Multi-source data analysis-based lamplight purse seine fishing strategy optimization method and device

Abstract

The invention relates to a lamplight purse seine fishing strategy optimization method and device based on multi-source data analysis, and belongs to the technical field of lamplight purse seine. Further acquiring multi-source data information of each sub-region in the target region, and constructing a biological population migration position point in combination with the multi-source data information and a biological population inhabitation prediction model, thereby constructing a migration route according to the biological population migration position point, setting different light fishing strategies, simulating the fishing effect of each light fishing strategy, and obtaining the target area. And finally, formulating a final light fishing strategy based on the fishing effect of each light fishing strategy and the migration route. According to the method, the migration position point of the target fishing organism population is optimized according to the survival habit data of the target organism population, the lamplight fishing strategy can be further optimized, and the fishing efficiency of the lamplight purse seine is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of light-purse seine technology, and in particular to a light-purse seine fishing strategy optimization method and device based on multi-source data analysis. Background Art

[0002] Light purse seine technology is a fishing technology that uses lights to attract fish to gather, and then uses seine nets to catch them. This technology relies on the natural tendency of fish to light. By using strong light at night, it can effectively concentrate fish and improve fishing efficiency. Light purse seine technology usually includes components such as light ships, net ships and light boats. The combination of underwater lights and surface lights can greatly improve the efficiency of attracting fish. However, when fishing with light purse seines in the existing technology, the migration of target fish schools is not fully considered, so the fishing with light purse seines is useless, resulting in the fishing efficiency of light purse seines not meeting expectations. Summary of the Invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method and device for optimizing light-purse seine fishing strategies based on multi-source data analysis.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is:

[0005] A first aspect of the present invention provides a method for optimizing a light-purse seine fishing strategy based on multi-source data analysis, comprising the following steps:

[0006] Obtaining data on the type of biological populations currently targeted for fishing and historical environmental preference characteristics, and constructing a biological population habitat prediction model based on the data on the type of biological populations currently targeted for fishing and historical environmental preference characteristics;

[0007] Acquire multi-source data information of each sub-region in the target region, and construct biological population migration location points by combining the multi-source data information and a biological population habitat prediction model;

[0008] Constructing a migration route based on the migration location of the biological population, setting different light fishing strategies, and simulating the fishing effect of each light fishing strategy;

[0009] The final light fishing strategy is formulated based on the fishing effect and migration route of each light fishing strategy.

[0010] Furthermore, in this method, a biological population habitat prediction model is constructed based on the biological population type data of the current target fishing target and the historical environmental preference characteristic data, specifically:

[0011] Constructing a biological population habitat prediction model based on a deep neural network, using the biological population type data of the current target fishing target and the historical environmental preference characteristic data as graph nodes, connecting the graph nodes to construct a topological structure graph;

[0012] Introducing a cyclic spatial attention mechanism, inputting the graph nodes into the cyclic spatial attention mechanism, focusing attention on the graph nodes, generating an attention feature map, and inputting the attention feature map into a hidden layer of a deep neural network;

[0013] The state of the hidden layer is updated, the model parameters of the biological population habitat prediction model are saved, and the biological population habitat prediction model is output.

[0014] Furthermore, in this method, multi-source data information of each sub-region in the target region is obtained, and the biomass migration location points are constructed by combining the multi-source data information and the biomass habitat prediction model, specifically including:

[0015] Acquiring multi-source data information of each sub-region in the target area and target fishing biological population type data, and inputting the multi-source data information of each sub-region in the target area and target fishing biological population type data into the biological population habitat prediction model for prediction;

[0016] Obtaining, by prediction, a sub-region in the target area where the current target biological population inhabits, and using the sub-region in the target area where the current target biological population inhabits as a migration location point of the biological population;

[0017] Acquire image data information of the migration location of each biological population through remote sensing technology, and calculate survival resource data based on the image data information of the migration location of each biological population, and set a survival resource data threshold;

[0018] The migration location of the biological population where the survival resource data is lower than the survival resource data threshold is used as a short-term migration location of the biological population, and the migration location of the biological population where the survival resource data is not lower than the survival resource data threshold is used as a long-term migration location of the biological population;

[0019] Biological population migration position points are constructed according to the short-term migration position points of the biological population and the long-term migration position points of the biological population, and the constructed biological population migration position points are output.

