Growth prediction device, growth prediction method, and growth prediction program

Abstract

This growth prediction device comprises: a maximum value estimation unit that, from environment information, cultivation information, and index value measurement data of an index value that changes in conjunction with the growth of a rice plant, estimates the maximum value of the index value; a normalization unit that normalizes the index value measurement data from the measurement data index value and the maximum value of the index value and calculates normalized measurement data; and a prediction unit that, by using a trained model generated by machine learning in which normalized measurement data of the index values obtained in advance is used as training data, predicts the future transition of the index value from the calculated normalized measurement data of the index values.

Description

[Technical Field]

[0001] The disclosed technology relates to a growth prediction device, a growth prediction method, and a growth prediction program. [Background technology]

[0002] Due to the decline in the agricultural population, there is a demand for new farmers. To enable new farmers to perform farm work in the same way as experienced farmers, a system to support farm work (hereinafter referred to as "farm work support system") has been launched. In order to perform the appropriate farm work at the appropriate time, the farm work support system needs to have the ability to predict and suggest farm work times.

[0003] When it comes to paddy rice, Japan's staple food, fertilization, which is carried out to improve yields, is considered best to be carried out approximately 15 to 20 days before heading. Fertilization refers to the application of fertilizer. Furthermore, mowing at least one week before heading is considered effective in preventing damage from spotted rice stink bugs. Many paddy rice farming practices (top dressing, pesticide application, etc.) are most effective when carried out around the heading date. Therefore, predicting the heading date or the growth stage (e.g., panicle differentiation, flag leaf formation) in advance is important for supporting paddy rice farming practices.

[0004] SIMRIW (Simulation Model for Rice-Weather Relationships) is a well-known technology related to rice growth models (see, for example, Non-Patent Document 1). SIMRIW is a highly reliable rice growth model that is widely used in Japan and has been adopted in the cultivation management support system provided by the National Agriculture and Food Research Organization (commonly known as NARO).

[0005] SIMRIW can calculate the heading date and optimum harvest time by calculating the rice growth index (DVI) from meteorological data and variety information, as shown in Figure 9. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Takeshi Horie, Hiroshi Nakagawa, "Studies on modeling and prediction of rice developmental processes", Journal of the Crop Science Society of Japan, 59(4): 687-695. (1990) Summary of the Invention [Problem to be solved by the invention]

[0007] However, because SIMRIW only uses weather data and variety information as input, it does not take into account human errors that occur in actual farming, resulting in discrepancies between actual growth conditions and calculated growth conditions.

[0008] In addition, the parameters for calculating DVI differ for each variety, and in order to obtain correct data, data must be collected for each variety, requiring a huge amount of data.

[0009] The Leaf Area Index (hereinafter referred to as "LAI") is a measurable parameter that changes as rice grows. This LAI represents the ratio of rice leaf area to a unit area of ​​the ground. As shown in Figure 10, for example, this LAI is an index of growth status that increases as the rice grows, and reaches its peak when the flag leaf appears, regardless of the variety. By predicting the peak timing, it is possible to determine 10 days before the heading date. This is expected to make it possible to plan agricultural work.

[0010] Like DVI, the maximum value of LAI also differs for each variety, and also differs depending on agricultural conditions such as weather conditions and whether or not fertilizer is applied. For example, as shown in Figure 11, the transition of LAI differs depending on whether or not fertilizer is applied.

[0011] In other words, when predicting the growth status of rice using LAI etc., it is necessary to do so for each variety, and therefore a prediction model must be built for each variety, which makes it impossible to predict the growth of new varieties.

[0012] The disclosed technology has been made in consideration of the above points, and aims to provide a growth prediction device, a growth prediction method, and a growth prediction program that can predict the growth status of rice without distinguishing between rice varieties. [Means for solving the problem]

[0013] A first aspect of the present disclosure is a growth prediction device comprising: a data input unit that accepts input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information related to rice cultivation, and environmental information related to the environment in which the rice is cultivated; a maximum value estimation unit that estimates a maximum value of the index value from the measurement data of the index values, the cultivation information, and the environmental information; a normalization unit that normalizes the measurement data of the index values ​​from the measurement data of the index values ​​and the maximum value of the index values ​​to calculate normalized measurement data; and a prediction unit that predicts future trends in the index values ​​from the normalized measurement data of the index values ​​calculated by the normalization unit using a trained model that is generated by machine learning using the normalized measurement data of the index values ​​obtained in advance as training data, and that takes the normalized measurement data of the index values ​​as input and outputs prediction data that represents future trends in the index values.

