A three-dimensional monitoring method, system and device for a marine ranch

Abstract

The present invention discloses a three-dimensional monitoring method, system and device for a marine ranch, including a plurality of water body sensors, underwater cameras and a data collection unit, wherein the data collection unit is used to receive the detection data and underwater image information of each marine ranch and send them out. By setting a plurality of water body sensors, the present invention can respectively collect the detection data of the marine ranch from multiple different positions, and use the underwater cameras to capture the underwater image information. The data collection unit is used to collect and send out the detection data and underwater image information of the marine ranch, so as to provide data support for the monitoring and analysis of the marine ranch; since the water body sensors and underwater cameras can be distributed in the marine ranch, data can be collected over a large range for real-time analysis, and the water body sensors and underwater cameras can be set at fixed points without the need for operations such as patrol observation, thus saving costs. The present invention is widely applied to the field of marine information technology.

Description

Technical Field

[0001] The present invention relates to the field of marine information technology, and in particular to a three-dimensional monitoring method, system and device for a marine ranch. Background Art

[0002] A "marine ranch" refers to a certain sea area where large-scale fishery facilities and a systematic management system are adopted. Utilizing the natural marine ecological environment, economically important marine organisms released artificially are gathered. Just like grazing cattle and sheep on land, the marine resources such as fish, shrimps, shellfish, and algae are stocked at sea in a planned and purposeful manner.

[0003] In a marine ranch, the conditions such as water quality, meteorology, and hydrology directly affect the distribution and growth of aquatic organisms, and are the key factors determining the yield and quality of aquatic products in the marine ranch. Monitoring the marine ranch can timely understand and master the whole process of the marine ranch environment, resource quality and development trend. The various indicators and data obtained through monitoring provide a scientific basis for the management of the marine ranch. However, due to the large spatial scope of the marine ranch, the existing technologies for monitoring the marine ranch have disadvantages such as small coverage, high cost, and poor real-time performance. Summary of the Invention

[0004] Aiming at the technical problems such as small coverage, high cost, and poor real-time performance existing in the current monitoring technologies for marine ranches, the purpose of the present invention is to provide a three-dimensional monitoring method, system and device for a marine ranch.

[0005] On the one hand, an embodiment of the present invention includes a three-dimensional monitoring system for a marine ranch, and the three-dimensional monitoring system for a marine ranch includes:

[0006] A plurality of water body sensors; the water body sensors are used to be distributed at corresponding positions in the marine ranch to detect and obtain marine ranch detection data;

[0007] An underwater camera; the underwater camera is used to be arranged underwater in the marine ranch to capture underwater image information;

[0008] A data collection unit; the data collection unit is used to establish connections with each of the water body sensors and the underwater camera, receive each of the marine ranch detection data and the underwater image information, and externally send each of the marine ranch detection data and the underwater image information.

[0009] Further, the three-dimensional monitoring system for a marine ranch further includes:

[0010] A data processing unit; the data processing unit is used to receive each of the marine ranch detection data and the underwater image information, analyze and process each of the marine ranch detection data and the underwater image information, and obtain marine monitoring indicators.

[0011] Further, analyzing and processing the detection data of each of the marine pastures and the underwater image information to obtain marine monitoring indicators, including:

[0012] Fusing the detection data of each of the marine pastures to obtain fused data;

[0013] Identifying the underwater image information to obtain fish population quantity time series data;

[0014] Performing principal component analysis on the fused data and the fish population quantity time series data to obtain at least one of the marine monitoring indicators.

[0015] Further, analyzing and processing the detection data of each of the marine pastures and the underwater image information to obtain marine monitoring indicators further includes:

[0016] Determining a comprehensive environment index according to each of the marine monitoring indicators and their respective corresponding first weights.

[0017] Further, fusing the detection data of each of the marine pastures to obtain fused data includes:

[0018] Determining the respective corresponding second weights of each of the water body sensors;

[0019] For any one of the detection data of the marine pastures, using the second weight corresponding to the water body sensor that detected the detection data of the marine pastures as the weight corresponding to the detection data of the marine pastures, and performing weighted averaging on the detection data of each of the marine pastures to obtain the fused data.

[0020] Further, determining the respective corresponding second weights of each of the water body sensors includes:

[0021] Obtaining the spatial distances between each of the water body sensors and the underwater camera;

[0022] For any one of the water body sensors, setting the magnitude of the second weight corresponding to the water body sensor according to the spatial distance corresponding to the water body sensor; wherein, the magnitude of the second weight is negatively correlated with the spatial distance.

[0023] Further, determining the respective corresponding second weights of each of the water body sensors includes:

[0024] Determining a plurality of fish movement directions according to the fish population quantity time series data;

[0025] Performing divergence analysis on the fish population vector field to obtain divergence data at a plurality of positions within the field of view of the underwater camera; wherein, the fish population vector field is a vector field formed by each of the fish movement directions.

[0026] Determine the second weight corresponding to each of the water body sensors according to each of the divergence data.