[0020] Furthermore, in this method, a migration route is constructed based on the migration locations of the biological population, specifically:

[0021] Obtaining the migration route of the target biological population within a preset time through remote sensing technology, and obtaining the survival habit data of the target biological population, and obtaining the migration location point of the target biological population at the current time stamp based on the migration route of the target biological population within the preset time;

[0022] Obtaining, from the biological population migration point, biological population migration point adjacent to the migration point of the target biological population at the current time stamp, based on the biological population migration point adjacent to the migration point of the target biological population at the current time stamp;

[0023] Constructing an estimated migration route based on the migration location points of the biological populations adjacent to the migration location point of the target biological population at the current time stamp and the migration location point of the target biological population at the current time stamp;

[0024] When there is a temporary migration location point of a biological population in the estimated migration route, obtaining the next adjacent biological population migration location point, and updating the estimated migration route based on the next adjacent biological population migration location point;

[0025] When there is no short-term migration point of the biological population in the estimated migration route, the estimated migration route is maintained unchanged and is used as the final migration route.

[0026] Furthermore, in this method, different light fishing strategies are set to simulate the fishing effect of each light fishing strategy, specifically including:

[0027] Set different light net working parameters and working quantities at the historical migration locations along the target migration route, build several light fishing strategies based on the different light net working parameters and working data, and perform light fishing simulations at the historical migration locations;

[0028] Through light simulation, the number of catches is counted, fishing effect evaluation index data is set, and the number of catches is evaluated according to the fishing effect evaluation index data to obtain the fishing effect of each light fishing strategy.

[0029] Furthermore, in this method, a final light fishing strategy is formulated based on the fishing effect and migration route of each light fishing strategy, specifically including:

[0030] Obtain a light fishing strategy with a greater fishing effect than the preset fishing effect, and arrange a light fishing strategy with a greater fishing effect than the preset fishing effect in advance along the migration route to collect real-time fishing quantity data;

[0031] When the real-time fishing quantity data is greater than the preset fishing quantity data, the current light fishing strategy is stopped; when the real-time fishing quantity data is not greater than the preset fishing quantity data, the current light fishing strategy is maintained.

[0032] The second aspect of the present invention provides a light-purse seine fishing strategy optimization device based on multi-source data analysis, including a memory and a processor, wherein the memory includes a light-purse seine fishing strategy optimization method program based on multi-source data analysis, and when the light-purse seine fishing strategy optimization method program based on multi-source data analysis is executed by the processor, any step of the light-purse seine fishing strategy optimization method based on multi-source data analysis is implemented.

[0033] The third aspect of the present invention provides a computer-readable storage medium, including a program for a method for optimizing a light-purse seine fishing strategy based on multi-source data analysis. When the program for optimizing a light-purse seine fishing strategy based on multi-source data analysis is executed by a processor, the steps of the method for optimizing a light-purse seine fishing strategy based on multi-source data analysis are implemented.

[0034] The present invention solves the defects existing in the background technology and has the following beneficial effects:

[0035] The present invention obtains data on the current target species and historical environmental preference characteristics, constructs a species habitat prediction model based on these data, and then obtains multi-source data information for each sub-region within the target area. Combining this multi-source data information with the species habitat prediction model, the present invention constructs species migration locations, constructs migration routes based on these migration locations, and sets different light fishing strategies. The results of each light fishing strategy are simulated, and finally, a final light fishing strategy is formulated based on the results and migration routes of each strategy. By optimizing the migration locations of the target species based on their survival habits, the present invention can further optimize light fishing strategies and improve the efficiency of light seine fishing. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying any creative work, they can also obtain drawings of other embodiments based on these drawings.

