A method, device and storage medium for predicting the population quantity of Leiocassis longirostris

Abstract

The present invention discloses a method, device and storage medium for predicting the population quantity of Leiocassis longirostris. The method includes obtaining the quantity of fish schools, detecting the length and width of fish, the scanning area, the detection distance and the detection depth; inputting the detected length and width of fish into a pre-configured standard frequency estimation model to obtain an estimated value of the occurrence frequency; performing weighted calculation based on the estimated value of the occurrence frequency and the equivalent area to obtain a weighted value of the occurrence frequency; wherein the equivalent area is calculated according to the detected length and width of fish and is an area value representing the size of fish; performing product calculation based on the scanning area, a preset horizontal rotation angle and a preset vertical rotation angle to obtain the detected water area; performing mean calculation based on the detection depth to obtain the average detection depth; calculating the effective volume; and based on a preset multi-source data fusion model, performing calculation by combining the effective volume, the quantity of fish schools, the weighted value of the occurrence frequency and the detection distance to obtain the population quantity. This method can improve the prediction accuracy.

Description

Technical Field

[0001] The invention relates to the technical field of fish population quantity estimation, and in particular to a method, a device and a storage medium for predicting the population quantity of a long-snout catfish. Background Art

[0002] In recent years, the number of longnose catfish farmed has increased dramatically, and the intensity of fishing has continued to increase. However, due to the deterioration of the living water environment and human destruction, the wild resources of longnose catfish have declined sharply, the germplasm has seriously degraded, and it is rarely seen in the waters of the main and tributary rivers of the Yangtze River. At present, the distribution area of ​​longnose catfish is shrinking, the number of surviving catfish is getting smaller and smaller, and the population is under great threat. In order to increase the number of longnose catfish resources, it is first necessary to estimate the population size of longnose catfish in the fishing ban area, and then implement the increase and release. However, longnose catfish is a cold-water upstream migratory fish, and it is difficult to find its migration channel during the fishing ban period. Therefore, the commonly used method is to use sonar to detect and estimate the population size of longnose catfish.

[0003] In one existing technology, sonar is directly used to detect whether there are schools of longnose catfish in the water area, and then the estimated number of echoes is used to predict the number of longnose catfish in the school. However, fish of different sizes have different behaviors, distributions, and reflection characteristics of sonar signals in the water body. For example, small fish may be more likely to avoid sonar detection due to their small size, or their reflection signals may be weak and difficult to detect. Large fish may be easier to detect because their reflection signals are stronger. The existing technical solutions only count the number of echoes, ignoring the different reflection characteristics of sonar signals in the water body for fish of different sizes, resulting in inaccurate quantity prediction results, which is not conducive to achieving accurate prediction of the longnose catfish population. Summary of the invention

[0004] The invention provides a method, a device and a storage medium for predicting the population quantity of a long-snouted catfish, so as to realize accurate prediction of the population quantity of the long-snouted catfish.

[0005] In a first aspect, in order to solve the above technical problems, the present invention provides a method for predicting the population size of long-nosed catfish, comprising:

[0006] Obtain the number of fish schools, length and width of detected fish, scanning area, detection distance and detection depth;

[0007] Input the length and width of the detected fish into a pre-configured standard frequency estimation model to obtain an estimated value of the occurrence frequency;

[0008] Performing weighted calculation according to the occurrence frequency estimation value and the equivalent area to obtain the occurrence frequency weighted value; wherein the equivalent area is an area value calculated according to the length and width of the detected fish and is used to represent the size of the fish;

[0009] The detected water area is obtained by performing a product calculation based on the scanning area, the preset lateral rotation angle and the preset vertical rotation angle;

[0010] Performing mean calculation according to the detection depth to obtain an average detection depth;

[0011] Calculate the product of the detected water area and the detected average depth to obtain the effective volume;

[0012] Based on a preset multi-source data fusion model, the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance are calculated to obtain the population number;

[0013] The standard frequency estimation model is constructed and trained based on the length and width of historically detected fish, the area range of fish of different sizes, the historical occurrence frequency and the number of historically detected fish.

[0014] In an optional implementation, the configuration process of the standard frequency estimation model includes:

[0015] Get the length, width and number of historically detected fish;

[0016] Calculate the historical equivalent area based on the length and width of the historical detected fish;

[0017] Divide the area according to the historical equivalent area to obtain area intervals of fish of different sizes;

[0018] According to the area interval and the number of historical detected fish, an initial frequency estimation model is constructed and a frequency statistics operation is performed to obtain an initial frequency estimation model and a historical occurrence frequency;

[0019] Iterate according to the initial frequency estimation model and the historical occurrence frequency, and when the number of iterations is greater than or equal to the preset maximum number of iterations, terminate the iteration and obtain the optimized coefficients, thereby obtaining a standard frequency estimation model;

[0020] The initial frequency estimation model construction formula is as follows:

[0021] ;

[0022] in, represents the estimated frequency of occurrence, represents the linear coefficient, represents the median of the corresponding area interval, represents the nonlinear coefficient.