[0027] Further, the determining the second weight corresponding to each of the water body sensors according to each of the divergence data includes:

[0028] Determine the convergence center and the divergence center within the field of view of the underwater camera according to each of the divergence data;

[0029] For any one of the water body sensors, when the distance between the water body sensor and the convergence center is less than the distance between the water body sensor and the divergence center and less than the distance threshold, determine the second weight corresponding to the water body sensor as a first value; when the distance between the water body sensor and the divergence center is less than the distance between the water body sensor and the convergence center and less than the distance threshold, determine the second weight corresponding to the water body sensor as a third value; otherwise, determine the second weight corresponding to the water body sensor as a second value;

[0030] Wherein, the first value is greater than the second value, and the second value is greater than the third value.

[0031] On the other hand, an embodiment of the present invention further includes a three-dimensional monitoring method for a marine ranch, and the three-dimensional monitoring method for a marine ranch includes:

[0032] Detect marine ranch detection data respectively through a plurality of water body sensors; each of the water body sensors is used to be distributed at corresponding positions of the marine ranch;

[0033] Obtain underwater image information by shooting with an underwater camera; the underwater camera is used to be arranged underwater in the marine ranch;

[0034] Receive each of the marine ranch detection data and the underwater image information through a data collection unit, and externally send each of the marine ranch detection data and the underwater image information; the data collection unit is used to establish connections with each of the water body sensors and the underwater camera;

[0035] Analyze and process each of the marine ranch detection data and the underwater image information to obtain marine monitoring indicators.

[0036] On the other hand, an embodiment of the present invention further includes a computer device, including a memory and a processor, the memory is used to store at least one program, and the processor is used to load at least one program to execute the three-dimensional monitoring method for a marine ranch in the embodiment.

[0037] The beneficial effects of the present invention are as follows: In the marine ranch three-dimensional monitoring system of the embodiment, by setting a plurality of water body sensors, the marine ranch detection data can be collected from multiple different positions of the marine ranch respectively, and the underwater image information can be obtained by using an underwater camera. The data collection unit is used to collect and externally send the marine ranch detection data and the underwater image information, so as to provide data support for the monitoring and analysis of the marine ranch; Since the water body sensors and the underwater camera can be distributed in the marine ranch, data can be collected over a large range for real-time analysis, and the water body sensors and the underwater camera can be set at fixed points without the need for operations such as patrol observation, thus saving costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a schematic structural diagram of the marine ranch three-dimensional monitoring system in the embodiment;

[0039] Figure 2 is a schematic layout diagram of the marine ranch three-dimensional monitoring system in the embodiment;

[0040] Figure 3 is a schematic principle diagram of the step of determining the respective second weights corresponding to each water body sensor in the embodiment;

[0041] Figure 4 is a schematic step diagram of the marine ranch three-dimensional monitoring method in the embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] In this embodiment, the structure of the marine ranch three-dimensional monitoring system is as Figure 1 shown, including an onshore base station, a power module, a data collection unit, an underwater camera, and a plurality of water body sensors.

[0043] Referring to Figure 2 , each water body sensor is distributed and set at the corresponding position of the marine ranch, such as the water surface, in the water, at the bottom, etc. The water body sensor detects the marine ranch detection data. Specifically, in this embodiment, there are different types of water body sensors such as type A, type B, and type C. Among them, the water body sensors belonging to type A are used to detect water quality detection data such as salinity, turbidity, and chlorophyll concentration, the water body sensors belonging to type B are used to detect hydrological detection data such as water temperature, pH value, and dissolved oxygen, and the water body sensors belonging to type C are used to detect meteorological detection data such as air temperature, air pressure, and wind speed. Therefore, in this embodiment, the marine ranch detection data includes water quality detection data, hydrological detection data, meteorological detection data, etc.

[0044] In this embodiment, there are multiple water body sensors of each type.

[0045] The underwater camera is used to be set at positions such as in the water or on the bottom of a marine ranch to capture underwater image information. The underwater image information can be single or multiple static images, or dynamic videos.

[0046] The power supply module can supply power to devices such as the data aggregation unit, the underwater camera, and the water body sensor.

[0047] In this embodiment, the data aggregation unit includes a single-chip microcomputer and a data sending end. Each water body sensor sends the detected marine ranch detection data (specifically, it may be water quality detection data, hydrological detection data, or meteorological detection data) to the single-chip microcomputer, and the underwater camera also sends the captured underwater image information to the single-chip microcomputer.

[0048] In this embodiment, each water body sensor can respectively perform preliminary processing on the collected marine ranch detection data. For example, the water body sensor of type A performs preliminary filtering and linearization processing on the collected water quality detection data to reduce noise and non-linear errors; the water body sensor of type B uses the Kalman Filter algorithm to smooth the collected hydrological detection data, thereby improving the stability and accuracy of the data; the water body sensor of type C compresses the collected meteorological detection data to reduce communication latency. Each water body sensor respectively sends out the preliminarily processed marine ranch detection data.

[0049] The single-chip microcomputer performs preprocessing on the received marine ranch detection data. For example, it converts the marine ranch detection data into binary and compresses the underwater image information, etc., and then sends the preprocessed data to the data sending end.