[0014] A second aspect of the present disclosure is a growth prediction method that receives input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information related to rice cultivation, and environmental information related to the environment in which the rice is cultivated, estimates the maximum value of the index value from the measurement data of the index value, the cultivation information, and the environmental information, normalizes the measurement data of the index value from the measurement data of the index value and the maximum value of the index value to calculate normalized measurement data, and predicts future changes in the index value from the calculated normalized measurement data of the index value using a trained model that is generated by machine learning using the normalized measurement data of the index value obtained in advance as training data, and that takes the normalized measurement data of the index value as input and outputs prediction data that represents future changes in the index value.

[0015] A third aspect of the present disclosure is a growth prediction program that causes a computer to execute the following steps: receive input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information related to rice cultivation, and environmental information related to the environment in which the rice is cultivated; estimate a maximum value of the index value from the measurement data of the index value, the cultivation information, and the environmental information; normalize the measurement data of the index value from the measurement data of the index value and the maximum value of the index value to calculate normalized measurement data; and predict future trends in the index value from the calculated normalized measurement data of the index value using a trained model that is generated by machine learning using the normalized measurement data of the index values ​​obtained in advance as training data, and that takes the normalized measurement data of the index values ​​as input and outputs prediction data that represents future trends in the index values. [Effects of the Invention]

[0016] The disclosed technology has the effect of making it possible to predict the growth status of rice without distinguishing between rice varieties. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a diagram illustrating an example of a configuration of a farm work support system according to an embodiment. [Figure 2] 1 is a block diagram showing an example of a hardware configuration of a growth prediction device according to an embodiment. [Figure 3] 1 is a block diagram showing an example of a functional configuration of a growth prediction device according to an embodiment. [Figure 4] FIG. 10 is a diagram illustrating an example of a plurality of variables included in environmental information according to the embodiment. [Figure 5] 10 is a graph showing an example of a relationship between actual measurement data and predicted data of LAI according to the embodiment. [Figure 6A] 10 is a graph showing an example of LAI when LAI is predicted using all environmental information. [Figure 6B] 10 is a graph showing an example of LAI when LAI is predicted using environmental information selected according to its contribution rate to LAI. [Figure 7] 10 is a flowchart showing an example of the flow of a learning process by the growth prediction program according to the embodiment. [Figure 8] 10 is a flowchart showing an example of the flow of a prediction process by a growth prediction program according to an embodiment. [Figure 9] FIG. 1 is a diagram illustrating the prior art. [Figure 10] FIG. 1 shows the relationship between LAI and heading date. [Figure 11] FIG. 1 shows the difference in LAI transition depending on whether fertilization is performed or not. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of the disclosed technology will be described below with reference to the drawings. Note that in each drawing, the same or equivalent components and parts are given the same reference numerals. Also, the dimensional proportions in the drawings are exaggerated for the convenience of explanation and may differ from the actual proportions.

[0019] The growth prediction device according to this embodiment provides specific improvements over conventional methods for predicting the growth status of each rice variety, and represents an advancement in the field of rice growth prediction.

[0020] The growth prediction device according to this embodiment reflects the current growth status of rice plants and predicts the growth status using LAI, which is an example of an index that can be measured actually, and determines the timing of agricultural work based on the predicted growth stage, thereby providing support for agricultural work. In other words, by learning time-series measurement data of LAI and predicting future trends in LAI, the device determines the time approximately 10 days before heading (when flag leaves emerge) and predicts rice growth.

[0021] In the growth prediction device according to this embodiment, when performing growth prediction using LAI, the maximum LAI value is used to normalize the LAI measurement data obtained by actual measurement to calculate normalized measurement data. This normalized measurement data is used for machine learning to generate a prediction model, which is then used to predict future trends in LAI. This allows predictions to be made using the normalized LAI measurement data as training data, regardless of rice variety. This makes it possible to predict growth even for new varieties.

[0022] FIG. 1 is a diagram showing an example of the configuration of a farm work support system 100 according to this embodiment.

[0023] The agricultural work support system 100 shown in FIG. 1 includes a growth prediction device 10, an LAI measurement sensor 20, a database 30, a farm work instruction unit 40, and a user terminal 50. The growth prediction device 10 is connected to each of the LAI measurement sensor 20, the database 30, and the farm work instruction unit 40. The farm work instruction unit 40 is connected to the user terminal 50. The database 30 may be stored in an external storage device or may be stored in the growth prediction device 10. Furthermore, the farm work instruction unit 40 may be realized as a function of an information processing device different from the growth prediction device 10, or may be realized as a function of the growth prediction device 10.