[0023] In an optional implementation, the iteration is performed according to the initial frequency estimation model and the historical occurrence frequency, and when the number of iterations is greater than or equal to a preset maximum number of iterations, the iteration is terminated and the optimized coefficients are obtained, thereby obtaining a standard frequency estimation model, including:

[0024] Initializing the initial frequency estimation model to obtain a first frequency estimation model;

[0025] Calculating according to the first frequency estimation model and the historical occurrence frequency to obtain first occurrence frequency estimation values ​​of fish of different sizes;

[0026] According to the first occurrence frequency estimation value and the historical occurrence frequency, error calculation is performed and parameters of the first frequency estimation model are updated to obtain a second frequency estimation model;

[0027] The iteration is continued, and when it is determined that the number of iterations is greater than or equal to the preset maximum number of iterations, the iteration is terminated and the optimized coefficients are obtained, thereby obtaining a standard frequency estimation model.

[0028] In an optional implementation, performing weighted calculation according to the occurrence frequency estimate value and the equivalent area to obtain the occurrence frequency weighted value includes:

[0029] The calculation formula of the occurrence frequency weighted value is as follows:

[0030] ;

[0031] in, represents the frequency weighted value, Indicates The estimated frequency of occurrence of fish of different sizes, Indicates Equivalent area for fish of different sizes.

[0032] In an optional implementation, the method of obtaining the detected water area by multiplying the scanned area, a preset lateral rotation angle, and a preset vertical rotation angle comprises:

[0033] The calculation formula for the detection water area is as follows:

[0034] ;

[0035] in, Indicates the area of ​​the detected water area. Indicates the preset conversion factor, represents the scan area, Indicates the preset lateral rotation angle. Indicates the preset vertical rotation angle.

[0036] In an optional implementation manner, the calculating the product of the detected water area and the detected average depth to obtain the effective volume includes:

[0037] The effective volume calculation formula is as follows:

[0038] ;

[0039] in, represents the effective volume, Indicates the area of ​​the detected water area. Indicates the average detection depth.

[0040] In an optional implementation, the preset multi-source data fusion model is used to calculate the effective volume, the number of fish, the weighted value of the occurrence frequency and the detection distance to obtain the population number, including:

[0041] The population calculation formula is as follows:

[0042] ;

[0043] in, represents the population size, Indicates the number of fish, represents the frequency weighted value, represents the effective volume, Indicates the preset single pulse quantity data, Indicates the preset single pulse energy data, Represents reflectivity coefficient data, Indicates the detection distance, Represents the preset energy constant.

[0044] In a second aspect, the present invention provides a device for predicting the population size of long-nosed catfish, comprising:

[0045] Input module, used to obtain the number of fish schools, length and width of detected fish, scanning area, detection distance and detection depth;

[0046] An occurrence frequency estimation module is used to input the length and width of the detected fish into a pre-configured standard frequency estimation model to obtain an occurrence frequency estimation value;

[0047] An appearance frequency weighting module, used for performing weighted calculation according to the appearance frequency estimation value and the equivalent area to obtain an appearance frequency weighted value; wherein the equivalent area is calculated according to the length and width of the detected fish and is used to represent the area value of the fish size;

[0048] A detection water area calculation module, used for performing product calculation according to the scanning area, a preset lateral turning angle and a preset vertical turning angle to obtain the detection water area;

[0049] A detection average depth calculation module is used to calculate the average value according to the detection depth to obtain the detection average depth;

[0050] An effective volume calculation module, used to calculate the product of the detected water area and the detected average depth to obtain an effective volume;

[0051] A population quantity calculation module, used to calculate the population quantity based on a preset multi-source data fusion model in combination with the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance;

[0052] The standard frequency estimation model is constructed and trained based on the length and width of historically detected fish, the area range of fish of different sizes, the historical occurrence frequency and the number of historically detected fish.

[0053] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements any one of the above-mentioned methods for predicting the population size of long-snouted catfish.

[0054] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the methods for predicting the population size of long-snouted catfish as described above.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] The invention provides a method for predicting the population of long-snout catfish, comprising obtaining the number of fish schools, the length and width of detected fish, a scanning area, a detection distance and a detection depth; inputting the length and width of the detected fish into a pre-configured standard frequency estimation model to obtain an occurrence frequency estimation value; performing weighted calculation according to the occurrence frequency estimation value and an equivalent area to obtain an occurrence frequency weighted value; wherein the equivalent area is an area value calculated according to the length and width of the detected fish and used to represent the size of the fish; performing product calculation according to the scanning area, a preset lateral turning angle and a preset vertical turning angle to obtain a detected water area; performing mean calculation according to the detection depth to obtain an average detection depth; calculating the product of the detected water area and the average detection depth to obtain an effective volume; performing calculation based on a preset multi-source data fusion model in combination with the effective volume, the number of fish schools, the occurrence frequency weighted value and the detection distance to obtain the population number; wherein the standard frequency estimation model is constructed and trained based on the length and width of historical detected fish, the area intervals of fish of different sizes, the historical occurrence frequency and the number of historical detected fish.

