Fishing support device and method

Abstract

To provide a fishing support device that allows a user to simply determine a place with a high possibility of catching fish.SOLUTION: A fishing support device includes: a first recognition unit for recognizing an activity for each fishing method for aquatic animals in a predetermined region and a parameter group for defining an activity model expressing the activity by associating them with each other; an analysis unit for defining the activity model with the activity as an object variable and the parameter group as an explanatory variable by executing machine learning analysis or statistical analysis on the basis of respective values of the active and the parameter group that are associated and recognized by the first recognition unit; a second recognition unit for recognizing the parameter group on the basis of either one of an actual measurement value and a prediction result predicted by a physical model; and an evaluation unit for evaluating the activity according to the activity model defined by the analysis unit on the basis of the parameter group recognized by the second recognition unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a fishing support device and method. [Background technology]

[0002] Conventionally, when planning fishing or fishing, anglers and fishermen usually select fishing locations based on weather classifications such as clear, cloudy, rainy, etc., the type of fish they are targeting, the fishing method, etc. Therefore, anglers and fishermen obtain weather forecasts and information on past fishing results, and then select fishing locations by making their own judgments based on a comprehensive analysis of this information.

[0003] Patent Document 1 discloses a fishing information providing program and fishing information providing device that can determine how suitable an area (predetermined region) is for fishing and assist in finding a suitable place for fishing.

[0004] Such a device can provide the user with wind direction information indicating the wind direction and waterfront orientation information indicating the orientation of the terrain at the waterfront, thereby providing the user with information on how suitable a specified area is for fishing.

[0005] In addition, Patent Document 2 discloses a method for predicting fishing results regardless of the angler's experience or intuition. The technology disclosed in Patent Document 2 predicts fishing results by correlating environmental information about fishing spots where catchable aquatic organisms live with actual fishing results information at the fishing spots. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-059358 [Patent Document 2] Patent Publication No. 2021-011768 Summary of the Invention [Problem to be solved by the invention]

[0007] However, while Patent Document 1 includes wind direction information indicating wind direction and waterfront orientation information indicating the orientation of the waterfront topography, which are information indicating the ease of fishing for the user, it does not include any information indicating the activity level of aquatic animals. As a result, even if a location is easy for the user to fish, it may be difficult to catch aquatic animals in such a location. Generally, aquatic animals' activity level (e.g., feeding intention) changes depending on the environment (e.g., water temperature), so if such information cannot be recognized, the possibility of catching them decreases. Furthermore, when making a comprehensive judgment based on past fishing results and information such as weather information, the user must make a judgment based on a huge amount of data, and even if the user recognizes fishing results and weather information, making these judgments is cumbersome.

[0008] Furthermore, while Patent Document 2 accumulates fishing information, it is unable to determine whether or not aquatic life is actually present. Fishing methods include using nets in addition to fishing, but in these cases, it is difficult to make such a determination based on catch reports alone. Furthermore, there are multiple types of fishing, and users have difficulty determining which fishing method to use, resulting in a lack of convenience. Furthermore, Patent Document 2 requires the installation of measuring devices at each fishing spot, which poses a problem in that it is difficult to properly determine which locations within a vast area, such as the ocean, have a high probability of catching fish.

[0009] In order to solve the above problems, the present invention aims to provide a fishing support device and method that allows users to easily and accurately determine the location, time, and fishing method that are most likely to result in a catch. [Means for solving the problem]

[0010] In order to achieve the above object, the fishing support device according to the present invention is characterized by having the following invention-specific features. (1): In one aspect, the fishing support device of the present invention is a first recognition unit that recognizes the activity of aquatic animals in a predetermined area for each fishing method in association with a group of parameters that define an activity model that expresses the activity; an analysis unit that defines the activity model by performing machine learning analysis or statistical analysis based on the values of the activity and the parameter group recognized in association with each other by the first recognition unit, the activity being a response variable and the parameter group being an explanatory variable; a second recognition unit that recognizes the group of parameters based on either an actual measurement result or a predicted result predicted by a physical model; and an evaluation unit that evaluates the activity level in accordance with the activity model defined by the analysis unit based on the group of parameters recognized by the second recognition unit.