[0037] Figure 1 The overall flow chart of the light purse seine fishing strategy optimization method based on multi-source data analysis is shown;

[0038] Figure 2 The device block diagram of the light purse seine fishing strategy optimization device based on multi-source data analysis is shown. DETAILED DESCRIPTION

[0039] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0040] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0041] like Figure 1 As shown, the first aspect of the present invention provides a light purse seine fishing strategy optimization method based on multi-source data analysis, comprising the following steps:

[0042] S102: Obtaining the current target fishery species population type data and historical environmental preference characteristic data, and constructing a species population habitat prediction model based on the current target fishery species population type data and historical environmental preference characteristic data;

[0043] S104: Acquire multi-source data information of each sub-region in the target region, and construct biological population migration location points by combining the multi-source data information and the biological population habitat prediction model;

[0044] S106: Constructing migration routes based on the migration locations of biological populations, setting different light fishing strategies, and simulating the fishing effects of each light fishing strategy;

[0045] S108: Formulate a final light fishing strategy based on the fishing effect and migration route of each light fishing strategy.

[0046] It should be noted that the present invention can further optimize the light fishing strategy and improve the fishing efficiency of the light seine by optimizing the migration location points of the target fishing biological population based on the survival habit data of the target biological population.

[0047] Furthermore, in this method, a biological population habitat prediction model is constructed based on the current target fishing species type data and historical environmental preference characteristic data, specifically:

[0048] A biological population habitat prediction model is constructed based on a deep neural network. The current target biological population type data and historical environmental preference characteristic data are used as graph nodes. The graph nodes are connected to construct a topological structure graph.

[0049] Introducing a cyclic spatial attention mechanism, inputting graph nodes into the cyclic spatial attention mechanism, focusing attention on the graph nodes, generating an attention feature map, and inputting the attention feature map into the hidden layer of the deep neural network;

[0050] Update the state of the hidden layer, save the model parameters of the biological population habitat prediction model, and output the biological population habitat prediction model.

[0051] It should be noted that environmental preference feature data include temperature, humidity, terrain structure and other data. By inputting graph nodes into the recurrent spatial attention mechanism, focusing attention on the graph nodes, generating an attention feature map, and inputting the attention feature map into the hidden layer of the deep neural network, and updating the state of the hidden layer, it is possible to suppress the interference of multi-scale data on model training, thereby optimizing the model training process and improving prediction accuracy.

[0052] Furthermore, in this method, multi-source data information of each sub-region in the target area is obtained, and the migration location points of biological populations are constructed by combining the multi-source data information and the biological population habitat prediction model, which specifically includes:

[0053] Acquire multi-source data information of each sub-region in the target area and data on the type of target fishing organism population, and input the multi-source data information of each sub-region in the target area and data on the type of target fishing organism population into a biological population habitat prediction model for prediction;

[0054] By prediction, a sub-area in the target area where the current target biological population lives is obtained, and the sub-area in the target area where the current target biological population lives is used as a migration location point of a biological population;

[0055] Obtain image data information of the migration location of each biological population through remote sensing technology, and calculate survival resource data based on the image data information of the migration location of each biological population, and set a survival resource data threshold;

[0056] The migration location of the biological population where the survival resource data is lower than the survival resource data threshold is regarded as a short-term migration location of the biological population, and the migration location of the biological population where the survival resource data is not lower than the survival resource data threshold is regarded as a long-term migration location of the biological population;

[0057] Biological population migration location points are constructed according to the short-term migration location points of the biological population and the long-term migration location points of the biological population, and the constructed biological population migration location points are output.

[0058] It should be noted that because different marine organisms have different predator-prey relationships, survival resource data includes the target prey animal and plant types. The migration locations of biological populations where survival resource data falls below the survival resource data threshold are used as temporary migration locations, while the migration locations of biological populations where survival resource data does not fall below the survival resource data threshold are used as long-term migration locations. This helps estimate the migration status of biological populations and improve the accuracy of migration route prediction. Multi-source data includes temperature, humidity, geographic location, terrain structure, and other data.