[0057] The present invention uses a standard frequency estimation model to predict the frequency of occurrence of fish. The model is constructed and trained based on the length and width of historically detected fish, the area interval of fish of different sizes, the historical frequency of occurrence, and the number of historically detected fish. This model can predict the frequency of occurrence of fish of different sizes in sonar detection according to the size distribution characteristics of fish, thereby more accurately reflecting the actual distribution of fish. At the same time, the frequency of occurrence weighted value further refines the consideration of the reflection characteristics of sonar signals of fish of different sizes. By adjusting the weighted value, the distribution density of fish in the detected waters can be more accurately estimated. Finally, the present invention is based on a multi-source data fusion model, combined with the effective volume, the number of fish, the frequency of occurrence weighted value and the detection distance for calculation, and the population size is obtained. This method can comprehensively utilize the information of multiple data sources, reduce the deviation and uncertainty that may be caused by a single data source, and improve the robustness of the prediction results, thereby realizing the accurate prediction of the population size of long-snout catfish. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a flow chart of a method for predicting the population quantity of long-snout catfish provided by an embodiment of the present invention;

[0059] Figure 2 The present invention is a schematic diagram of the structure of a device for predicting the population size of long-snout catfish provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] Reference Figure 1 The embodiment of the present invention provides a method for predicting the population size of long-nosed catfish, comprising the following steps:

[0062] S11, obtaining the number of fish schools, length and width of detected fish, scanning area, detection distance and detection depth;

[0063] S12, inputting the length and width of the detected fish into a pre-configured standard frequency estimation model to obtain an estimated value of the occurrence frequency;

[0064] S13, performing weighted calculation according to the occurrence frequency estimation value and the equivalent area to obtain an occurrence frequency weighted value; wherein the equivalent area is an area value calculated according to the length and width of the detected fish and is used to represent the size of the fish;

[0065] S14, performing product calculation according to the scanning area, a preset lateral turning angle and a preset vertical turning angle to obtain the detected water area;

[0066] S15, performing mean calculation according to the detection depth to obtain an average detection depth;

[0067] S16, calculating the product of the detected water area and the detected average depth to obtain an effective volume;

[0068] S17, based on a preset multi-source data fusion model, calculating in combination with the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance to obtain the number of populations;

[0069] The standard frequency estimation model is constructed and trained based on the length and width of historically detected fish, the area range of fish of different sizes, the historical occurrence frequency and the number of historically detected fish.

[0070] In step S11, it is necessary to obtain the number of fish, the length and width of the detected fish, the scanning area, the detection distance and the detection depth.

[0071] It should be noted that the number of fish refers to the number of individuals in a specific area obtained by detection means. This data directly reflects the abundance of fish in the area. For example, the present invention uses a monophonic sonar as a detection means. The sonar detects underwater objects by emitting sound waves and receiving reflected sound wave signals. Of course, other detection means can also be selected according to user needs and application scenarios, and the present invention does not limit this. The length and width of the detected fish refers to the measurement data of the length and width of the individuals in the detected fish. These data help to understand the size distribution of the fish, and then infer its age structure and growth status. The scanning area refers to the area of ​​water covered by a detection device such as a sonar during a detection process. This parameter affects the representativeness and reliability of the detection results. The detection distance refers to the maximum distance at which a device such as a sonar can effectively detect a school of fish. It determines the distance of the detection range, thereby affecting the detection efficiency and coverage. The detection depth refers to the depth of the water body that a device such as a sonar can detect.

[0072] Exemplarily, the present invention estimates the number of fish by analyzing sonar images and distribution. Of course, according to different user needs and application scenarios, you can also choose to analyze the intensity of sonar echo signals or other detection methods, which are not limited by the present invention. The length and width of the detected fish can be obtained by analyzing the morphology of individual fish in the sonar image. The scanning area can be calculated based on the scanning angle and detection distance of the sonar. Exemplarily, the present invention sets the scanning angle of the sonar to , the detection distance is , then the scan area It can be approximated as:

[0073] ;

[0074] The detection distance and detection depth are determined by the performance parameters of the sonar equipment, which are generally clearly marked in the technical specifications of the equipment. For example, the detection distance of the sonar used in the present invention is preset to 100 meters, and the detection depth is preset to 50 meters. After obtaining these data, they will be used for subsequent analysis and calculation.