[0011] According to the fishery support device of (1), the activity of aquatic animals in a specified area is evaluated by correlating it with each value of a group of parameters that define an activity model that expresses the activity of the aquatic animals. Therefore, a user of the fishery support device can quantitatively recognize the relationship between the environmental, physical, and chemical parameters in the area (specified area) that is the target of fishing and the specified aquatic animals. Therefore, a user of the fishery support device can quantitatively select fishing locations based on the activity, and ultimately improve the efficiency of fishing.

[0012] Also, (2): In the fishing support device of the present invention, It is preferable that the parameter group includes a first parameter group consisting of at least one parameter based on current environmental data of a predetermined area, and a second parameter group consisting of a differential value of either a time derivative or a spatial derivative of at least one parameter of the first parameter group.

[0013] According to the fishery support device of (2), the parameter group includes parameters for a predetermined area and spatiotemporal derivatives of the parameters for the predetermined area. Therefore, the user of the fishery support device can quantitatively recognize not only the environmental, physical, and chemical parameters in the area (predetermined area) targeted for fishing, but also the relationship between the parameters and a predetermined aquatic animal according to the spatiotemporal fluctuations of the parameters, thereby more reliably quantifying the activity level. Therefore, the user of the fishery support device can more efficiently and quantitatively select fishing locations based on the activity level, thereby further improving the efficiency of fishing.

[0014] Also, (3): In the fishing support device of the present invention, The activity level is preferably the catch per unit effort (CPUE) of aquatic animals.

[0015] According to the fishery support device of (3), the activity level is based on CPUE. Therefore, even if the fishing equipment or fishing skill of the user of the fishery support device changes, the activity level can be quantitatively evaluated. Therefore, the user of the fishery support device can select fishing locations more efficiently and quantitatively based on the more precise activity level, and ultimately, the efficiency of fishing can be further improved.

[0016] Also, (4): In the fishing support device of the present invention, It is preferable that the first parameter includes at least one parameter selected from the current water temperature, salinity, air temperature, atmospheric pressure, wind speed, tide, solar radiation, wave height, wind direction, wave direction, ocean current speed, ocean current direction, chlorophyll concentration, dissolved oxygen content, and transparency in a specified area.

[0017] According to the fishing support device of (4), the first parameter is based on the current environmental data of the predetermined area. Therefore, the user of the fishing support device can quantitatively recognize the relationship between these parameters and the predetermined aquatic animals in the area (predetermined area) that is the target of fishing. Therefore, the user of the fishing support device can quantitatively select fishing locations based on the activity level, and ultimately improve the efficiency of fishing.

[0018] In order to achieve the above object, the fishing support device and method according to the present invention are characterized by having the following invention-specific features. (5): The fishing support device of the present invention is A fishing support device according to any one of (1) to (4), an information terminal device that acquires location information and time information of the predetermined area; The fishing support device evaluates the activity level of the specified area based on the location information and time information acquired by the information terminal.

[0019] In addition, in order to achieve the above object, the fishery support method of the present invention is characterized by having the following invention-specific features. (5): The fishery support method of the present invention is a first recognition step of associating and recognizing the activity level of aquatic animals in a predetermined area with a group of parameters that define an activity model that expresses the activity level; an analysis unit that defines the activity model by performing machine learning analysis or statistical analysis based on the values of the activity and the parameter group that are associated and recognized in the first recognition step, with the activity as a response variable and the parameter group as explanatory variables; a second recognition step of recognizing the parameter group based on either actual measurement results or prediction results predicted by a physical model; and a step of evaluating the activity level in accordance with the activity level model defined in the analysis step, based on the group of parameters recognized by the second recognition unit. [Effects of the Invention]