[0059] Furthermore, in this method, a migration route is constructed based on the migration locations of biological populations, specifically:

[0060] Obtain the migration route of the target biological population within a preset time through remote sensing technology, and obtain the survival habit data of the target biological population, and obtain the migration location point of the target biological population at the current time stamp based on the migration route of the target biological population within the preset time;

[0061] Obtaining, from the biological population migration position points, the biological population migration position points adjacent to the migration position point of the target biological population at the current time stamp, based on the biological population migration position points adjacent to the migration position point of the target biological population at the current time stamp;

[0062] Constructing an estimated migration route based on the migration location points of the biological populations adjacent to the migration location point of the target biological population at the current time stamp and the migration location point of the target biological population at the current time stamp;

[0063] When there is a temporary migration point of a biological population in the estimated migration route, the next adjacent biological population migration point is obtained, and the estimated migration route is updated based on the next adjacent biological population migration point;

[0064] When there is no temporary migration point of a biological population in the estimated migration route, the estimated migration route is maintained unchanged and is used as the final migration route.

[0065] It should be noted that when a species' transient migration point is included in the estimated migration route, the next adjacent species' migration point is obtained and the estimated migration route is updated based on this point. This allows for the estimation of transient stopover points, allowing for the pre-setting of multiple light-purse seine fishing locations and improving fishing efficiency. If no transient migration point is included in the estimated migration route, the estimated migration route is maintained and used as the final route, thus avoiding unnecessary effort.

[0066] Furthermore, in this method, different light fishing strategies are set to simulate the fishing effect of each light fishing strategy, specifically including:

[0067] Set different light net working parameters and working quantities at the historical migration locations along the target migration route, build several light fishing strategies based on the different light net working parameters and working data, and perform light fishing simulations at the historical migration locations;

[0068] Through light simulation, the number of catches is counted, the fishing effect evaluation index data is set, and the number of catches is evaluated based on the fishing effect evaluation index data to obtain the fishing effect of each light fishing strategy.

[0069] It should be noted that the operating parameters of light seines include the type of light, light intensity, etc. Fishing efficiency includes low efficiency, medium efficiency, high efficiency, etc.

[0070] Furthermore, in this method, a final light fishing strategy is formulated based on the fishing effect and migration route of each light fishing strategy, specifically including:

[0071] Obtain a light fishing strategy with a greater fishing effect than the preset fishing effect, and arrange a light fishing strategy with a greater fishing effect than the preset fishing effect in advance along the migration route to collect real-time fishing quantity data;

[0072] When the real-time fishing quantity data is greater than the preset fishing quantity data (such as can be set to medium efficiency), stop the current light fishing strategy; when the real-time fishing quantity data is not greater than the preset fishing quantity data, maintain the current light fishing strategy.

[0073] It should be noted that overfishing can be avoided by this method.

[0074] In addition, the method further comprises:

[0075] Obtain data on the natural enemy species corresponding to the target fishing species through big data, and obtain remote sensing image data information for each estimated migration location within a preset time through remote sensing technology;

[0076] Identify remote sensing image data information of each estimated migration location within a preset time using AI recognition technology, and determine whether the remote sensing image data information of each estimated migration location within the preset time contains natural enemy species data corresponding to the target fishing species;

[0077] When natural enemy type data corresponding to the target fishing organism type exists in the remote sensing image data information of the estimated migration location within a preset time, adjusting the migration route of the target fishing organism until the natural enemy type data corresponding to the target fishing organism type no longer exists;

[0078] When the natural enemy type data corresponding to the target fishing organism type does not exist in the remote sensing image data information of the estimated migration location within the preset time, the migration route of the target fishing organism is maintained unchanged.

[0079] It should be noted that when there are natural enemies along the migration route, according to the rules of the food chain, the target fishing organisms will escape or not choose the migration location, indicating that the migration route is not suitable for the target organisms. This method can further optimize the fishing strategy of the light seine.