[0075] In step S12, the length and width of the detected fish are input into a pre-configured standard frequency estimation model to obtain an estimated value of the occurrence frequency. The configuration process of the standard frequency estimation model includes:

[0076] Get the length, width and number of historically detected fish;

[0077] Calculate the historical equivalent area based on the length and width of the historical detected fish;

[0078] Divide the area according to the historical equivalent area to obtain area intervals of fish of different sizes;

[0079] According to the area interval and the number of historical detected fish, an initial frequency estimation model is constructed and a frequency statistics operation is performed to obtain an initial frequency estimation model and a historical occurrence frequency;

[0080] Iterate according to the initial frequency estimation model and the historical occurrence frequency, and when the number of iterations is greater than or equal to the preset maximum number of iterations, terminate the iteration and obtain the optimized coefficients, thereby obtaining a standard frequency estimation model;

[0081] The initial frequency estimation model construction formula is as follows:

[0082] ;

[0083] in, represents the estimated frequency of occurrence, represents the linear coefficient, represents the median of the corresponding area interval, represents the nonlinear coefficient.

[0084] It should be noted that the length and width of the detected fish in step S12 are obtained by sonar detection of the school of fish, and include the length and width information of the detected fish. The standard frequency estimation model is a pre-built mathematical model for estimating the frequency of occurrence of long snout catfish according to the length and width of the detected fish. This model is fitted based on a large amount of historical data, and can output corresponding frequency estimation values ​​according to the input length and width of the detected fish. The construction process of the model includes analysis and statistics of historical data, and uses mathematical methods such as regression analysis to determine the parameters of the model. The length and width of the historical detected fish refer to the length and width data of the long snout catfish collected in the past, and these data are used to construct the initial frequency estimation model. Through the analysis of these historical data, the size distribution characteristics of the long snout catfish can be understood, providing a basis for the construction of the model. The number of historical detected fish refers to the number of schools of fish detected under the same or similar conditions in the past. These data, together with the length and width of the historical detected fish, are used to construct the initial frequency estimation model to help determine the frequency of occurrence of fish of different sizes in the school of fish. The historical equivalent area is obtained by calculating the length and width of the historically detected fish, which reflects the area occupied by each fish. Exemplarily, the calculation method adopted by the present invention is to multiply the length and width of the fish to obtain an approximate area value. This data is used for subsequent area interval division operations. Area interval division is to divide the historical equivalent area into different intervals according to certain rules, so as to classify and count fish of different sizes. Exemplarily, the present invention divides the equivalent area into three intervals: small, medium and large, and each interval corresponds to a specific area range. Of course, other division methods can also be selected according to different user needs and application scenarios, and the present invention does not limit this. The occurrence frequency estimate is obtained by performing frequency statistics operations on the area interval and the number of historically detected fish. This data reflects the frequency of occurrence of fish in different area intervals in historical data.

[0085] In the initial frequency estimation model construction formula, It represents the estimated value of occurrence frequency, that is, the frequency of fish occurrence in different area intervals; Represents the linear coefficient, which is a model parameter used to adjust the linear part of the model; It represents the median of the corresponding area interval, that is, the median of the interval to which the equivalent area of ​​the fish belongs; Represents the nonlinear coefficient, which is used to adjust the nonlinear part of the model. This formula is used to describe the relationship between the size of the fish and its frequency of occurrence, which is determined by fitting historical data and This model can be used to predict the frequency with which fish of different sizes are likely to appear under new detection conditions.

[0086] It should be noted that the iteration is performed according to the initial frequency estimation model and the historical occurrence frequency, and when the number of iterations is greater than or equal to the preset maximum number of iterations, the iteration is terminated and the optimized coefficients are obtained, thereby obtaining a standard frequency estimation model, including:

[0087] Initializing the initial frequency estimation model to obtain a first frequency estimation model;

[0088] Calculating according to the first frequency estimation model and the historical occurrence frequency to obtain first occurrence frequency estimation values ​​of fish of different sizes;

[0089] According to the first occurrence frequency estimation value and the historical occurrence frequency, error calculation is performed and parameters of the first frequency estimation model are updated to obtain a second frequency estimation model;

[0090] Continue iterating, and when it is determined that the number of iterations is greater than or equal to the preset maximum number of iterations, end the iteration and obtain the optimized coefficients, thereby obtaining a standard frequency estimation model.