[0020] According to the present invention, a fishing support device and method can be provided that allow a user to easily and accurately determine the location, time, and fishing method that are most likely to result in a catch. [Brief explanation of the drawings]

[0021] [Figure 1] FIG. 1 is a schematic diagram showing an embodiment of a fishing support device of the present invention. [Figure 2] FIG. 2 is a flowchart showing an embodiment of the fishery support method of the present invention. [Figure 3] FIG. 3 is a flowchart showing an embodiment of a method for generating an activity model in the fishery support method of the present invention. [Figure 4] FIG. 4 is a diagram showing the importance of parameters that explain the activity level. [Figure 5] FIG. 5 is a diagram showing an ROC curve of the activity model. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, an embodiment of a fishing support device and method according to the present invention will be described with reference to the drawings.

[0023] (Configuration of fishing support device and method) Fig. 1 is a schematic diagram showing one embodiment of the present invention. As shown in Fig. 1, the fishery support system 1 of this embodiment includes a fishery support device 10 and an information terminal device 20. The fishery support device 10 and the information terminal device 20 are communicably connected via a network. The fishery support device 10 evaluates the activity level of aquatic animals in a specified area based on information acquired by the information terminal device 20.

[0024] The fishery support device 10 includes a first recognition unit 111, a second recognition unit 112, an analysis unit 120, an evaluation unit 130, a recording unit 140, and an information transmission / reception unit 150. Each of the first recognition unit 111, the second recognition unit 112, the analysis unit 120, the evaluation unit 130, the recording unit 140, and the information transmission / reception unit 150 is configured by a storage device (memory such as RAM, ROM, EEPROM, SSD, HDD, etc.) that stores and retains programs (software) and data, an arithmetic processing device (single-core processor, multi-core processor, CPU, etc.) that reads the necessary programs and / or data from the storage device and then executes predetermined arithmetic processing, an I / O circuit, etc.

[0025] The storage device stores and holds various data, as well as programs (software), which will be described later. The processor reads the necessary programs and data from the memory and executes arithmetic processing in accordance with the programs based on the data, thereby executing the arithmetic processing or tasks, as will be described later.

[0026] The information terminal device 20 includes a time-space information acquisition unit 210, an input interface 221, an output interface 222, a control unit 230, a recording unit 240, and an information transmission / reception unit 250. The information terminal device 20 includes, for example, equipment such as a personal computer, a mobile phone (smartphone), a home appliance, an electric reel, an electric lure, or a mobile object such as an electric bicycle.

[0027] The control unit 230 is composed of a processor (arithmetic processing unit), a memory (storage device), an I / O circuit, etc. The memory or a separate storage device stores and holds various data such as activity levels. The processor constituting the control unit 230 reads necessary programs and data from the memory, and executes the assigned arithmetic processing in accordance with the programs based on the data.

[0028] The fishery support device 10 may be mounted on the information terminal device 20. In this case, a software server (not shown) may transmit fishery support software to a processing device constituting the control unit 230 of the information terminal device 20, thereby imparting the functions of the fishery support device 10 to the processing device.

[0029] (Configuration of fishing support equipment) As described above, the fishery support device 10 includes the first recognition unit 111, the second recognition unit 112, the analysis unit 120, the evaluation unit 130, the recording unit 140, and the information transmission / reception unit 150. Here, "fishing" refers to any method of acquiring fish catches, and includes any fishing method that would normally be selected by a person skilled in the art, such as fishing, fixed net fishing, gill net fishing, purse seine fishing, and pole and net fishing.

[0030] The first recognition unit 111 recognizes the activity levels of aquatic animals in a predetermined area by associating them with a group of parameters that define an activity model that expresses the activity levels. Here, the "predetermined area" refers to a specific location, expressed by latitude, longitude, and water depth. "Aquatic animals" is a concept that includes not only fish, but also mollusks such as squid and octopus, and arthropods such as blue crabs. "Activity level" is a quantity that expresses the ease (possibility, probability) of catching an aquatic animal. Even for the same aquatic animal, the activity level will differ depending on the fishing method. For example, in fishing, this means that feeding behavior is reflected in activity level, while in gill nets and purse seines, movement amount is reflected in activity level.

