis for fishery resource assessment and aquaculture planning. Furthermore, this solution organically integrates navigation control and resource monitoring, forming a closed-loop optimization mechanism: navigation optimization ensures the efficient implementation of resource monitoring, while monitoring data in turn feeds back into navigation strategy adjustments, achieving synergistic effects between data collection and ship operation. This powerfully promotes the scientific and efficient development and management of aquatic resources.

[0027] Figure 10 This is a schematic diagram of the structure of a flowing water seafood quantity measuring device 900 provided in an embodiment of the present invention. This flowing water seafood quantity measuring device 900 can vary considerably due to different configurations or performance. It may include one or more central processing units (CPUs) 910 (e.g., one or more processors) and a memory 920, and one or more media 930 (e.g., one or more mass storage devices) storing application programs 933 or data 932. The memory 920 and media 930 can be temporary or persistent storage. The program stored in the media 930 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the flowing water seafood quantity measuring device 900. Furthermore, the processor 910 may be configured to communicate with the media 930 and execute a series of instruction operations in the media 930 on the flowing water seafood quantity measuring device 900 to implement the steps of the flowing water seafood quantity measuring method provided in the above-described method embodiments.

[0028] A seafood quantity measuring device 900 for flowing water may further include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 10The illustrated structure of a seafood quantity measuring device in flowing water does not constitute a limitation on a seafood quantity measuring device 900 in flowing water. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

Claims

1. A method for measuring the quantity of seafood in flowing water, characterized in that, include: Acquire ship attitude sensing data and water area sensing parameter data; Based on preset fluid dynamics principles, preset shape parameters, hull attitude sensing data, and water area sensing data, smooth propeller speed parameters are generated. The navigation route is obtained and adjusted according to the propeller smoothing speed parameters to obtain an optimized navigation route; Obtain navigation status based on the optimized navigation route; The navigation status is analyzed to obtain the navigation status analysis results; When the navigation status analysis result indicates that the hull is located in the detection area, optical imaging data, sonar detection data, and water quality parameter data are acquired. Based on a pre-defined multimodal convolutional neural network model and a pre-defined Poisson surface reconstruction algorithm, optical imaging data, sonar detection data, and water quality parameter data are analyzed to obtain the first analysis result. A three-dimensional distribution map of seafood resources was generated based on the first analysis results; We analyzed the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density.

2. The method for measuring the quantity of seafood in flowing water as described in claim 1, characterized in that, The process of generating smooth propeller speed parameters based on preset fluid dynamics principles, preset shape parameters, hull attitude sensor data, and water area sensor data includes: The drag coefficient is generated based on fluid mechanics principles and shape parameters; The optimal counterweight is generated based on the preset fluid dynamics model, drag coefficient, hull attitude sensor data, water area sensor data, and preset target anchoring coordinates. Propeller speed parameters are generated based on the optimal counterweight and the preset propeller thrust coefficient; The propeller speed parameters are smoothed according to the preset smoothing curve and preset smoothing coefficient to obtain the smoothed propeller speed parameters.

3. The method for measuring the quantity of seafood in flowing water as described in claim 1, characterized in that, The acquisition of optical imaging data, sonar detection data, and water quality parameter data includes: Obtain camera intrinsic parameters and calibration error pixels; Optical imaging data is obtained based on camera intrinsic parameters and calibration error pixels; Sonar detection data is obtained based on the preset beamwidth and preset sampling frequency; The intensity of the scattered light is obtained based on the preset principle of light scattering; Acquire sensor data on the current depth and pH value of the detected water area; The intensity of scattered light, pH value sensor data, and current depth of the detected water area are analyzed to obtain the second analysis result; Water quality parameter data are generated based on the results of the second analysis.

4. The method for measuring the quantity of seafood in flowing water as described in claim 1, characterized in that, The method, based on a preset multimodal convolutional neural network model and a preset Poisson surface reconstruction algorithm, analyzes optical imaging data, sonar detection data, and water quality parameter data to obtain a first analysis result, including: Acquire historical sonar echo data; The sonar detection data is enhanced based on historical sonar echo data and preset Gaussian white noise data to obtain enhanced echo data. The three-dimensional volume of the detection area is calculated based on the enhanced echo data, preset trigonometric functions, and preset sound velocity values. The three-dimensional volume, water quality parameter data, and optical imaging data of the detection area are analyzed based on a multimodal convolutional neural network model and a Poisson surface reconstruction algorithm to obtain the first analysis result.