[0091] It should be noted that the key features involved in this step include an initial frequency estimation model, an estimated frequency of occurrence, a preset maximum number of iterations, a first frequency estimation model, a first estimated frequency of occurrence, and a second frequency estimation model. The initial frequency estimation model is a mathematical model based on historical data for estimating the frequency of occurrence of fish schools of different sizes. The estimated frequency of occurrence refers to the number of long-nosed catfish populations actually measured in a specific water area in the past, and these data are used for calibration and optimization to ensure that the output of the model is closer to the actual situation. The preset maximum number of iterations is a parameter for controlling the model optimization process, which determines the upper limit of the number of iterations performed. Exemplarily, the preset maximum number of iterations of the present invention is set to 10,000 times, which means that the optimization process of the model parameters is performed up to 10,000 iterations. The first frequency estimation model is the first instantiation of the initial frequency estimation model in the iterative process, and its parameters are based on the initial value. According to the first frequency estimation model and the historical frequency of occurrence, the theoretical calculation of the frequency of occurrence estimation is performed, that is, the theoretical frequency of occurrence of fish schools of different sizes is predicted using the model to obtain the first frequency of occurrence estimate. This calculation process applies the model parameters to historical data and simulates the expected results of the fish school size distribution through mathematical formulas or algorithms. Subsequently, an error calculation is performed based on the first occurrence frequency estimate and the historical occurrence frequency. The purpose of the error calculation is to quantify the difference between the model prediction value and the actual observation value. Exemplarily, the error calculation method adopted by the present invention is the mean square error (MSE). Of course, according to different user needs and application scenarios, the absolute error or other error calculation methods can also be selected, and the present invention is not limited to this. Based on the calculated error, the parameters of the first frequency estimation model are updated to obtain the second frequency estimation model. The parameter update process adopts an optimization algorithm. Exemplarily, the present invention adopts the gradient descent method. Of course, according to different user needs and application scenarios, other parameter update methods can also be selected. Continue to iterate the above process, and each iteration will perform a theoretical occurrence frequency estimation calculation based on the latest model parameters, and then calculate the error and update the model parameters. When the number of iterations reaches or exceeds the preset maximum number of iterations, the iterative process ends, and the model obtained at this time is the standard frequency estimation model.

[0092] In step S13, performing weighted calculation according to the occurrence frequency estimation value and the equivalent area to obtain the occurrence frequency weighted value includes:

[0093] The calculation formula of the occurrence frequency weighted value is as follows:

[0094] ;

[0095] in, represents the frequency weighted value, Indicates The estimated frequency of occurrence of fish of different sizes, Indicates The equivalent area of ​​the fish of different sizes is calculated based on the length and width of the detected fish and is used to represent the area value of the fish size.

[0096] It should be noted that in step S13, the key features include the estimated value of the frequency of occurrence and the length and width of the detected fish, and these data are used to perform calculation operations to obtain the weighted value of the frequency of occurrence. The estimated value of the frequency of occurrence is obtained based on the pre-built model and the input length and width of the detected fish, which reflects the possibility of fish of different sizes being detected. The length and width of the detected fish are directly obtained through sonar detection, and contain the body length and width information of the detected fish. These data are used to calculate the equivalent area of ​​each size of fish. The equivalent area calculation operation is performed based on the length and width of the detected fish, with the aim of converting the size of each fish into an area value representing its size. Exemplarily, the present invention assumes that the shape of the fish is approximately rectangular, and is calculated by a simple geometric formula. The length of the fish is , the width is , then the equivalent area Can be approximated as Of course, other equivalent area calculation methods can be selected according to different application scenarios and user needs, and the present invention does not limit this.

[0097] In this embodiment, the above calculation formula is used to comprehensively consider the estimated frequency of occurrence of fish of different sizes and their equivalent areas to calculate a comprehensive frequency weighted value In this formula, Representative The estimated frequency of occurrence of fish of different sizes, Represents the equivalent area of ​​fish of corresponding sizes. A weighted average frequency of occurrence is obtained by summing the product of the frequency estimate of each size of fish with its equivalent area and then dividing it by the sum of all equivalent areas. This method takes into account the different contributions of fish of different sizes in the detection process. In population prediction, directly using a single frequency estimate may not accurately reflect the actual situation, especially when the fish school contains fish of multiple sizes. The frequency weighting value can more accurately estimate the actual probability of detection of the fish school by considering the frequency estimate of fish of different sizes and the equivalent area. For example, if an area is mainly composed of small fish, but the frequency of small fish is low, directly using a single frequency estimate may overestimate the population size. The frequency weighting value can adjust the final frequency of occurrence according to the frequency of occurrence and the proportion of the number of large and small fish, so as to more accurately reflect the population size.

[0098] In step S14, the product calculation is performed according to the scanning area, the preset lateral angle and the preset vertical angle to obtain the detected water area, including:

[0099] The calculation formula for the detection water area is as follows:

[0100] ;

[0101] in, Indicates the area of ​​the detected water area. Indicates the preset conversion factor, represents the scan area, Indicates the preset lateral rotation angle. Indicates the preset vertical rotation angle.

[0102] It should be noted that in step S14, the key features include the scanning area, the preset conversion coefficient, the preset lateral angle, and the preset vertical angle. These features are used to perform calculation operations on the detected water area to obtain the detected water area. The scanning area is directly measured by the sonar device under specific operating conditions, reflecting the water area that the sonar device can cover in one scanning process. The preset conversion coefficient is a proportional factor used to adjust the calculation result, and its value is preset according to the characteristics of the sonar device and the water conditions. Exemplarily, the present invention is set to 0.8, which is used to correct the difference between the actual detection area and the theoretical scanning area. The preset lateral angle and the preset vertical angle represent the scanning angles of the sonar device in the lateral and vertical directions, respectively. These two parameters are also preset according to the specifications of the equipment and the detection requirements. Exemplarily, the present invention sets the lateral angle to 45 degrees and the vertical angle to 30 degrees. They affect the coverage and shape of the sonar beam. When performing the calculation operation of the detected water area, the above features are substituted into the calculation formula for the detected water area. Among them, is the preset conversion factor, is the scan area, is the preset horizontal angle. is the preset vertical turning angle. Through calculation, the detected water area is obtained This data reflects the actual water area that can be effectively detected after taking into account the scanning characteristics and angle settings of the sonar equipment.