[0031] For this reason, the quantity used to express activity is preferably the catch per unit effort (CPUE) of an aquatic animal. By using CPUE, it is possible to express the activity of each aquatic animal and also to evaluate the activity of each fishing method. However, in this embodiment, activity can also be defined using a simple catch amount in addition to CPUE. In this case, for example, the number of catches or the weight of catches can be defined as activity.

[0032] In this embodiment, the "activity model" refers to a function with variables that are derived from a set of parameters derived by machine learning or statistical techniques, as described below. Examples of such a function include a multiple regression model, a decision tree model, and a deep learning model (such as a convolutional neural network). Other models and techniques that can be selected by those skilled in the art may also be used. The "set of parameters" refers to environmental data that affect the behavior of aquatic animals. Examples of such parameters include the current water temperature, salinity, air temperature, atmospheric pressure, wind speed, tides, solar radiation, wave height, wind direction, wave direction, hydrosphere flow velocity, hydrosphere flow direction, chlorophyll concentration, dissolved oxygen content, and transparency in a given area.

[0033] In addition, in this embodiment, it is preferable to further include a time derivative or a spatial derivative of the parameter group in order to take into account the influence of spatiotemporal environmental changes on aquatic animals. These time derivatives or spatial derivatives are not limited to first-order derivatives, but may also employ n-th-order derivative values (n is an integer equal to or greater than 2). Furthermore, not only a single spatial derivative but also a product of multiple derivative values may be included. For example, since there are countless vortices in the hydrosphere, the rotation of the velocity field can be employed as one of the parameters in order to consider the influence of these vortices on aquatic animals.

[0034] The analysis unit 120 defines an activity model using the activity as a response variable and the parameter group as an explanatory variable by performing machine learning analysis or statistical analysis based on the values of the activity and the parameter group recognized in association with each other by the first recognition unit 111. Here, it is preferable to use decision tree analysis as the machine learning or statistical analysis method, but this is not limitative, and for example, multiple regression analysis, deep learning analysis, etc. can also be used. In addition, by generating the obtained parameters as multiple two-dimensional arrays, a generative adversarial network (GAN, etc.) can also be used.

[0035] The second recognition unit 112 recognizes the set of parameters based on either actual measurement results for the predetermined area or prediction results based on a physical model. When a user or the like starts fishing, information is transmitted to the second recognition unit 112 according to the location information of the predetermined area and the current time acquired by the spatiotemporal information acquisition unit 210 of the information terminal device 20 (described later). The second recognition unit 112 acquires the set of parameters based on the location information and time information for the predetermined area. In this embodiment, even if the user is not present in the predetermined area, the information may be transmitted to the second recognition unit 112 by specifying the location and time of the predetermined area in advance.

[0036] The second recognition unit 112 acquires a set of parameters for the predetermined area based on the information. In this case, actual measured values for the predetermined area may be used, or predicted values from a physical model may be used as the set of parameters. Here, "measurement results" may refer to, for example, measurement data acquired by the device itself, or measurement data acquired by a national government, local government, or private organization through fixed observations, moored observations, satellite observations, or floating observations such as "ARGO floats." Here, "physical models" refer to, for example, forecast data issued by a national government, local government, or private organization, or forecast data obtained by machine learning analysis or statistical analysis of past or current measurement results.

[0037] The evaluation unit 130 evaluates the activity level according to the activity model defined by the analysis unit 120, based on the group of parameters recognized by the second recognition unit 112. This allows the user to predict the activity level of aquatic animals in a predetermined area.

[0038] The recording unit 140 records the activity model defined by the analysis unit 120 for each predetermined area and each aquatic animal. The information transmitting / receiving unit 150 transmits and receives information to and from the information terminal device 20 or other network services via the network, as will be described later.