[0024] The LAI measurement sensor 20 is a sensor that measures the LAI, an example of an index value that changes with the growth of rice, from an image obtained by actually photographing rice plants. The LAI measurement sensor 20 is assumed to be capable of measuring weather (temperature, etc.), soil solution concentration, etc., and capable of measuring the LAI. The LAI may be estimated using other methods, not just a method of estimating it from an image. It is conceivable that the types of sensors installed in each field will vary from field to field. For example, in field A, water temperature, air temperature, and humidity are acquired by sensors, while in field B, water temperature, EC (electrical conductivity), air temperature, and humidity are acquired by sensors. In other words, the items that can be acquired from the sensors may differ in each field, and items that cannot be acquired may be acquired from the database 30.

[0025] The database 30 stores cultivation information related to rice cultivation and environmental information related to the environment in which the rice is cultivated. The database 30 associates the LAI measurement data and environmental information acquired from the LAI measurement sensor 20 with dates and stores the LAI measurement data and environmental information as time-series data. The time-series LAI measurement data (hereinafter simply referred to as "LAI measurement data") is also stored in association with the cultivation information. The cultivation information includes multiple variables related to rice cultivation. The multiple variables related to rice cultivation include, for example, the transplanting date, variety, timing of top dressing for the variety, and whether or not fertilization was performed (fertilization or no fertilization). The environmental information also includes multiple variables related to the environment in which the rice is cultivated. The environmental information includes, for example, variables related to the weather during the rice cultivation period and variables related to the soil in which the rice is cultivated. The weather-related variables include temperature, solar radiation, humidity, etc., and the soil-related variables include pH (hydrogen ion exponent) and soil solution concentration (ion concentration), etc.

[0026] The growth prediction device 10 acquires LAI measurement data, cultivation information, and environmental information from the database 30. The growth prediction device 10 may acquire this information periodically or in response to a user request. The LAI measurement data may be acquired directly from the LAI measurement sensor 20. The growth prediction device 10 determines the day when the LAI will peak and the maximum value of the LAI from the acquired LAI measurement data, cultivation information, and environmental information, and outputs these to the farm work instruction unit 40. The day when the LAI will peak and the maximum value of the LAI are used as data representing the future growth status.

[0027] When the agricultural work instruction unit 40 obtains the day when the LAI will peak and the maximum value of the LAI from the growth prediction device 10, it determines the need for additional fertilization and the timing of agricultural work based on the obtained day when the LAI will peak and the maximum value of the LAI, and outputs the determination result to the user terminal 50.

[0028] The user terminal 50 is a terminal device used by the user, a farmer, and may be, for example, a personal computer (PC), a smartphone, etc. The user terminal 50 displays the determination results from the farm work instruction unit 40 and instructs the user on the need for additional fertilization and the timing of farm work.

[0029] Next, with reference to FIG. 2, the hardware configuration of the growth prediction device 10 according to this embodiment will be described.

[0030] FIG. 2 is a block diagram showing an example of the hardware configuration of the growth prediction device 10 according to this embodiment.

[0031] 2, the growth prediction device 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and a communication interface (I / F) 17. Each component is connected to each other via a bus 18 so as to be able to communicate with each other.

[0032] The CPU 11 is a central processing unit that executes various programs and controls each component. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a work area. The CPU 11 controls each of the above components and performs various arithmetic processing in accordance with the program stored in the ROM 12 or the storage 14. In this embodiment, the ROM 12 or the storage 14 stores a growth prediction program for executing a growth prediction process.

[0033] The ROM 12 stores various programs and various data. The RAM 13 temporarily stores programs or data as a working area. The storage 14 is configured with an HDD (Hard Disk Drive) or SSD (Solid State Drive) and stores various programs including the operating system and various data.

[0034] The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used to input various information to the device itself.

[0035] The display unit 16 is, for example, a liquid crystal display, and displays various information. The display unit 16 may also function as the input unit 15 by adopting a touch panel system.

[0036] The communication interface 17 is an interface for the device itself to communicate with other external devices. For this communication, a wired communication standard such as Ethernet (registered trademark) or FDDI (Fiber Distributed Data Interface) or a wireless communication standard such as 4G, 5G, or Wi-Fi (registered trademark) is used.

[0037] The growth prediction device 10 according to this embodiment is implemented by a general-purpose computer such as a server computer or a personal computer (PC).

[0038] Next, the functional configuration of the growth prediction device 10 will be described with reference to FIG.

[0039] FIG. 3 is a block diagram showing an example of the functional configuration of the growth prediction device 10 according to this embodiment.

[0040] 3, the growth prediction device 10 has, as its functional components, a data input unit 101, a maximum value estimation unit 102, a normalization unit 103, a variable selection unit 104, a learning unit 105, a prediction unit 106, and a data output unit 107. Each functional component is realized by the CPU 11 reading out a growth prediction program stored in the ROM 12 or storage 14, expanding it in the RAM 13, and executing it.