[0103] In step S15, it should be noted that in step S15, the key feature is the detection depth, which is a collection of water depth information obtained by sonar equipment at different measurement points. These data points are distributed at multiple locations in the water area to be measured, and are used to fully understand the depth of the water area. Exemplarily, the present invention obtains these data by emitting sound waves to the water body through a sonar device, and receiving sound waves reflected from the bottom of the water, and calculating the depth by measuring the time it takes for the sound waves to go back and forth. The mean calculation operation is the process of processing these detection depths, the purpose of which is to obtain a value representing the average depth of the entire detection area, that is, the detection average depth. This operation is achieved by adding all the detection depths and then dividing them by the total number of data points. The detection average depth can be used to evaluate the overall depth characteristics of the water area to be measured, and provide basic information for terrain analysis of the water area, fishery resource assessment, and related oceanographic research.

[0104] In step S16, the calculating the product of the detected water area and the detected average depth to obtain the effective volume includes:

[0105] The effective volume calculation formula is as follows:

[0106] ;

[0107] in, represents the effective volume, Indicates the area of ​​the detected water area. Indicates the average detection depth.

[0108] It should be noted that in step S16, the key features include the area of ​​detected waters and the average depth of detection. These data are used to perform an effective volume calculation operation to obtain the effective volume. The area of ​​detected waters is directly measured by the sonar equipment under specific operating conditions, reflecting the area of ​​waters that the sonar equipment can cover in one scanning process. The average depth of detection is obtained by calculating the mean of the detection depth, representing the average depth of the waters to be measured. The effective volume calculation formula is used to multiply the area of ​​detected waters and the average depth of detection to obtain a value representing the effective volume of the waters to be measured. The obtained effective volume is of great significance in subsequent applications. It can be used to evaluate the total volume of the waters to be measured, and provide basic information for terrain analysis of the waters, fishery resource assessment, and related oceanographic research.

[0109] In step S17, in one embodiment, the calculation based on the preset multi-source data fusion model, combined with the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance, to obtain the population number includes:

[0110] The population calculation formula is as follows:

[0111] ;

[0112] in, represents the population size, Indicates the number of fish, represents the frequency weighted value, represents the effective volume, Indicates the preset single pulse quantity data, Indicates the preset single pulse energy data, Represents reflectivity coefficient data, Indicates the detection distance, Represents the preset energy constant.

[0113] It should be noted that in this embodiment, the key features involved include effective volume, number of fish, weighted value of occurrence frequency, detection distance, preset single pulse energy data, preset single pulse number data and preset energy constant. These features are used to perform population quantity calculation operations to obtain the population quantity. The effective volume is obtained by multiplying the area of ​​the detected water area by the average detection depth, which represents the volume of water actually covered by the sonar device during the detection process. The number of fish is directly obtained through sonar detection, reflecting the number of individual fish detected in the detection area. The weighted value of occurrence frequency is calculated based on the estimated value of occurrence frequency and the length and width of the detected fish, which represents the probability of fish of different sizes being detected. The detection distance is the maximum distance at which the sonar device can effectively detect the fish school, and this parameter is determined by the performance of the sonar device. The preset single pulse energy data refers to the energy carried when each sonar pulse is emitted. Exemplarily, the present invention is preset to 10 joules. This parameter affects the propagation distance and penetration ability of the sonar wave. The preset single pulse number data refers to the number of pulses emitted by the sonar during a detection process. Exemplarily, the present invention presets it to 100 times, which affects the coverage of the detection and the density of the data. The preset energy constant is a proportional factor used to calibrate the calculation results. Exemplarily, the present invention presets it to 0.8, which is used to adjust the influence of factors such as energy loss during the calculation process. The population calculation formula is used to comprehensively consider all the above characteristics and calculate a value representing the population of long-snouted catfish in the waters to be tested.

[0114] In this formula, represents the population size, Indicates the number of fish, represents the frequency weighted value, represents the effective volume, Indicates the preset single pulse quantity data, Indicates the preset single pulse energy data, Represents reflectivity coefficient data, Indicates the detection distance, represents the preset energy constant. By substituting these data into the formula, an estimated population size that takes into account detection efficiency, coverage and energy factors can be calculated. The obtained population size M is of great significance in subsequent applications. It can be used to assess the overall number of long-snout catfish in the waters to be tested.

[0115] In another embodiment, the preset multi-source data fusion model uses a data fusion algorithm to integrate information from different data sources. For example, a Bayesian fusion algorithm is used to integrate information from different data sources and recalculate the population size:

[0116] (1) Define the prior distribution and define the prior distribution of the population size based on historical data. For example, assuming that the population size follows a normal distribution, its mean and variance can be estimated based on historical data.