[0050] Specifically, the single-chip microcomputer transmits the processed marine ranch detection data and the underwater image information compressed by H.264 encoding to the data sending end through the UART (Universal Asynchronous Receiver-Transmitter) interface. The data sending end uses the CRC (Cyclic Redundancy Check) algorithm to generate a check code to ensure data integrity during the transmission process. Subsequently, the data is encapsulated in a UDP (User Datagram Protocol) data packet and is ready for modulation processing.

[0051] In this embodiment, the data sending end uses an underwater acoustic communication protocol for data transmission. Specifically, the data sending end uses MFSK (Multi-Frequency Shift Keying) as the modulation method to convert the digital-form data into an acoustic signal, effectively reducing the influence of multipath effects and frequency-selective fading; before signal modulation, the data sending end encodes the data through LDPC (Low-Density Parity-Check), significantly improving the bit error rate performance in underwater acoustic communication, so that the system can still work stably in an underwater environment with interference.

[0052] In this embodiment, the data sender converts the modulated electrical signal into an acoustic signal through a piezoelectric transducer and transmits the acoustic signal externally. In this way, the marine ranch detection data and underwater image information are transmitted in the form of acoustic signals.

[0053] Since the processes of preprocessing, modulation, encoding, and transducer conversion do not change the content of the information itself, such as the marine ranch detection data and underwater image information, in this embodiment, after the processes of preprocessing, modulation, encoding, and transducer conversion, the marine ranch detection data and underwater image information are still referred to as marine ranch detection data and underwater image information.

[0054] In this embodiment, since the acoustic signal propagates in water at a specific frequency, the data sender can adjust the transmission power and frequency to ensure that the signal can penetrate seawater during long-distance underwater transmission and maintain a high signal strength in a multipath propagation environment. To cope with the multipath propagation effect and delay spread in the underwater channel, the data sender can adopt an adaptive equalization algorithm to perform real-time channel equalization and optimize the quality of the received signal.

[0055] In this embodiment, the three-dimensional monitoring system of the marine ranch can collect marine ranch detection data from multiple different positions of the marine ranch by setting multiple water body sensors, and obtain underwater image information by using an underwater camera. The data aggregation unit is used to aggregate and externally transmit the marine ranch detection data and underwater image information, so as to provide data support for the monitoring and analysis of the marine ranch; since the water body sensors and underwater cameras can be distributed and set in the marine ranch, data can be collected over a large range for real-time analysis, and the water body sensors and underwater cameras can be set at fixed points without the need for operations such as patrol observation, thus saving costs.

[0056] In this embodiment, referring to Figure 2 , the onshore base station performs steps such as reception and demodulation, data storage and processing. These steps specifically include:

[0057] Reception and demodulation:

[0058] 1) Signal reception: The onshore base station receives the acoustic signal transmitted underwater through a hydrophone array. The signal reception ability in a specific direction is enhanced through beamforming technology, and background noise is suppressed.

[0059] 2) Signal demodulation: After the received signal is amplified and filtered, the acoustic signal is converted back to digital data through an MFSK demodulator. The demodulator performs soft decision decoding using the Viterbi Algorithm to further correct possible transmission errors.

[0060] 3) Error detection and correction: The demodulated data is again subjected to CRC check and combined with LDPC decoding for error detection and correction to ensure high reliability of the data.

[0061] Data storage and processing:

[0062] 1) Data storage: The successfully received data is stored in the SQL database of the onshore base station, along with a timestamp and location information for subsequent data analysis and retrieval.

[0063] 2) Real-time processing and feedback: The system can analyze the received environmental parameters in real time and trigger an alarm or execute a predefined control strategy according to preset thresholds, such as adjusting the water quality or environment in the aquaculture area.

[0064] By setting up an onshore base station, it is possible to receive the marine ranch detection data and underwater image information collected by the three-dimensional monitoring system of the marine ranch with the main body set in the water body for further analysis.

[0065] In this embodiment, the three-dimensional monitoring system of the marine ranch further includes a data processing unit. Referring to Figure 1 and Figure 2 , the data processing unit can be set at positions such as the onshore base station. After receiving the various marine ranch detection data and underwater image information sent by the data aggregation unit, the data processing unit analyzes and processes the various marine ranch detection data and underwater image information to obtain marine monitoring indicators. The marine monitoring indicators can be used to assist the operators of the marine ranch in analysis and decision-making.

[0066] In this embodiment, when the data processing unit executes the step of analyzing and processing the various marine ranch detection data and underwater image information to obtain marine monitoring indicators, it can specifically execute the following steps:

[0067] S401. Fuse the various marine ranch detection data to obtain fused data;

[0068] S402. Identify the underwater image information to obtain the time series data of the fish population quantity;

[0069] S403. Perform principal component analysis on the fused data and the time series data of the fish population quantity to obtain at least one marine monitoring indicator;

[0070] S404. Determine the comprehensive environmental index according to the various marine monitoring indicators and their respective corresponding first weights.

[0071] Before performing step S401, the detection data of each ocean ranch can be standardized. For example, normalization can be performed on the detection data of each ocean ranch to eliminate the influence of different dimensions of the detection data of ocean ranches detected by different types of water body sensors, so that the detection data of ocean ranches detected by different types of water body sensors can be compared and analyzed under the same framework.