[0041] The storage 14 stores a first trained model 141 and a second trained model 142 generated by the learning unit 105. The first trained model 141 is an example of a trained model and is a prediction model generated by machine learning using previously obtained normalized measurement data of the LAI as training data. The first trained model 141 receives the normalized measurement data of the LAI as input and outputs prediction data representing a future change in the LAI. The second trained model 142 is an example of another trained model and is a prediction model generated by machine learning using previously obtained normalized measurement data of the LAI and a predetermined number of variables selected from environmental information in descending order of contribution rate to the change in the LAI as training data. The second trained model 142 receives the normalized measurement data of the LAI and the selected predetermined number of variables as input and outputs prediction data representing a future change in the LAI. Note that, for example, a long-shot-term memory (LSTM) or the like is used for the machine learning of each of the first trained model 141 and the second trained model 142.

[0042] The data input unit 101 periodically or upon request from a user receives input of LAI measurement data, cultivation information (such as transplanting date, variety, and whether or not fertilization was performed), and environmental information (such as weather and soil solution concentration) from the database 30 (or the LAI measurement sensor 20). Because LAI changes relatively slowly, the time-series LAI measurement data may be processed, for example, by averaging daily or thinning out every few days. Hereinafter, the LAI measurement data will be referred to as LAI measurement data LAI(t). The data input unit 101 converts the LAI measurement data LAI(t) into a format that can be input to each unit. The data input unit 101 outputs the LAI measurement data LAI(t), cultivation information, and environmental information to the maximum value estimation unit 102, and outputs the environmental information to the variable selection unit 104.

[0043] The maximum value estimation unit 102 estimates the maximum value of LAI from the LAI measurement data LAI(t), cultivation information, and environmental information acquired from the data input unit 101. Hereinafter, the maximum value of LAI is referred to as the LAI maximum value LAI. MAX The maximum value of LAI is LAI MAXSince there are trends such as fertilization > no fertilization and high temperature > low temperature, it can be calculated using the following formula (1).

[0044] LAI MAX = α × β × LAI ave ···(1)

[0045] where α is a coefficient determined by the presence or absence of fertilization, β is a coefficient determined by the transplanting date and temperature, and LAI ave is the maximum average LAI (approximately 5). Transplanting dates can be expressed in three stages: early, middle, and late, depending on the planting time. LAI ave is calculated from measurement data obtained from the past to the present of LAI.

[0046] The maximum value estimation unit 102 estimates the maximum LAI value LAI using the LAI measurement data LAI(t) and the above equation (1). MAX to the normalization unit 103. Note that the maximum value estimation unit 102 may have the correspondence relationship of the above equation (1) in the form of a table.

[0047] The normalization unit 103 calculates the LAI measurement data LAI(t) and the LAI maximum value LAI MAX The LAI measurement data LAI(t) is normalized to calculate normalized measurement data. Hereinafter, this normalized measurement data will be referred to as normalized measurement data LAI nom (t) Normalized measurement data LAI nom (t) is calculated by the following equation (2), that is, the LAI measurement value LAI(t) is converted into the maximum LAI value LAI MAX It can be found by dividing by .

[0048] LAI nom (t)=LAI(t) / LAI MAX ···(2)

[0049] The normalization unit 103 calculates the LAI normalized measurement data LAI using the above formula (2). nom (t) is output to the variable selection unit 104.

[0050] The variable selection unit 104 acquires the environmental information from the data input unit 101 and also acquires the LAI normalized measurement data LAI from the normalization unit 103. nom The variable selection unit 104 selects a predetermined number of variables from all variables included in the environmental information in descending order of their contribution rate to the transition of LAI. The variable selection unit 104 analyzes the relationship between multiple explanatory variable candidates (weather, soil solution concentration, etc.) and the objective variable LAI, selects variables with a high contribution rate to describing LAI, and calculates the selected variables and the LAI normalized measurement data LAI. nom (t) is output to the learning unit 105 or the prediction unit 106. Note that if the only variable to be used is LAI, the variable selection function is not used, and the LAI normalized measurement data LAI nom Only (t) is output to the learning unit 105 or the prediction unit 106.

[0051] For example, environmental information highly correlated with growth may differ for each field. Therefore, by appropriately selecting variables with high correlation (i.e., high contribution rates) from among the multiple variables included in the environmental information, it is expected that the accuracy of predicting LAI trends will be improved. Here, the variable selection process performed by the variable selection unit 104 will be specifically described with reference to Figures 4, 5, 6A, and 6B.

[0052] FIG. 4 is a diagram showing an example of a plurality of variables included in the environment information according to this embodiment.