[0080] like Figure 2 As shown, the second aspect of the present invention provides a light-purse seine fishing strategy optimization device 4 based on multi-source data analysis, including a memory 41 and a processor 42. The memory 41 includes a light-purse seine fishing strategy optimization method program based on multi-source data analysis. When the light-purse seine fishing strategy optimization method program based on multi-source data analysis is executed by the processor 42, the following steps are implemented:

[0081] Obtaining data on the current target species population type and historical environmental preference characteristics, and constructing a species population habitat prediction model based on the current target species population type data and historical environmental preference characteristics data;

[0082] Obtain multi-source data information for each sub-region in the target area, and construct the migration location points of biological populations by combining the multi-source data information with the biological population habitat prediction model;

[0083] Construct migration routes based on the migration locations of biological populations, set different light fishing strategies, and simulate the fishing effects of each light fishing strategy;

[0084] The final light fishing strategy is formulated based on the fishing effect of each light fishing strategy and the migration route.

[0085] It should be noted that the present invention can further optimize the light fishing strategy and improve the fishing efficiency of the light seine by optimizing the migration location points of the target fishing biological population based on the survival habit data of the target biological population.

[0086] Furthermore, in this device, a biological population habitat prediction model is constructed based on the current target fishing biological population type data and historical environmental preference characteristic data, specifically:

[0087] A biological population habitat prediction model is constructed based on a deep neural network. The current target biological population type data and historical environmental preference characteristic data are used as graph nodes. The graph nodes are connected to construct a topological structure graph.

[0088] Introducing a cyclic spatial attention mechanism, inputting graph nodes into the cyclic spatial attention mechanism, focusing attention on the graph nodes, generating an attention feature map, and inputting the attention feature map into the hidden layer of the deep neural network;

[0089] Update the state of the hidden layer, save the model parameters of the biological population habitat prediction model, and output the biological population habitat prediction model.

[0090] It should be noted that environmental preference feature data include temperature, humidity, terrain structure and other data. By inputting graph nodes into the recurrent spatial attention mechanism, focusing attention on the graph nodes, generating an attention feature map, and inputting the attention feature map into the hidden layer of the deep neural network, and updating the state of the hidden layer, it is possible to suppress the interference of multi-scale data on model training, thereby optimizing the model training process and improving prediction accuracy.

[0091] Furthermore, in this device, multi-source data information of each sub-region in the target area is obtained, and the migration location of the biological population is constructed by combining the multi-source data information and the biological population habitat prediction model, specifically including:

[0092] Acquire multi-source data information of each sub-region in the target area and data on the type of target fishing organism population, and input the multi-source data information of each sub-region in the target area and data on the type of target fishing organism population into a biological population habitat prediction model for prediction;

[0093] By prediction, a sub-area in the target area where the current target biological population lives is obtained, and the sub-area in the target area where the current target biological population lives is used as a migration location point of a biological population;

[0094] Obtain image data information of the migration location of each biological population through remote sensing technology, and calculate survival resource data based on the image data information of the migration location of each biological population, and set a survival resource data threshold;

[0095] The migration location of the biological population where the survival resource data is lower than the survival resource data threshold is regarded as a short-term migration location of the biological population, and the migration location of the biological population where the survival resource data is not lower than the survival resource data threshold is regarded as a long-term migration location of the biological population;

[0096] Biological population migration location points are constructed according to the short-term migration location points of the biological population and the long-term migration location points of the biological population, and the constructed biological population migration location points are output.

[0097] It should be noted that since different marine organisms have different predation relationships, the survival resource data include the predator type and predator plant type of the target fishing object. The migration location point of the biological population where the survival resource data is lower than the survival resource data threshold is used as a short-term migration location point of the biological population, and the migration location point of the biological population where the survival resource data is not lower than the survival resource data threshold is used as a long-term migration location point of the biological population, so as to estimate the migration situation of the biological population and improve the prediction accuracy of the migration route.