[0072] When performing step S401, that is, fusing the detection data of each ocean ranch to obtain the fused data, the following steps can be specifically executed:

[0073] S40101. Determine the respective second weights corresponding to each water body sensor;

[0074] S40102. For any detection data of an ocean ranch, use the second weight corresponding to the water body sensor that detected the detection data of the ocean ranch as the weight corresponding to the detection data of the ocean ranch, and perform weighted averaging on the detection data of each ocean ranch to obtain the fused data.

[0075] In step S40101, the reliability can be determined according to the historical working parameters and real-time working parameters of each water body sensor, and the corresponding second weight is assigned according to the reliability. Among them, the value of the second weight assigned to each water body sensor is positively correlated with the reliability of this water body sensor.

[0076] In step S40102, assume that there are a total of M water body sensors, and the value of the second weight corresponding to the mth water body sensor is w′ m , and the value of the detection data of the ocean ranch detected by the mth water body sensor after standardization is value m , then the weighted average value of the detection data of the ocean ranch detected by all M water body sensors is

[0077]

[0078] Then the weighted average value VALUE can be used as the fused data obtained by performing step S401.

[0079] In this embodiment, when performing step S401, that is, fusing the detection data of each ocean ranch to obtain the fused data, the data fusion can also be performed in the way of Bayesian Estimation. Specifically, Bayesian Estimation updates the fusion result by introducing prior information and observation data to improve the accuracy of data fusion.

[0080] In step S402, the data processing unit identifies the underwater image information to obtain the time series data of the fish school quantity.

[0081] Specifically, the data processing unit can first adjust the frame rate and eliminate noise from the received underwater image information. For example, Gaussian Blur and Median Filtering are used to reduce the noise in the underwater image information, and Histogram Equalization is used to enhance the contrast of the underwater image information. Then, the data processing unit uses the Yolov8 algorithm to detect and count fish in the preprocessed underwater image information.

[0082] YOLO (You Only Look Once) is an efficient object detection algorithm based on convolutional artificial neural network (CNN). It transforms the object detection task into a regression problem and predicts the location and category of the object in one go through a single neural network, thus achieving fast object detection. Compared with other object detection algorithms, such as those based on sliding windows or region proposals, YOLO is faster. Yolov8 is a SOTA model that builds on previous versions of the Yolo series and introduces new features and improvements to further enhance performance and flexibility, making it a good choice for tasks such as object detection and image segmentation.

[0083] Taking the underwater image information in the form of a video as an example, the data processing unit uses the Yolov8 algorithm to identify each frame in the underwater image information, marks and counts the target fish in each frame, and the system calculates the number of fish in the fish school per unit time based on the counting results, thereby obtaining the fish quantity data for this frame. Since the underwater image information contains multiple frames and each frame has corresponding fish quantity data, multiple fish quantity data can form a time series, thus constituting the fish school quantity time series data series = {number 1 , number 2 ……}, where number 1 represents the number of fish in the first frame of the underwater image information (it can also include information such as the location of certain specific fish), and so on.

[0084] For the fish school quantity time series data obtained by processing with the Yolov8 algorithm, the Exponential Smoothing algorithm can be used to predict the fish school quantity time series data to provide a trend analysis of the dynamic changes of the fish school.

[0085] In this embodiment, the "fish school" can refer to groups of fish such as Pacific saury and tuna that are biologically classified as fish, or can also include groups such as squid that are not biologically fish but are caught like fish.

[0086] In step S403, principal component analysis (PCA) is performed on the fused data VALUE and the time series data series of the fish school quantity to obtain at least one ocean monitoring index.

[0087] Specifically, before performing PCA, all input fused data VALUE and the time series data series of the fish school quantity can be standardized to ensure that data with different units or magnitudes can be compared on the same scale. Then, the covariance matrix of the fused data VALUE and the time series data serises of the fish school quantity is calculated, and eigenvalue decomposition is performed on the covariance matrix to obtain the eigenvalues and corresponding eigenvectors of each component. The larger the eigenvalue of a component, the greater the variance contribution of the corresponding eigenvector of this component in the original data, that is, it can better represent the main change trend of the data. The system sorts according to the size of the eigenvalues and selects the largest k eigenvectors (i.e., the principal components) as the basis vectors of the new data space to construct the main indicators.

[0088] In step S403, the obtained eigenvectors are used as ocean monitoring indicators, and the corresponding first weights are calculated according to the eigenvalues corresponding to the eigenvectors. By executing step S403, PC 1 、PC 2 ……PC k and other k principal components, namely ocean monitoring indicators, and the eigenvalues λ 1 corresponding to the ocean monitoring indicator PC 1 、the eigenvalues λ 2 corresponding to the ocean monitoring indicator PC 2 ……the eigenvalues λ k corresponding to the ocean monitoring indicator PC k .