5. The method for measuring the quantity of seafood in flowing water as described in claim 4, characterized in that, The sonar detection data is enhanced based on historical sonar echo data and preset Gaussian white noise data to obtain enhanced echo data, including: Feature analysis was performed on historical sonar echo data to obtain source-level features, target intensity features, and noise level features; The echo intensity equation is constructed based on the preset propagation loss value, sound source level characteristics, target intensity characteristics, and noise level characteristics. Multi-beam echo data were obtained from sonar detection data; The multi-beam echo data is enhanced based on the echo intensity equation and preset Gaussian white noise data to obtain enhanced echo data.

6. The method for measuring the quantity of seafood in flowing water as described in claim 4, characterized in that, The calculation of the three-dimensional volume of the detection area based on enhanced echo data, preset trigonometric functions, and preset sound velocity values ​​includes: The echo duration is obtained from the enhanced echo data; The length of the detection area is calculated based on the echo duration and sound velocity. The maximum sonar beamwidth, maximum vertical beam angle, and sonar signal propagation distance are obtained from the enhanced echo data. The horizontal and vertical widths of the detection area are calculated based on trigonometric functions, the maximum sonar beamwidth, the maximum vertical beam half-angle, and the sonar signal propagation distance. The three-dimensional volume of the detection area is calculated based on the length, horizontal width, and vertical width of the detection area.

7. The method for measuring the quantity of seafood in flowing water as described in claim 4, characterized in that, The analysis of the three-dimensional volume, water quality parameter data, and optical imaging data of the detection area based on the multimodal convolutional neural network model and Poisson surface reconstruction algorithm yields the first analysis result, including: A multimodal convolutional neural network model is used to fuse optical imaging data, sonar detection data, and water quality parameter data to obtain fused multi-source data. The three-dimensional volume and optical imaging data of the detection area are analyzed based on the preset edge computing unit and the preset three-dimensional coordinate system to obtain the third analysis result; Generate rotation matrices and translation vectors based on the third analysis results; The first analysis result is obtained by analyzing the fused multi-source data based on the Poisson surface reconstruction algorithm, rotation matrix and translation vector.

8. The method for measuring the quantity of seafood in flowing water as described in claim 1, characterized in that, The analysis of the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density includes: Obtain the number of sampling regions, the dataset of sampling areas, and the set of single sampling times; The relative abundance index is calculated based on the number of sampling regions, the sampling area dataset, and the set of single sampling times. The three-dimensional distribution map of seafood resources was analyzed based on the relative abundance index to obtain a dataset of seafood quantity and population density.

9. A device for measuring the quantity of seafood in flowing water, characterized in that, include: The first data acquisition module is used to acquire ship attitude sensing data and water area sensing parameter data. The parameter generation module is used to generate smooth propeller speed parameters based on preset fluid dynamics principles, preset shape parameters, hull attitude sensing data, and water area sensing data. The route optimization module is used to obtain the flight route and adjust the flight route according to the propeller smooth speed parameters to obtain an optimized flight route; The navigation status acquisition module is used to acquire navigation status based on the optimized navigation route; The first analysis module is used to analyze the navigation status in order to obtain the navigation status analysis results; The second data acquisition module is used to acquire optical imaging data, sonar detection data, and water quality parameter data when the navigation status analysis result indicates that the ship is located in the detection area. The second analysis module is used to analyze optical imaging data, sonar detection data, and water quality parameter data based on a preset multimodal convolutional neural network model and a preset Poisson surface reconstruction algorithm to obtain the first analysis result. The map generation module is used to generate a three-dimensional distribution map of seafood resources based on the first analysis results. The third analysis module is used to analyze the three-dimensional distribution map of seafood resources to obtain a dataset of seafood quantity and population density.

10. A device for measuring the quantity of seafood in flowing water, characterized in that, include: A memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the flowing water seafood quantity measuring device to perform the steps of the flowing water seafood quantity measuring method as claimed in any one of claims 1-7.

Images