[0117] (2) Collect new observation data, and collect current observation data from sonar detection and other sensors, including the effective volume, the number of fish schools, the weighted value of the occurrence frequency, and the detection distance, etc.

[0118] (3) Construct a likelihood function based on the new observation data. For example, assuming that the observation data follows a Poisson distribution or a normal distribution, its parameters can be estimated based on the observation data.

[0119] (4) Bayesian update: Using Bayes’ theorem, combined with the prior distribution and likelihood function, the posterior distribution of the population size is calculated. The posterior distribution reflects the most reasonable estimate of the population size after considering the new data.

[0120] (5) Estimate the population size and extract the expected value or median of the population size from the posterior distribution as the final estimated value, which is used as the final output population size.

[0121] It should be noted that the Bayesian fusion algorithm allows the model to dynamically update predictions based on new data. This means that as more data is acquired, the prediction results can be continuously optimized and adjusted to reflect the latest information. By combining multiple data sources, the Bayesian fusion algorithm can reduce the bias that may be introduced by a single data source, thereby improving the accuracy and reliability of the prediction.

[0122] In summary, the present invention uses a standard frequency estimation model to predict the frequency of occurrence of fish. The model is constructed and trained based on the length and width of historically detected fish, the area interval of fish of different sizes, the historical frequency of occurrence, and the number of historically detected fish. This model can predict the frequency of occurrence of fish of different sizes in sonar detection according to the size distribution characteristics of fish, thereby more accurately reflecting the actual distribution of fish. At the same time, the frequency of occurrence weighted value further refines the consideration of the reflection characteristics of sonar signals of fish of different sizes. By adjusting the weighted value, the distribution density of fish in the detected waters can be more accurately estimated. Finally, the present invention is based on a multi-source data fusion model, combined with the effective volume, the number of fish, the frequency of occurrence weighted value and the detection distance for calculation, and the population size is obtained. This method can comprehensively utilize the information of multiple data sources, reduce the deviation and uncertainty that may be caused by a single data source, and improve the robustness of the prediction results, thereby achieving an accurate prediction of the population size of the long-snout catfish.

[0123] Reference Figure 2 The embodiment of the present invention provides a device for predicting the population size of long-nosed catfish, comprising:

[0124] Input module, used to obtain the number of fish schools, length and width of detected fish, scanning area, detection distance and detection depth;

[0125] An occurrence frequency estimation module is used to input the length and width of the detected fish into a pre-configured standard frequency estimation model to obtain an occurrence frequency estimation value;

[0126] An appearance frequency weighting module, used for performing weighted calculation according to the appearance frequency estimation value and the equivalent area to obtain an appearance frequency weighted value; wherein the equivalent area is calculated according to the length and width of the detected fish and is used to represent the area value of the fish size;

[0127] A detection water area calculation module, used for performing product calculation according to the scanning area, a preset lateral turning angle and a preset vertical turning angle to obtain the detection water area;

[0128] A detection average depth calculation module is used to calculate the average value according to the detection depth to obtain the detection average depth;

[0129] An effective volume calculation module, used to calculate the product of the detected water area and the detected average depth to obtain an effective volume;

[0130] A population quantity calculation module, used to calculate the population quantity based on a preset multi-source data fusion model in combination with the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance;

[0131] The standard frequency estimation model is constructed and trained based on the length and width of historically detected fish, the area range of fish of different sizes, the historical occurrence frequency and the number of historically detected fish.

[0132] It should be noted that the device for predicting the population size of long-snouted catfish provided in an embodiment of the present invention is used to execute all the process steps of the method for predicting the population size of long-snouted catfish in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and thus will not be described in detail.

[0133] The embodiment of the present invention also provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method for predicting the population size of long-nosed catfish are implemented, for example Figure 1 Alternatively, the processor implements the functions of each module / unit in the above-mentioned device embodiments when executing the computer program.