[0089] In step S403, the first weight w

[0090]

[0091] corresponding to the ocean monitoring indicator PC i can be calculated according to the formula i . Traverse i = 1, 2... k, so as to obtain the first weight w 1 corresponding to the ocean monitoring indicator PC 1 、the first weight w 2 corresponding to the ocean monitoring indicator PC 2 ……the first weight w k corresponding to the ocean monitoring indicator PC k and so on. Each ocean monitoring indicator represents the environmental status of a certain aspect of the ocean ranch, and the first weight wk reflects the contribution of the ocean monitoring index PC k to the total variance.

[0092] In step S404, the data processing unit can calculate the comprehensive environment index CEI according to the ocean monitoring indexes such as PC 1 , PC 2 ... PC k and the corresponding first weights w 1 , w 2 ... w k and so on. In this embodiment, the calculation formula of the comprehensive environment index CEI is:

[0093]

[0094] In this embodiment, the calculation formula of the comprehensive environment index CEI takes into account the weighted impacts of water quality, hydrology, meteorological conditions and fish quantity, and can quantitatively consider the environmental conditions of water quality, hydrology, meteorological conditions and fish quantity, providing a scientific basis for the management and decision-making of the marine ranch.

[0095] In this embodiment, when performing step S40101, that is, the step of determining the respective second weights corresponding to each water body sensor, the following steps can be specifically performed:

[0096] S4010101A. Obtain the spatial distances between each water body sensor and the underwater camera;

[0097] S4010102A. For any water body sensor, set the magnitude of the second weight corresponding to the water body sensor according to the corresponding spatial distance; wherein, the magnitude of the second weight is negatively correlated with the spatial distance.

[0098] Steps S4010101A - S4010102A are the first implementation manner of step S40101.

[0099] In step S4010101A, the spatial distance between the water body sensor and the underwater camera specifically refers to the distance between the spatial position of the water body sensor in the water body and the spatial position of the underwater camera in the water body, and can be calculated by recording the coordinates of both when arranging the water body sensor and the underwater camera. In actual use, it can be considered that the spatial distance between each water body sensor and the underwater camera is unchanged.

[0100] In step S4010102A, the value of the second weight corresponding to each water body sensor can be set to be negatively correlated with the spatial distance corresponding to this water body sensor. For example, for the m-th water body sensor, assuming the spatial distance between it and the underwater camera is distance m , then according to the formula

[0101]

[0102] Calculate the value w' of the second weight corresponding to the m-th water body sensor m , where DISTANCE is a fixed value.

[0103] In this embodiment, the principle of performing steps S4010101A - S4010102A is as follows: The fish population quantity time series data is obtained by analyzing the underwater image information. Since the underwater image information is captured by an underwater camera, this indicates that the fish population is active near the underwater camera. By performing steps S4010101A - S4010102A, it is possible to set a larger second weight for the water body sensors closer to the underwater camera. Therefore, the marine ranch detection data detected at positions closer to the fish population will be given a larger second weight. Thus, when monitoring the marine ranch, it is possible to focus on using the data collected near the fish population, thereby increasing the monitoring attention to the fish population and being beneficial to realizing the economy of the marine ranch.

[0104] In this embodiment, when performing step S40101, that is, the step of determining the second weight corresponding to each water body sensor, the following steps can be specifically performed:

[0105] S4010101B. Determine multiple fish population movement directions according to the fish population quantity time series data;

[0106] S4010102B. Perform divergence analysis on the fish population vector field to obtain divergence data at multiple positions within the field of view of the underwater camera;

[0107] S4010103B. Determine the second weight corresponding to each water body sensor according to each divergence data.

[0108] Steps S4010101B - S4010103B are the second execution manner of step S40101.

[0109] The principle of steps S4010101B - S4010103B is as Figure 3 shown.

[0110] Refer to Figure 3, in step S4010101B, for the fish population quantity time series data series output by the Yolov8 algorithm, multiple target fish can be marked from it, and the positions of these target fish in each frame of the fish population quantity time series data series are tracked. The fish in the fish population that do not belong to the target fish are ordinary fish. In any two adjacent frames of the fish population quantity time series data series, a fish population movement direction can be calculated based on the positions of the same target fish. In this way, when there are multiple frames in the fish population quantity time series data series and multiple target fish in each frame, multiple fish population movement directions can be obtained. Each fish population movement direction can be expressed in vector form, and multiple fish population movement directions form a vector field, that is, the fish population vector field. The distribution range of the fish population vector field is the field of view of the underwater camera.

[0111] In step S4010102B, for the fish population vector field, the divergence data at each position can be solved. Among them, the positions where the divergence data is greater than 0 are the convergence centers, indicating the positions that the fish population as a whole swims towards; the positions where the divergence data is less than 0 are the divergence centers, indicating the positions that the fish population as a whole swims away from; the positions where the divergence data is equal to 0 neither belong to the convergence centers nor the divergence centers, indicating the positions that the fish population as a whole swims through.

[0112] Suppose that by executing step S4010102B, multiple convergence centers and multiple divergence centers are determined.