[0053] As shown in Fig. 4, the environmental information includes a number of variables related to weather, such as maximum temperature, minimum temperature (lowest temperature), average temperature, emissivity, solar radiation, humidity, precipitation, water temperature, and soil temperature. The environmental information also includes a number of variables related to soil, such as EC representing electrical conductivity, pH representing hydrogen ion exponent, ORP representing oxidation-reduction potential, DO representing dissolved oxygen, NH4+ representing ammonium ions, NO3- representing nitrate ions, Na+ representing sodium ions, K+ representing potassium ions, and Ca2+ representing calcium ions. NDVI represents the normalized difference vegetation index (also known as the normalized difference vegetation index).

[0054] As described above, when predicting LAI, it is expected that prediction accuracy will be improved by appropriately selecting the environmental information (weather, soil solution concentration, etc.) to be used. Specifically, a regression analysis is performed using the LAI during the cultivation period as the objective variable and the soil solution concentration (concentration of each ion) and meteorological data (temperature, solar radiation, humidity, etc.) as explanatory variables. From the coefficients obtained by the regression analysis (see, for example, Figure 4), variables that have a high correlation with LAI, i.e., a high contribution rate, are selected. Specifically, the larger the coefficient of a variable, the higher the contribution rate to the transition of LAI. For this reason, a predetermined number of variables (for example, 1 to 5) are selected in descending order of coefficient.

[0055] 5 is a graph showing an example of the relationship between actual measurement data A and predicted data P of LAI according to this embodiment. The vertical axis represents LAI, and the horizontal axis represents the number of days after transplantation.

[0056] In FIG. 5, actual measurement data A indicates the actual measurement value of LAI, and predicted data P indicates the predicted value of LAI predicted using environmental information selected according to the contribution rate to LAI.

[0057] Figure 6A is a graph showing an example of LAI when predicted using all environmental information. Figure 6B is a graph showing an example of LAI when predicted using environmental information selected according to the contribution rate to LAI. The vertical axis shows LAI. The solid line shows the measured LAI data, and the dotted line shows the predicted LAI data.

[0058] It can be seen that the example of Fig. 6B makes it easier to identify the position of the peak of the LAI compared to the example of Fig. 6A. Hereinafter, the environmental information selected according to the contribution rate to the LAI will be referred to as "selected environmental information."

[0059] 3, the learning unit 105 performs a learning process for the prediction model. The learning unit 105 receives the LAI normalized measurement data LAI from the variable selection unit 104. nom If only (t) is acquired, the LAI normalized measurement data LAI nom(t) is used as training data for machine learning to generate a first trained model 141. The first trained model 141 is generated by using the LAI normalized measurement data LAI nom This model takes (t) as input and outputs forecast data showing future trends in LAI.

[0060] On the other hand, the learning unit 105 receives the LAI normalized measurement data LAI from the variable selection unit 104. nom (t) and selected environmental information, LAI normalized measurement data LAI nom The second trained model 142 is generated by performing machine learning using the LAI normalized measurement data LAI (t) and the selected environment information as training data. nom This model receives as input (t) and selected environment information, and outputs prediction data representing future changes in LAI. Note that, as described above, for the machine learning of each of the first trained model 141 and the second trained model 142, for example, LSTM or the like is used.

[0061] The prediction unit 106 performs a prediction process using the prediction model learned by the learning unit 105. The prediction unit 106 receives the LAI normalized measurement data LAI from the variable selection unit 104. nom When only (t) is acquired, the first trained model 141 is used to obtain the LAI normalized measurement data LAI nom The future transition of the LAI is predicted from (t). nom (t) is the LAI normalized measurement data LAI acquired by the learning unit 105. nom In other words, the LAI normalized measurement data LAI used for prediction in the prediction unit 106 is data that is later in time than (t). nom (t) is data obtained after the first trained model 141 is generated.

[0062] On the other hand, the prediction unit 106 receives the LAI normalized measurement data LAI from the variable selection unit 104. nom When (t) and the selected environment information are acquired, the LAI normalized measurement data LAI is obtained using the second trained model 142. nomThe future transition of the LAI is predicted from (t) and the selected environment information. As in the above, the LAI normalized measurement data LAI obtained by the prediction unit 106 is nom (t) and the selected environment information are obtained by the learning unit 105 as the LAI normalized measurement data LAI nom (t) and the selected environment information. In other words, the LAI normalized measurement data LAI used for prediction in the prediction unit 106 is nom (t) and the selected environment information are data obtained after the second trained model 142 was generated.

[0063] Furthermore, the prediction unit 106 may predict future trends in the LAI using one of the first trained model 141 and the second trained model 142, whichever has higher LAI prediction accuracy. The prediction accuracy of the model may be determined using a method that determines the difference between an actual measurement value and a predicted value, such as RMSE (Root Mean Squared Error), or other methods.