[0039] (Configuration of information terminal device) As described above, the information terminal device 20 includes the time-space information acquisition unit 210 , the input interface 221 , the output interface 222 , the control unit 230 , the recording unit 240 , and the information transmission / reception unit 250 .

[0040] The spatiotemporal information acquisition unit 210 automatically or manually acquires location information or time information of a user, etc. At this time, the spatiotemporal information acquisition unit 210 may acquire the time information in any unit of month, day, hour, or second. Furthermore, the spatiotemporal information acquisition unit 210 may acquire the spatial information in any unit of degree, hour, or second for latitude and longitude, and may acquire the depth in meters or centimeters from a reference surface such as a geoid surface, mean sea level, or sea level taking tides into account at the latitude and longitude.

[0041] The time-space information acquisition unit 210 automatically acquires the location information of the user, etc., for example, by using GPS from an artificial satellite. Alternatively, any GNSS method may be used. The time-space information acquisition unit 210 automatically acquires the time information of the user, etc., for example, by acquiring the time from a radio-controlled clock at the current time.

[0042] The time-space information acquisition unit 210 manually acquiring the position information of the user, etc. means, for example, that the user, etc. inputs an arbitrary point as text information, that the user, etc. directly inputs latitude, longitude, and depth, etc. The time-space information acquisition unit 210 manually acquiring the time information of the user, etc. means, for example, that the user, etc. inputs an arbitrary time, etc.

[0043] The input interface 221 is used, for example, for a user to input spatiotemporal information. The input interface 221 may be an audio input device, a keypad, a smartphone screen, or any other existing method. The output interface 222 is used, for example, for a user to recognize an evaluation value. The output interface 222 may be an existing method, for example, a screen, an audio output device, or any other existing method.

[0044] The recording unit 240 records location information, aquatic animal information, or time information of a predetermined area where the user has previously visited, or location information, aquatic animal information, or time information of a predetermined area where the user frequently visits, etc. The information transmitting / receiving unit 250 transmits and receives information to and from the fishing support device 10 or other network services via a network, as described above.

[0045] (Fishing support method using fishing support device) Next, the fishery support method of the present invention will be described with reference to the drawings. FIG. 2 is a flowchart showing an embodiment of the fishery support method according to the present invention.

[0046] First, the user installs the application or software on his / her own information terminal device 20 via a network, a hard drive, etc. This allows the user to record his / her own catch information and spatiotemporal information in the recording unit 240 through the application, etc., and to check past excess information and spatiotemporal information, as well as environmental information or activity level information of specific organisms linked to the spatiotemporal information (described later), through the output interface 222.

[0047] As shown in FIG. 2, in this embodiment, the user first launches an application or the like installed on the information terminal device 20. The application includes a planned fishing start button, and the user inputs location information and time information for the location where the user will be fishing, thereby performing an input action such as tapping the planned fishing start button (STEP 10). At this time, a start fishing button or the like may be provided in addition to the planned fishing start button, and the user may input location information and time information for the location where the user will actually be fishing, thereby assuming a virtual start fishing button operation. Alternatively, the user may input location information and time information for a location where the user has previously fished, thereby assuming a virtual start fishing button operation.

[0048] After the fishing start plan button is input in STEP 10, the information terminal device 20 uses the time-space information acquisition unit 210 to acquire location information and time information of the location where the fishing start plan button was input (STEP 11). At this time, if the time-space information acquisition unit 210 cannot acquire location information or time information, the user may directly input location information or time information through the input interface 221.

[0049] After the position information and time information at the start of fishing are acquired in STEP 11, the position information and time information are then transmitted by the information transmitting / receiving unit 250 to the fishery support device 10 (STEP 12). After that, the information transmitting / receiving unit 150 of the fishery support device 10 receives the position information and time information (STEP 20).

[0050] When the fishery support device 10 receives the position information and time information by the information transmitting / receiving unit 150 in STEP 20, the second recognition unit 112 acquires a parameter group in the position information and time information (STEP 21).