[0113] In step S4010103B, taking the m-th water body sensor as an example, the convergence center closest to the m-th water body sensor is searched, and the distance between them is d convergence ; the divergence center closest to the m-th water body sensor is searched, and the distance between them is d divergence ;

[0114] The following judgment conditions are set:

[0115] (1) d convergence <distance threshold, and d convergence <d divergence ;

[0116] (2) d divergence <distance threshold, and d divergence <d convergence ;

[0117] (3) Other cases that do not meet judgment condition (1) or (2).

[0118] When judgment condition (1) is met, then the second weight w′ corresponding to the m-th water body sensor mDetermine it as a relatively large first value; when the judgment condition (2) is met, then the second weight w' corresponding to the m-th water body sensor m Determine it as a relatively small third value; when the judgment condition (3) is met, then the second weight w' corresponding to the m-th water body sensor m Determine it as a medium second value.

[0119] In this embodiment, the principle of executing steps S4010101B - S4010103B is as follows: In judgment condition (1), d convergence < distance threshold indicates that the m-th water body sensor is quite close to the convergence center, d convergence < d divergence indicates that the convergence center is closer to the m-th water body sensor than the divergence center, that is, the m-th water body sensor is closer to the position where the fish school as a whole swims towards. Therefore, set the second weight with the largest value (first value) for the m-th water body sensor; correspondingly, in judgment condition (2), d divergence < distance threshold indicates that the m-th water body sensor is quite close to the divergence center, d divergence < d convergence indicates that the divergence center is closer to the m-th water body sensor than the convergence center, that is, the m-th water body sensor is closer to the position where the fish school as a whole swims away. Therefore, set the second weight with the smallest value (third value) for the m-th water body sensor; if neither judgment condition (1) nor (2) holds, then judgment condition (3) holds, indicating that the m-th water body sensor is not relatively closer to any convergence center or divergence center, and the m-th water body sensor is at the position where the fish school as a whole swims through. Therefore, set the second weight with a medium value (second value) for the m-th water body sensor;

[0120] By executing steps S4010101B - S4010103B, it is possible to set a larger second weight for the water body sensor closer to the convergence center and a smaller second weight for the water body sensor closer to the divergence center. Therefore, the marine ranch detection data detected at the position closer to the position where the fish school converges will be given a greater second weight, so that when monitoring the marine ranch, it is possible to focus on using the data collected near the fish school, thereby increasing the monitoring attention to the fish school and being beneficial to realizing the economy of the marine ranch.

[0121] In this embodiment, by running the three-dimensional monitoring system of the marine ranch, the three-dimensional monitoring method of the marine ranch can be executed. Refer to Figure 4 , the three-dimensional monitoring method of the marine ranch includes the following steps:

[0122] S1. Through multiple water body sensors, respectively detect the marine ranch detection data; each water body sensor is used to be distributed and set at the corresponding position of the marine ranch;

[0123] S2. Obtain underwater image information by taking pictures with an underwater camera; the underwater camera is used to be set underwater in a marine ranch;

[0124] S3. Receive the detection data and underwater image information of each marine ranch through a data aggregation unit, and send out the detection data and underwater image information of each marine ranch; the data aggregation unit is used to establish connections with each water body sensor and underwater camera;

[0125] S4. Analyze and process the detection data and underwater image information of each marine ranch to obtain marine monitoring indicators.

[0126] In this embodiment, when the three-dimensional monitoring system of the marine ranch executes the three-dimensional monitoring method of the marine ranch, it can be divided into four parts: a data acquisition layer, a data transmission layer, a data processing and analysis layer, and a user interaction layer.

[0127] Data acquisition layer: In the data acquisition layer, the system comprehensively acquires the environmental data of the marine ranch by arranging various types of sensors and high-definition underwater cameras. The sensors include salinity, turbidity, and chlorophyll sensors for water quality detection, water temperature, pH value, and dissolved oxygen sensors for hydrological detection, and air temperature, air pressure, and wind speed sensors for meteorological detection. These sensors are all designed to be waterproof and corrosion-resistant, and are connected to the single-chip microcomputer through I2C, SPI, or UART interfaces to ensure the accuracy and stability of data acquisition. The single-chip microcomputer is responsible for acquiring sensor data at a high sampling rate and performing preliminary processing, such as noise filtering and data linearization. The high-definition underwater camera is used to acquire underwater videos in real time, and is connected to the single-chip microcomputer through the MIPI interface. The video data is stored in an external memory for subsequent processing.

[0128] Data transmission layer: In the data transmission layer, the system first compresses and encodes the acquired data to reduce the transmission bandwidth requirement and improve the reliability of the data. The video data is compressed using the H.264 encoding algorithm, and the sensor data reduces redundancy through differential encoding technology. All data is error-corrected through LDPC encoding before transmission to cope with the noise interference in the underwater environment. The compressed and encoded data is converted into acoustic signals through MFSK modulation and transmitted to the onshore base station through an underwater acoustic communication module. The receiving end of the onshore base station is equipped with a hydrophone array, which is responsible for receiving and demodulating the underwater acoustic signals, and restoring the original digital data through an LDPC decoder to ensure the integrity and accuracy of the data.