[0098] Furthermore, in this device, a migration route is constructed according to the migration location of the biological population, specifically:

[0099] Obtain the migration route of the target biological population within a preset time through remote sensing technology, and obtain the survival habit data of the target biological population, and obtain the migration location point of the target biological population at the current time stamp based on the migration route of the target biological population within the preset time;

[0100] Obtaining, from the biological population migration position points, the biological population migration position points adjacent to the migration position point of the target biological population at the current time stamp, based on the biological population migration position points adjacent to the migration position point of the target biological population at the current time stamp;

[0101] Constructing an estimated migration route based on the migration location points of the biological populations adjacent to the migration location point of the target biological population at the current time stamp and the migration location point of the target biological population at the current time stamp;

[0102] When there is a temporary migration point of a biological population in the estimated migration route, the next adjacent biological population migration point is obtained, and the estimated migration route is updated based on the next adjacent biological population migration point;

[0103] When there is no temporary migration point of a biological population in the estimated migration route, the estimated migration route is maintained unchanged and is used as the final migration route.

[0104] It should be noted that when a species' transient migration point is included in the estimated migration route, the next adjacent species' migration point is obtained and the estimated migration route is updated based on this point. This allows for the estimation of transient stopover points, allowing for the pre-setting of multiple light-purse seine fishing locations and improving fishing efficiency. If no transient migration point is included in the estimated migration route, the estimated migration route is maintained and used as the final route, thus avoiding unnecessary effort.

[0105] Furthermore, in this device, different light fishing strategies are set to simulate the fishing effect of each light fishing strategy, specifically including:

[0106] Set different light net working parameters and working quantities at the historical migration locations along the target migration route, build several light fishing strategies based on the different light net working parameters and working data, and perform light fishing simulations at the historical migration locations;

[0107] Through light simulation, the number of catches is counted, the fishing effect evaluation index data is set, and the number of catches is evaluated based on the fishing effect evaluation index data to obtain the fishing effect of each light fishing strategy.

[0108] It should be noted that the working parameters of the light fence include data such as the type of light and the light intensity.

[0109] Furthermore, in this device, a final light fishing strategy is formulated based on the fishing effect and migration route of each light fishing strategy, specifically including:

[0110] Obtain a light fishing strategy with a greater fishing effect than the preset fishing effect, and arrange a light fishing strategy with a greater fishing effect than the preset fishing effect in advance along the migration route to collect real-time fishing quantity data;

[0111] When the real-time fishing quantity data is greater than the preset fishing quantity data, the current light fishing strategy is stopped; when the real-time fishing quantity data is not greater than the preset fishing quantity data, the current light fishing strategy is maintained.

[0112] The third aspect of the present invention provides a computer-readable storage medium, including a program for a method for optimizing a light-purse seine fishing strategy based on multi-source data analysis. When the program for optimizing a light-purse seine fishing strategy based on multi-source data analysis is executed by a processor, the steps of the method for optimizing a light-purse seine fishing strategy based on multi-source data analysis are implemented.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0114] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0115] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0116] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0117] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0118] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A light seine fishing strategy optimization method based on multi-source data analysis, characterized by: The following steps are involved: Obtaining data on the type of biological populations currently targeted for fishing and historical environmental preference characteristics, and constructing a biological population habitat prediction model based on the data on the type of biological populations currently targeted for fishing and historical environmental preference characteristics; Acquire multi-source data information of each sub-region in the target region, and construct biological population migration location points by combining the multi-source data information and a biological population habitat prediction model; Constructing a migration route based on the migration location of the biological population, setting different light fishing strategies, and simulating the fishing effect of each light fishing strategy; The final light fishing strategy is formulated based on the fishing effect and migration route of each light fishing strategy.

2. The light purse seine fishing strategy optimization method based on multi-source data analysis according to claim 1 is characterized in that: A biological population habitat prediction model is constructed based on the biological population type data of the current target fishing target and the historical environmental preference characteristic data, specifically: Constructing a biological population habitat prediction model based on a deep neural network, using the biological population type data of the current target fishing target and the historical environmental preference characteristic data as graph nodes, connecting the graph nodes to construct a topological structure graph; Introducing a cyclic spatial attention mechanism, inputting the graph nodes into the cyclic spatial attention mechanism, focusing attention on the graph nodes, generating an attention feature map, and inputting the attention feature map into a hidden layer of a deep neural network; The state of the hidden layer is updated, the model parameters of the biological population habitat prediction model are saved, and the biological population habitat prediction model is output.