[0064] That is, the prediction unit 106 predicts the transition of the LAI in the next unit time. For the prediction, for example, a method such as the above-mentioned LSTM is used. When predicting the transition of the LAI, the following variations are assumed to be available for the selection of variables to be used. (1) Forecasting is performed using only LAI. (2) Predictions are made using LAI and selected environmental information. (3) Perform both predictions (1) and (2) above and select the one with the better prediction accuracy.

[0065] The prediction unit 106 outputs prediction data that predicts future changes in the LAI to the data output unit 107.

[0066] The data output unit 107 identifies the day when the LAI will peak from the prediction data acquired from the prediction unit 106, and outputs the day when the LAI will peak and the maximum value of the LAI to the farm work instruction unit 40. The growth prediction device 10 may output this information periodically or in response to a request from a user.

[0067] Returning to FIG. 1 , the agricultural work instruction unit 40 acquires the LAI peak date and maximum LAI value from the growth prediction device 10, determines the need for additional fertilization and the timing of agricultural work based on the acquired LAI peak date and maximum LAI value, and outputs the determination result to the user terminal 50. Specifically, for example, the recommended timing of agricultural work for the cultivated variety is acquired from the cultivation information stored in the database 30. The following items are estimated from the LAI peak date, maximum LAI value, and the current date, and instructions are given to the user.

[0068] -Need for additional fertilization (determined by the maximum LAI value) Top dressing time (day when LAI peaks) - Pesticide application date (approximately 3 days after the peak of LAI) - Best harvest time (about one and a half months after the peak LAI date)

[0069] The necessity of top dressing is determined based on whether the maximum LAI value is equal to or greater than a threshold value. The time for top dressing (top dressing timing) is determined, for example, as a period of X days before and after the day when LAI peaks. The optimal date for X may differ depending on the variety. For example, for variety A, it is 15 days before the heading date (≒ 5 days before the LAI peak), and for variety B, it is 12 days before the heading date (≒ 2 days before the LAI peak). Therefore, when the top dressing time for each variety is based on the day when LAI peaks, it is desirable to set the time for top dressing Y days before the heading date minus 10 days (Y-10). In this case, information on the time for top dressing for each variety can be stored in the database 30 as cultivation information.

[0070] Next, the operation of the growth prediction device 10 according to this embodiment will be described with reference to FIGS.

[0071] 7 is a flowchart showing an example of the flow of a learning process by the growth prediction program according to this embodiment. The learning process by the growth prediction program is realized by the CPU 11 of the growth prediction device 10 writing the growth prediction program stored in the ROM 12 or the storage 14 into the RAM 13 and executing it.

[0072] In step S101 of FIG. 7, the CPU 11 periodically or in response to a user request acquires the LAI measurement data LAI(t), cultivation information (transplanting date, variety, whether or not fertilization was performed, etc.), and environmental information (weather, soil solution concentration, etc.) from the database 30.

[0073] In step S102, the CPU 11 calculates the maximum LAI value LAI from the LAI measurement data LAI(t), cultivation information, and environmental information acquired in step S101, using, for example, the above-mentioned formula (1). MAX Estimate.

[0074] In step S103, the CPU 11 calculates the LAI measurement data LAI(t) acquired in step S101 and the maximum LAI value LAI estimated in step S102. MAX Then, as an example, the LAI measurement data LAI(t) is normalized using the above-mentioned formula (2) to obtain normalized measurement data LAI nom Calculate (t).

[0075] In step S104, the CPU 11 selects, as selected environmental information, a predetermined number of variables from all the variables included in the environmental information acquired in step S101 in descending order of their contribution rate to the change in LAI, as shown in FIG. 4 above, for example.

[0076] In step S105, the CPU 11 calculates the LAI normalized measurement data LAI normalized in step S103. nom (t) and the selected environment information selected in step S104 are used as training data to perform machine learning to generate a second trained model 142. The second trained model 142 is generated by performing machine learning on the LAI normalized measurement data LAI nomThis model takes as input the LAI normalized measurement data LAI (t) and selected environment information, and outputs predicted data that indicates the future transition of LAI. nom If only (t) is acquired, the LAI normalized measurement data LAI nom (t) is used as training data for machine learning to generate a first trained model 141. The first trained model 141 is generated by using the LAI normalized measurement data LAI nom This model takes (t) as input and outputs forecast data showing future trends in LAI.

[0077] In step S106, the CPU 11 stores the first trained model 141 and the second trained model 142 generated in step S105 in the storage 14, and ends the learning process by this growth prediction program.