[0129] Data Processing and Analysis Layer: In the data processing and analysis layer, the system fuses and analyzes various sensor data received. First, the data is standardized to eliminate the dimensional differences between different sensors, and then noise is eliminated through Kalman filtering and smoothing processing. The data fusion combines the weighted average method and Bayesian estimation, adaptively adjusts the weights according to the historical confidence of the sensor data, and generates comprehensive environmental parameters. After the video data is preprocessed by Gaussian blur and histogram equalization, the Yolov8 deep learning algorithm is used for fish target detection and counting to obtain the number of fish per unit time. Subsequently, the system combines the fused sensor data with the fish number data, extracts the key monitoring indicators reflecting the overall situation of the marine ranch through principal component analysis, and calculates the Comprehensive Environmental Index (CEI) to provide a scientific decision-making basis for users.

[0130] User Interaction Layer: In the user interaction layer, the system provides intuitive monitoring and warning functions for users through the Human-Machine Interface (HMI). The real-time monitoring data is displayed on the interface, and users can view the historical and real-time change trends of various environmental parameters. The interface also provides a customized report generation function, facilitating users to export detailed monitoring reports. When the monitoring indicators exceed the preset safety thresholds, the system automatically triggers an alarm and sends the alarm information to relevant personnel via text message or email to ensure that necessary countermeasures can be taken in a timely manner. The user interface is designed to be simple and intuitive, supporting access from multiple devices, making it convenient for users to monitor the operation status of the marine ranch anytime and anywhere.

[0131] It is possible to write a computer program that executes the three-dimensional monitoring method of the marine ranch in this embodiment, write this computer program into a computer device or a storage medium. When the computer program is read and run, it executes the three-dimensional monitoring method of the marine ranch in this embodiment, thereby achieving the same technical effects as the three-dimensional monitoring method of the marine ranch in the embodiment.

[0132] It should be noted that, unless otherwise specified, when a certain feature is referred to as "fixed" or "connected" to another feature, it can be directly fixed or connected to the other feature, or indirectly fixed or connected to the other feature. In addition, the up, down, left, right, etc. descriptions used in this disclosure are only relative to the mutual positional relationship of the components of this disclosure in the drawings. The singular forms of "a", "an" and "the" used in this disclosure are also intended to include the plural forms, unless the context clearly indicates otherwise. In addition, unless otherwise defined, all technical and scientific terms used in this embodiment have the same meaning as commonly understood by those skilled in the technical field of this invention. The terms used in the description of this embodiment are only for describing specific embodiments, rather than for limiting the present invention. The term "and / or" used in this embodiment includes any combination of one or more of the related listed items.

[0133] It should be understood that although the terms first, second, third, etc. may be used in this disclosure to describe various elements, these elements should not be limited to these terms. These terms are only used to distinguish elements of the same type from each other. For example, without departing from the scope of this disclosure, the first element may also be referred to as the second element, and similarly, the second element may also be referred to as the first element. The use of any and all examples or exemplary language ("for example", "such as", etc.) provided in this embodiment is only intended to better illustrate the embodiments of the present invention and will not impose a limitation on the scope of the present invention unless otherwise required.

[0134] It should be recognized that embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The methods can be implemented in a computer program using standard programming techniques - including a non-transitory computer-readable storage medium configured with the computer program, where the storage medium so configured causes the computer to operate in a specific and predefined manner - according to the methods and drawings described in the specific embodiments. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, if desired, the program can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. In addition, for this purpose, the program can run on a programmed application-specific integrated circuit.

[0135] In addition, the operations of the processes described in this embodiment can be performed in any suitable order, unless this embodiment otherwise indicates or is otherwise clearly inconsistent with the context. The processes described in this embodiment (or variations and / or combinations thereof) can be executed under the control of one or more computer systems configured with executable instructions and can be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) executed jointly on one or more processors, by hardware, or a combination thereof. A computer program includes a plurality of instructions executable by one or more processors.

[0136] Further, the method can be implemented in any type of computing platform operatively connected, including but not limited to personal computers, minicomputers, mainframes, workstations, network or distributed computing environments, separate or integrated computer platforms, or communicating with charged particle tools or other imaging devices, etc. Aspects of the present invention can be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into the computing platform, such as a hard disk, optical read and / or write storage medium, RAM, ROM, etc., such that it is readable by a programmable computer and can be used to configure and operate the computer to perform the processes described herein when the storage medium or device is read by the computer. Additionally, the machine-readable code, or portions thereof, can be transmitted via wired or wireless networks. When such media includes instructions or programs that implement the above steps in conjunction with a microprocessor or other data processor, the invention of this embodiment includes these and other different types of non-transitory computer-readable storage media. When programmed according to the methods and techniques of the present invention, the present invention also includes the computer itself.

[0137] A computer program can be applied to input data to perform the functions of this embodiment, thereby transforming the input data to generate output data stored in non-volatile memory. The output information can also be applied to one or more output devices such as a display. In a preferred embodiment of the present invention, the transformed data represents physical and tangible objects, including a specific visual depiction of the physical and tangible objects generated on the display.

[0138] The above are only the preferred embodiments of the present invention. The present invention is not limited to the above-described embodiments. As long as it achieves the technical effects of the present invention by the same means, any modifications, equivalent replacements, improvements, etc., made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention. Within the scope of protection of the present invention, its technical solutions and / or implementation manners can have various different modifications and changes.