3. The light purse seine fishing strategy optimization method based on multi-source data analysis according to claim 1 is characterized in that: Acquiring multi-source data information for each sub-region in the target region, and constructing biological population migration locations based on the multi-source data information and the biological population habitat prediction model, specifically including: Acquiring multi-source data information of each sub-region in the target area and target fishing biological population type data, and inputting the multi-source data information of each sub-region in the target area and target fishing biological population type data into the biological population habitat prediction model for prediction; Obtaining, by prediction, a sub-region in the target area where the current target biological population inhabits, and using the sub-region in the target area where the current target biological population inhabits as a migration location point of the biological population; Acquire image data information of the migration location of each biological population through remote sensing technology, and calculate survival resource data based on the image data information of the migration location of each biological population, and set a survival resource data threshold; The migration location of the biological population where the survival resource data is lower than the survival resource data threshold is used as a short-term migration location of the biological population, and the migration location of the biological population where the survival resource data is not lower than the survival resource data threshold is used as a long-term migration location of the biological population; Biological population migration position points are constructed according to the short-term migration position points of the biological population and the long-term migration position points of the biological population, and the constructed biological population migration position points are output.

4. The light seine fishing strategy optimization method based on multi-source data analysis according to claim 1 is characterized in that: Construct a migration route based on the migration location points of the biological population, specifically: Obtaining the migration route of the target biological population within a preset time through remote sensing technology, and obtaining the survival habit data of the target biological population, and obtaining the migration location point of the target biological population at the current time stamp based on the migration route of the target biological population within the preset time; Obtaining, from the biological population migration point, biological population migration point adjacent to the migration point of the target biological population at the current time stamp, based on the biological population migration point adjacent to the migration point of the target biological population at the current time stamp; Constructing an estimated migration route based on the migration location points of the biological populations adjacent to the migration location point of the target biological population at the current time stamp and the migration location point of the target biological population at the current time stamp; When there is a temporary migration location point of a biological population in the estimated migration route, obtaining the next adjacent biological population migration location point, and updating the estimated migration route based on the next adjacent biological population migration location point; When there is no short-term migration point of the biological population in the estimated migration route, the estimated migration route is maintained unchanged and is used as the final migration route.

5. The light purse seine fishing strategy optimization method based on multi-source data analysis according to claim 1 is characterized in that: Set different light fishing strategies and simulate the fishing effects of each light fishing strategy, including: Set different light net working parameters and working quantities at the historical migration locations along the target migration route, build several light fishing strategies based on the different light net working parameters and working data, and perform light fishing simulations at the historical migration locations; Through light simulation, the number of catches is counted, fishing effect evaluation index data is set, and the number of catches is evaluated according to the fishing effect evaluation index data to obtain the fishing effect of each light fishing strategy.

6. The light seine fishing strategy optimization method based on multi-source data analysis according to claim 1 is characterized in that: Based on the fishing results and migration routes of each light fishing strategy, the final light fishing strategy is formulated, including: Obtain a light fishing strategy with a greater fishing effect than the preset fishing effect, and arrange a light fishing strategy with a greater fishing effect than the preset fishing effect in advance along the migration route to collect real-time fishing quantity data; When the real-time fishing quantity data is greater than the preset fishing quantity data, the current light fishing strategy is stopped; when the real-time fishing quantity data is not greater than the preset fishing quantity data, the current light fishing strategy is maintained.

7. A light seine fishing strategy optimization device based on multi-source data analysis, characterized in that: It includes a memory and a processor, wherein the memory includes a program for a light-purse fishing strategy optimization method based on multi-source data analysis. When the program for the light-purse fishing strategy optimization method based on multi-source data analysis is executed by the processor, the steps of the light-purse fishing strategy optimization method based on multi-source data analysis as described in any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium, characterized in that It includes a light-purse seine fishing strategy optimization method program based on multi-source data analysis. When the light-purse seine fishing strategy optimization method program based on multi-source data analysis is executed by a processor, it implements the steps of the light-purse seine fishing strategy optimization method based on multi-source data analysis as described in any one of claims 1 to 6.

Images