[0078] 8 is a flowchart showing an example of the flow of a prediction process by the growth prediction program according to this embodiment. The prediction process by the growth prediction program is realized by the CPU 11 of the growth prediction device 10 writing the growth prediction program stored in the ROM 12 or the storage 14 into the RAM 13 and executing it.

[0079] In step S111 of FIG. 8, the CPU 11 periodically or in response to a user request acquires the LAI measurement data LAI(t), cultivation information (transplanting date, variety, whether fertilization was performed, etc.), and environmental information (weather, soil solution concentration, etc.) from the database 30.

[0080] In step S112, the CPU 11 calculates the maximum LAI value LAI from the LAI measurement data LAI(t), cultivation information, and environmental information acquired in step S111, using, for example, the above-mentioned formula (1). MAX Estimate.

[0081] In step S113, the CPU 11 calculates the LAI measurement data LAI(t) acquired in step S111 and the maximum LAI value LAI estimated in step S112. MAXThen, as an example, the LAI measurement data LAI(t) is normalized using the above-mentioned formula (2) to obtain normalized measurement data LAI nom Calculate (t).

[0082] In step S114, the CPU 11 selects, as selected environmental information, a predetermined number of variables from all the variables included in the environmental information acquired in step S111 in descending order of their contribution rate to the transition of the LAI, as shown in FIG. 4 above, for example.

[0083] In step S115, the CPU 11 calculates the LAI normalized measurement data LAI normalized in step S113 for the second trained model 142. nom (t) and the selected environment information selected in step S114 are input, and prediction data representing a future change in LAI is output from the second trained model 142. Note that the CPU 11 outputs the LAI normalized measurement data LAI nom When only (t) is acquired, the LAI normalized measurement data LAI normalized in step S113 is used for the first trained model 141. nom (t) and causes the first trained model 141 to output prediction data representing a future change in the LAI. Alternatively, the CPU 11 may use either the first trained model 141 or the second trained model 142, whichever model has a higher LAI prediction accuracy, to predict a future change in the LAI.

[0084] In step S116, the CPU 11 identifies the day on which the LAI will peak from the prediction data obtained in step S115, outputs the day on which the LAI will peak and the maximum value of the LAI to the agricultural work instruction unit 40 as future growth conditions, and terminates the prediction processing by this growth prediction program.

[0085] As described above, according to this embodiment, by predicting the LAI, it is possible to predict the growth status (the timing of flag leaf emergence and the approximate heading date).

[0086] Furthermore, by using the measured LAI, it is possible to grasp the current growth status, making it possible to respond to growth delays caused by human factors.

[0087] Furthermore, by normalizing the LAI measurement data, it is possible to use the LAI measurement data for learning without distinguishing between varieties, which eliminates the need to build a model for each variety and makes it possible to predict the growth of new varieties.

[0088] In the above embodiment, the growth prediction process executed by the CPU 11 after reading the growth prediction program may be executed by various processors other than the CPU 11. Examples of such processors include dedicated electrical circuits, such as programmable logic devices (PLDs) (such as field-programmable gate arrays (FPGAs)) whose circuit configuration can be changed after fabrication, and application-specific integrated circuits (ASICs) that are processors with circuit configurations specifically designed to execute specific processes. The growth prediction process may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (e.g., multiple FPGAs, or a combination of a CPU and an FPGA). The hardware structure of these various processors is, more specifically, an electrical circuit that combines circuit elements such as semiconductor devices.

[0089] In the above embodiment, the growth prediction program is described as being pre-stored (also referred to as "installed") in the ROM 12 or the storage 14, but the present invention is not limited to this. The growth prediction program may be provided in a form stored in a non-transitory storage medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. The growth prediction program may also be downloaded from an external device via a network.

[0090] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0091] The following additional notes are provided regarding the above-described embodiments.

[0092] (Additional note 1) Memory and at least one processor coupled to said memory; Including, The processor: The system accepts input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information on rice cultivation, and environmental information on the environment in which rice is cultivated, estimating a maximum value of the index value from the measurement data of the index value, the cultivation information, and the environmental information; Normalizing the measurement data of the index values ​​from the measurement data of the index values ​​and the maximum value of the index values ​​to calculate normalized measurement data; A trained model is generated by machine learning the normalized measurement data of the index value obtained in advance as learning data, and the trained model receives the normalized measurement data of the index value as input and outputs prediction data representing a future change in the index value. The trained model predicts a future change in the index value from the calculated normalized measurement data of the index value. The growth prediction device is configured as follows.