[0051] In STEP 21, a set of parameters based on the location information and time information is acquired. Then, the evaluation unit 140 evaluates the activity level of the aquatic animals according to an activity model defined by the analysis unit 120 (described later) (STEP 22). While the evaluation unit 140 typically evaluates the activity level of all aquatic animals, the user may select an aquatic animal or a type of fishing in STEP 10. The fishery support device 10 receives the selected information and evaluates the activity level of each aquatic animal or each type of fishing based on the selected information. For example, if the user selects the order Perciformes, the activity level of all aquatic animals in the order Perciformes may be evaluated. If the user selects the order Carangidae, the activity level of all aquatic animals in the order Carangidae may be evaluated, which is more detailed than the order Perciformes. If the user selects yellowtail, the activity level of yellowtail may be evaluated. For example, if the user selects lure fishing, the activity level may be evaluated based on a high probability of catching a fish with a lure. If the user selects krill fishing, the activity level may be evaluated based on a high probability of catching a fish with krill bait. Also, if the user selects yellowtail as the aquatic animal and krill fishing as the type of fishing, the activity level may be evaluated as a state in which there is a high probability of catching a yellowtail using krill as bait.

[0052] When the activity level of the aquatic animal is evaluated in STEP 22, the fishery support device 10 transmits the activity level to the information terminal device 20 via the information transmitting / receiving unit 150 (STEP 23). When the information transmitting / receiving unit 250 of the information terminal device 20 receives the activity level (STEP 13), the activity level is output to the output interface 222 (STEP 14). In addition to evaluating the activity level of the location information and time information acquired by the user at this time, the activity levels of nearby location information, recent time information, nearby aquatic animal species (e.g., amberjack when yellowtail is selected), nearby fishing method (e.g., sabiki fishing when krill fishing is selected), etc. may be calculated in STEP 22, and the activity level may be output to the output interface 222 together with the original activity level information according to the activity level value.

[0053] As a result, users can efficiently evaluate activity levels and appropriately select fishing locations and dates.

[0054] (Activity evaluation method) Next, the activity evaluation method of the present invention will be described with reference to the drawings. FIG. 3 is a flowchart illustrating one embodiment of the activity evaluation method of the present invention. In this embodiment, the user first launches an application or the like installed on the information terminal device 20. The application is provided with a fishing start button, and the user performs an input operation such as tapping the fishing start button (STEP 30). At this time, the user may input location information and time information of a past fishing location, which can serve as a substitute for the operation of virtually pressing the fishing start button.

[0055] After the fishing start button is pressed in STEP 30, the information terminal device 20 uses the time-space information acquisition unit 210 to acquire location information and time information about the location where the fishing start button was pressed (STEP 31). If the time-space information acquisition unit 210 cannot acquire the location information or time information, the user may directly input the location information or time information through the input interface 221.

[0056] In STEP 30, the user starts fishing at the location after pressing the fishing start plan button. At this time, the user can select the type of fishing in advance or after starting fishing.

[0057] After starting fishing, the user will catch a number of aquatic animals. At this time, the user inputs information about the aquatic animals they caught (aquatic animal information) through the input interface 221 of the information terminal device 20 (STEP 31). Here, the aquatic animal information includes biological information (including kingdom, phylum, class, order, family, genus, and species), weight, number of catches, body length (including body height, etc.), time and location information at the time of capture, and time required for capture. When inputting the aquatic animal information through the input interface 221, it is preferable for the user to directly input it using a keyboard, but for example, an imaging device or an audio input device may also be used. When an imaging device is used for the input interface 221, the captured aquatic animals may be photographed, and the captured images may be used to estimate and obtain biological information (including kingdom, phylum, class, order, family, genus, and species), weight, number of catches, size, body length (including body height, etc.), time required for capture, etc. Furthermore, if the user is unable to recognize accurate aquatic animal information, the user may obtain the aquatic animal information by selecting one option from a plurality of options in advance.