Claims

1. A three-dimensional monitoring system for marine ranches, characterized in that: The marine ranch three-dimensional monitoring system comprises: A plurality of water body sensors; the water body sensors are used to be distributed and arranged at corresponding positions of the marine ranch to detect and obtain marine ranch detection data; Underwater camera; the underwater camera is used to be set underwater in the ocean ranch to capture underwater image information; Data collection unit; the data collection unit is used to establish a connection with each of the water body sensors and the underwater camera, receive each of the marine ranch detection data and the underwater image information, and send each of the marine ranch detection data and the underwater image information to the outside; Data processing unit; the data processing unit is used to receive the marine ranch detection data and the underwater image information, analyze and process the marine ranch detection data and the underwater image information, and obtain the marine monitoring index; The analysis and processing of the marine ranch detection data and the underwater image information to obtain marine monitoring indicators includes: fusing the marine ranch detection data to obtain fused data; Identifying the underwater image information to obtain time series data of the number of fish schools; Performing principal component analysis on the fused data and the fish population time series data to obtain at least one of the marine monitoring indicators; The step of fusing the marine ranch detection data to obtain fused data includes: Determine a second weight corresponding to each of the water body sensors; For any of the marine ranch detection data, the second weight corresponding to the water body sensor that detects the marine ranch detection data is used as the weight corresponding to the marine ranch detection data, and weighted averaging is performed on each of the marine ranch detection data to obtain the fused data.

2. The three-dimensional monitoring system for marine ranching according to claim 1 is characterized in that: The analyzing and processing of the marine ranch detection data and the underwater image information to obtain the marine monitoring index also includes: A comprehensive environmental index is determined based on each of the marine monitoring indicators and their corresponding first weights.

3. The three-dimensional monitoring system for marine ranching according to claim 1 is characterized in that: The determining of the second weight corresponding to each of the water body sensors includes: Obtaining the spatial distance between each of the water body sensors and the underwater camera; For any of the water body sensors, the size of the second weight corresponding to the water body sensor is set according to the spatial distance corresponding to the water body sensor; wherein the size of the second weight is negatively correlated with the spatial distance.

4. The three-dimensional monitoring system for marine ranching according to claim 1 is characterized in that: The determining of the second weight corresponding to each of the water body sensors includes: Determining the movement directions of multiple schools of fish according to the time series data of the number of the school of fish; Performing divergence analysis on the fish school vector field to obtain divergence data of multiple positions within the field of view of the underwater camera; wherein the fish school vector field is a vector field formed by the movement directions of each of the fish schools; The second weight corresponding to each of the water body sensors is determined according to each of the divergence data.

5. The three-dimensional monitoring system for marine ranching according to claim 4 is characterized in that: The determining, according to each of the divergence data, the second weight corresponding to each of the water body sensors, comprises: Determine the convergence center and the divergence center within the field of view of the underwater camera according to each of the divergence data; For any of the water body sensors, when the distance between the water body sensor and the convergence center is less than the distance between the water body sensor and the divergence center and less than a distance threshold, the second weight corresponding to the water body sensor is determined to be a first value; when the distance between the water body sensor and the divergence center is less than the distance between the water body sensor and the convergence center and less than a distance threshold, the second weight corresponding to the water body sensor is determined to be a third value; otherwise, the second weight corresponding to the water body sensor is determined to be a second value; The first value is greater than the second value, and the second value is greater than the third value.

6. A three-dimensional monitoring method for marine ranches, characterized in that: The marine ranch three-dimensional monitoring method comprises: The marine ranch detection data is obtained by respectively detecting through a plurality of water body sensors; each of the water body sensors is used to be distributed and arranged at a corresponding position of the marine ranch; Underwater image information is obtained by shooting with an underwater camera; the underwater camera is used to be set underwater in the ocean ranch; The marine ranch detection data and the underwater image information are received by a data collection unit, and the marine ranch detection data and the underwater image information are sent to the outside; the data collection unit is used to establish a connection with each of the water body sensors and the underwater camera; Analyzing and processing the marine ranch detection data and the underwater image information to obtain marine monitoring indicators; The analysis and processing of the marine ranch detection data and the underwater image information to obtain marine monitoring indicators includes: fusing the marine ranch detection data to obtain fused data; Identifying the underwater image information to obtain time series data of the number of fish schools; Performing principal component analysis on the fused data and the fish population time series data to obtain at least one of the marine monitoring indicators; The step of fusing the marine ranch detection data to obtain fused data includes: Determine a second weight corresponding to each of the water body sensors; For any of the marine ranch detection data, the second weight corresponding to the water body sensor that detects the marine ranch detection data is used as the weight corresponding to the marine ranch detection data, and weighted averaging is performed on each of the marine ranch detection data to obtain the fused data.

7. A computer device, characterized in that: It includes a memory and a processor, the memory is used to store at least one program, and the processor is used to load at least one program to execute the marine ranch stereoscopic monitoring method described in claim 6.

Images