[0093] (Additional note 2) A non-transitory storage medium storing a program executable by a computer to perform growth prediction processing, The growth prediction process includes: The system accepts input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information on rice cultivation, and environmental information on the environment in which rice is cultivated, estimating a maximum value of the index value from the measurement data of the index value, the cultivation information, and the environmental information; Normalizing the measurement data of the index values ​​from the measurement data of the index values ​​and the maximum value of the index values ​​to calculate normalized measurement data; A trained model is generated by machine learning the normalized measurement data of the index value obtained in advance as learning data, and the trained model receives the normalized measurement data of the index value as input and outputs prediction data representing a future change in the index value. The trained model predicts a future change in the index value from the calculated normalized measurement data of the index value. Non-transitory storage medium. [Explanation of symbols]

[0094] 10 Growth prediction device 11 CPU 12 ROM 13 RAM 14. Storage 15 Input section 16 Display 17 Communication I / F 18 Bus 20 LAI measurement sensor 30 databases 40 Agricultural Work Instruction Department 50 User Terminals 100 Agricultural work support system 101 Data Entry Section 102 Maximum value estimation unit 103 Normalization section 104 Variable Selection Section 105 Learning Department 106 Prediction Department 107 Data output section 141 First trained model 142 Second trained model

Claims

1. a data input unit that receives input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information related to rice cultivation, and environmental information related to the environment in which rice is cultivated; a maximum value estimation unit that estimates a maximum value of the index value from the measurement data of the index value, the cultivation information, and the environmental information; a normalization unit that normalizes the measurement data of the index values ​​from the measurement data of the index values ​​and the maximum value of the index values ​​to calculate normalized measurement data; a prediction unit that predicts future trends in the index values ​​from the normalized measurement data of the index values ​​calculated by the normalization unit, using a trained model that is generated by machine learning using normalized measurement data of the index values ​​obtained in advance as training data, and that receives the normalized measurement data of the index values ​​as input and outputs prediction data that represents future trends in the index values; A growth prediction device equipped with the above.

2. The cultivation information includes a transplanting date indicating the date on which the rice was transplanted and whether or not fertilization was performed indicating whether or not fertilizer was applied, the environmental information includes temperature during the rice cultivation period; The maximum value estimation unit estimates the maximum value of the index value from a coefficient determined by the presence or absence of fertilization, a coefficient determined by the transplanting date and the temperature, and the average maximum value of the index value. The growth prediction device according to claim 1 .

3. the environmental information includes a plurality of variables related to an environment in which rice is grown; a variable selection unit that selects a predetermined number of variables from all variables included in the environmental information in descending order of contribution rate to the transition of the index value; The prediction unit predicts a future transition of the index value from the normalized measurement data of the index value calculated by the normalization unit and the predetermined number of variables selected by the variable selection unit, using another trained model instead of the trained model, which is generated by machine learning using, as learning data, normalized measurement data of the index value obtained in advance and a predetermined number of variables selected from the environmental information in descending order of contribution rate to the transition of the index value, and which receives as input the normalized measurement data of the index value and the selected predetermined number of variables and outputs prediction data representing a future transition of the index value. The growth prediction device according to claim 1 or 2.

4. The prediction unit predicts a future change in the index value using one of the trained model and the other trained model, which has a higher prediction accuracy of the index value. The growth prediction device according to claim 3.

5. The environmental information includes variables related to the weather during the rice cultivation period and variables related to the soil in which the rice is cultivated. The growth prediction device according to claim 3 or 4.

6. The index value is a leaf area index that represents the ratio of rice leaf area to a unit area of ​​the ground. The growth prediction device according to any one of claims 1 to 5.

7. The system accepts input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information on rice cultivation, and environmental information on the environment in which rice is cultivated, estimating a maximum value of the index value from the measurement data of the index value, the cultivation information, and the environmental information; Normalizing the measurement data of the index values ​​from the measurement data of the index values ​​and the maximum value of the index values ​​to calculate normalized measurement data; a trained model that is generated by machine learning normalized measurement data of the index value obtained in advance as learning data, and that receives the normalized measurement data of the index value as input and outputs prediction data that represents a future change in the index value, and predicts a future change in the index value from the calculated normalized measurement data of the index value; A computer-implemented growth prediction method.

8. The system accepts input of time-series measurement data of index values ​​that change with the growth of rice, cultivation information on rice cultivation, and environmental information on the environment in which rice is cultivated, estimating a maximum value of the index value from the measurement data of the index value, the cultivation information, and the environmental information; Normalizing the measurement data of the index values ​​from the measurement data of the index values ​​and the maximum value of the index values ​​to calculate normalized measurement data; a trained model that is generated by machine learning normalized measurement data of the index value obtained in advance as learning data, and that receives the normalized measurement data of the index value as input and outputs prediction data that represents a future change in the index value, and predicts a future change in the index value from the calculated normalized measurement data of the index value; A growth prediction program to be run on a computer.

Images