[0058] When the user has finished fishing, the user performs an input action such as tapping the fishing end button included in the application (STEP 33). If the user forgets to tap the fishing end button at this time, the user may input location information and time information of a previous fishing location, which can be used as a substitute for the operation of virtually tapping the fishing end button.

[0059] After the fishing end button is input in STEP 33, the information terminal device 20 causes the time-space information acquisition unit 210 to acquire the location information and time information of the location where the fishing end button was input (STEP 34). At this time, if the time-space information acquisition unit 210 cannot acquire the location information or time information, the user may directly input the location information or time information through the input interface 221.

[0060] After the position information and time information after the start of fishing are acquired in STEP 34, the position information and time information at the start of fishing, aquatic animal information, and position information and time information at the end of fishing are transmitted to the fishery support device 10 by the information transmitting / receiving unit 250 (STEP 35).Then, the information transmitting / receiving unit 150 of the fishery support device 10 receives the information (STEP 40).

[0061] In STEP 40, when the fishery support device 10 receives the information via the information transmission / reception unit 150, the second recognition unit 112 acquires a group of parameters for the location information and time information at the start of the fishing, when the fish is caught, and when the fishing ends (STEP 41).

[0062] In STEP 41, a group of parameters based on the information is acquired, and then pre-processing for activity model analysis is performed (STEP 42). Pre-processing for activity model analysis is a screening process that removes information that cannot be clearly measured, such as the specific activity or location information derived from the aquatic animal information and time information.

[0063] Here, identifying activity levels means calculating CPUE or any quantity calculated from multiple aquatic animal information and parameter groups. For example, the product of the number of catches and the time required for catching them. Activity levels can also be defined by setting a threshold for the number of catches, amount of catch, etc., and performing clustering such that when the threshold is met, there is activity (fish can be caught), and when it is met, there is no activity (fish cannot be caught).

[0064] After the activity model analysis preprocessing is performed in STEP 42, an activity model is then defined based on the activity and the parameter group (STEP 43). Here, the activity model used is, for example, an activity model created by using a decision tree or light BGM to identify the influence of each parameter constituting the parameter group as an explanatory variable that affects the activity as the objective variable, based on the activity and the parameter group. Furthermore, machine learning or deep learning techniques such as CNN or GAN may be used to define the activity model.

[0065] The above describes one embodiment of the fishing support device and fishing support system according to the present invention, along with the accompanying drawings. However, the present invention is not limited to the above embodiment and may be modified within the scope of ordinary skill in the art. [Example]

[0066] According to the above embodiment, an activity model was defined based on the Wright GBM. Here, the activity level was determined as the objective variable by clustering the number of sea bass (standard Japanese names: Marusuzuki and Hirasuzuki) caught by a person during a single fishing trip in a certain area (Tokyo Bay in this example). That is, the activity level was determined by using the number of sea bass caught per fishing trip to represent the probability of catching a sea bass in Tokyo Bay at any given date and time. The clustering data was for 2019. Note that, by using a different activity level, it would also be possible to display information about what fish can be caught in Tokyo Bay tomorrow. For example, whether or not one or more sea bass were caught could be used as the activity level for clustering. In this case, zero data can be used, allowing users to build an intuitive and easy-to-understand model.

[0067] In addition, when defining the activity model based on the Wright GBM, the following parameters were used as explanatory variables: water temperature (JcopeT), tidal range between high and low tide (trange), wind speed (x1), moon phase (moonage), area (area), latitude (lat), date and time (time), longitude (lon), temperature (airt), wind direction (windd), tide, month (month), and location (place). Note that location represents the granularity by prefecture, and area represents the granularity by city, ward, town, or village. Tides are classified into spring tide, neap tide, neap tide, and young tide.