[0134] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0135] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0136] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0137] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store operating devices, at least one application required for a function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0138] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0139] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0140] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for predicting the population size of long-nosed catfish, characterized in that: include: Obtain the number of fish schools, length and width of detected fish, scanning area, detection distance and detection depth; Input the length and width of the detected fish into a pre-configured standard frequency estimation model to obtain an estimated value of the occurrence frequency; Performing weighted calculation according to the occurrence frequency estimation value and the equivalent area to obtain the occurrence frequency weighted value; wherein the equivalent area is an area value calculated according to the length and width of the detected fish and is used to represent the size of the fish; The detected water area is obtained by performing a product calculation based on the scanning area, the preset lateral rotation angle and the preset vertical rotation angle; Performing mean calculation according to the detection depth to obtain an average detection depth; Calculate the product of the detected water area and the detected average depth to obtain the effective volume; Based on a preset multi-source data fusion model, the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance are calculated to obtain the population number; The standard frequency estimation model is constructed and trained based on the length and width of historically detected fish, the area intervals of fish of different sizes, the historical occurrence frequency, and the number of historically detected fish; The configuration process of the standard frequency estimation model includes: Get the length, width and number of historically detected fish; Calculate the historical equivalent area based on the length and width of the historical detected fish; Divide the area according to the historical equivalent area to obtain area intervals of fish of different sizes; According to the area interval and the number of historical detected fish, an initial frequency estimation model is constructed and a frequency statistics operation is performed to obtain an initial frequency estimation model and a historical occurrence frequency; Iterate according to the initial frequency estimation model and the historical occurrence frequency, and when the number of iterations is greater than or equal to the preset maximum number of iterations, terminate the iteration and obtain the optimized coefficients, thereby obtaining a standard frequency estimation model; The initial frequency estimation model construction formula is as follows: ; in, represents the estimated frequency of occurrence, represents the linear coefficient, represents the median of the corresponding area interval, represents the nonlinear coefficient; The calculation based on the preset multi-source data fusion model, combined with the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance, to obtain the population number includes: The population calculation formula is as follows: ; in, represents the population size, Indicates the number of fish, represents the frequency weighted value, represents the effective volume, Indicates the preset single pulse quantity data, Indicates the preset single pulse energy data, Represents reflectivity coefficient data, Indicates the detection distance, Represents the preset energy constant.

2. The method for predicting the population size of long-snouted catfish according to claim 1, characterized in that: The iterating according to the initial frequency estimation model and the historical occurrence frequency, when the number of iterations is greater than or equal to the preset maximum number of iterations, the iteration is terminated and the optimized coefficients are obtained, thereby obtaining a standard frequency estimation model, including: Initializing the initial frequency estimation model to obtain a first frequency estimation model; Calculating according to the first frequency estimation model and the historical occurrence frequency to obtain first occurrence frequency estimation values ​​of fish of different sizes; According to the first occurrence frequency estimation value and the historical occurrence frequency, error calculation is performed and parameters of the first frequency estimation model are updated to obtain a second frequency estimation model; The iteration is continued, and when it is determined that the number of iterations is greater than or equal to the preset maximum number of iterations, the iteration is terminated and the optimized coefficients are obtained, thereby obtaining a standard frequency estimation model.

3. The method for predicting the population size of long-snouted catfish according to claim 1, characterized in that: The step of performing weighted calculation according to the occurrence frequency estimation value and the equivalent area to obtain the occurrence frequency weighted value includes: The calculation formula of the occurrence frequency weighted value is as follows: ; in, represents the frequency weighted value, Indicates The estimated frequency of occurrence of fish of different sizes, Indicates Equivalent area for fish of different sizes.

4. The method for predicting the population size of long-snouted catfish according to claim 1, characterized in that: The method of obtaining the detected water area by multiplying the scanned area, the preset lateral rotation angle and the preset vertical rotation angle comprises: The calculation formula for the detected water area is as follows: ; in, Indicates the area of ​​the detected water area. Indicates the preset conversion factor, represents the scan area, Indicates the preset lateral rotation angle. Indicates the preset vertical rotation angle.

5. The method for predicting the population size of long-snouted catfish according to claim 1, characterized in that: The calculating the product of the detected water area and the detected average depth to obtain the effective volume includes: The effective volume calculation formula is as follows: ; in, represents the effective volume, Indicates the area of ​​the detected water area. Indicates the average detection depth.

6. A device for predicting the population size of long-nosed catfish, characterized in that: The method for predicting the population size of long-nosed catfish as claimed in any one of claims 1 to 5 comprises: Input module, used to obtain the number of fish schools, length and width of detected fish, scanning area, detection distance and detection depth; An occurrence frequency estimation module is used to input the length and width of the detected fish into a pre-configured standard frequency estimation model to obtain an occurrence frequency estimation value; An appearance frequency weighting module, used for performing weighted calculation according to the appearance frequency estimation value and the equivalent area to obtain an appearance frequency weighted value; wherein the equivalent area is calculated according to the length and width of the detected fish and is used to represent the area value of the fish size; A detection water area calculation module, used for performing product calculation according to the scanning area, a preset lateral turning angle and a preset vertical turning angle to obtain the detection water area; A detection average depth calculation module is used to calculate the average value according to the detection depth to obtain the detection average depth; An effective volume calculation module, used to calculate the product of the detected water area and the detected average depth to obtain an effective volume; A population quantity calculation module, used to calculate the population quantity based on a preset multi-source data fusion model in combination with the effective volume, the number of fish schools, the weighted value of the occurrence frequency and the detection distance; The standard frequency estimation model is constructed and trained based on the length and width of historically detected fish, the area range of fish of different sizes, the historical occurrence frequency and the number of historically detected fish.

7. An electronic equipment device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for predicting the population size of long-snouted catfish as claimed in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for predicting the population size of long-snouted catfish as described in any one of claims 1 to 5.

Images