[0068] Figure 4 shows the importance of the parameters that explain the activity. Referring to Figure 4, it is suggested that the most influential factors are water temperature, the difference in tidal levels between high and low tide, and wind speed, in that order. It is also suggested that location and month are not particularly related to the catch. In other words, it has become clear that water temperature is an important parameter for catching sea bass.

[0069] Figure 5 shows the ROC curve of this activity model. As shown in Figure 5, the AUC (area under an ROC curve) was 0.83. This is thought to be due to the small amount of training data used in the light GBM. However, accuracy can be improved by increasing the number of data or the number of parameter sets used as explanatory variables. This can be improved by increasing the number of catch data.

[0070] An activity model like the one described above is created. Therefore, by inserting a new set of parameters into the activity model, the user can easily determine the activity level (i.e., whether or not sea bass can be caught) based on the parameters, and can determine whether or not to go fishing or operate a fishing business based on the predicted catch or catch volume. [Advantages and effects of this embodiment]

[0071] With the fishing support system of the present embodiment as described above, the user can accurately measure information such as the ease of fishing with just a few simple operations (tapping a button to start, taking a photo when fishing, and tapping a button to finish). Furthermore, with the fishing support system configured as described above, information on when fishing is unsuccessful can also be used for analysis, making it possible to provide the user with accurate activity levels. [Explanation of symbols]

[0072] 1·· Fisheries support system, 10·· Fisheries support device, 20·· Information terminal device, 111·· First recognition unit, 112·· Second recognition unit, 120·· Analysis unit, 130·· Evaluation unit, 140·· Recording unit, 150·· Information transmission and reception unit (fisheries support device), 210·· Spatiotemporal information acquisition unit (fisheries support device), 221·· Input interface, 222·· Output interface, 230·· Control unit, 240·· Recording unit (information terminal device), 250·· Information transmission and reception unit (information terminal device).

Claims

1. a first recognition unit that recognizes the activity of aquatic animals in a predetermined area for each fishing method in association with a group of parameters that define an activity model that expresses the activity; an analysis unit that defines the activity model by performing machine learning analysis or statistical analysis based on the values of the activity and the parameter group recognized in association with each other by the first recognition unit, the activity being a response variable and the parameter group being an explanatory variable; a second recognition unit that recognizes the parameter group based on either an actual measurement result or a predicted result predicted by a physical model; an evaluation unit that evaluates the activity level according to the activity level model defined by the analysis unit based on the parameter group recognized by the second recognition unit, Fishery support equipment.

2. The fishing support device according to claim 1, The parameter group includes a first parameter group consisting of at least one parameter based on current environmental data of a predetermined area, and a second parameter group consisting of a differential value of either a time differential or a spatial differential of at least one parameter of the first parameter group. Fishery support equipment.

3. The fishing support device according to claim 1, The activity is the catch per unit effort (CPUE) of aquatic animals. Fishery support equipment.

4. The fishing support device according to claim 2, The first parameter includes at least one parameter selected from the current water temperature, salinity, air temperature, atmospheric pressure, wind speed, tide, solar radiation, wave height, wind direction, wave direction, ocean current speed, ocean current direction, chlorophyll concentration, dissolved oxygen content, and transparency of the predetermined area. Fishery support equipment.

5. A fishing support device according to any one of claims 1 to 4; an information terminal device that acquires location information and time information of the predetermined area; The fishing support device evaluates the activity level of the predetermined area based on the position information and time information acquired by the information terminal. Fisheries support system.

6. a first recognition step of associating and recognizing the activity level of aquatic animals in a predetermined area with a group of parameters that define an activity model that expresses the activity level; an analysis unit that defines the activity model by performing machine learning analysis or statistical analysis based on the activity and the parameter group values associated and recognized in the first recognition step, with the activity as a response variable and the parameter group as explanatory variables; a second recognition step of recognizing the parameter group based on either actual measurement results or prediction results predicted by a physical model; and evaluating the activity level in accordance with the activity level model defined in the analysis step based on the parameter group recognized by the second recognition unit. How to support fisheries.

Images