Pesticide composition, information processing device and computer program

Abstract

The present invention relates to an agrochemical composition, an information processing device, and a computer program.

Description

[Technical Field]

[0001] The present invention relates to a pesticide composition, an information processing device, and a computer program. [Background technology]

[0002] Currently, a wide variety of pesticide formulations developed for use on paddy rice are commercially available, including, for example, herbicides, insecticides, and fungicides.

[0003] Many herbicides have been developed and sold with different active ingredients and optimal spraying times, including early-stage herbicides that are sprayed before weed emergence, before sowing rice seeds or transplanting rice, or immediately after sowing or transplanting; early- to mid-stage one-shot herbicides that are sprayed after transplanting; and late-stage and mid- to late-stage herbicides that are sprayed from the middle to late stages of rice growth to control weeds that could not be controlled with these early-stage and early- to mid-stage one-shot herbicides or difficult-to-control weeds. Furthermore, as for insecticides and fungicides, seed treatment agents, box-applied agents, and field-applied agents have been developed and sold with different active ingredients.

[0004] From the viewpoint of efficient control, mixtures containing multiple active ingredients have been developed, and the use of such mixtures can efficiently control a wide variety of diseases, pests, and / or weeds. However, products that are suitable for the occurrence of diseases, pests, and / or weeds specific to a field are not always available on the market, and sometimes more active ingredient than necessary is sprayed, raising concerns about the burden on the environment. In light of this situation, applying an appropriate amount of agrochemical formulations containing single or multiple active ingredients depending on the field conditions will lead to a reduction in the environmental burden.

[0005] Furthermore, if the occurrence of diseases, pests, and / or weeds in a field cannot be accurately understood, it may be impossible to select appropriate pesticides, and it may be impossible to completely control the diseases, pests, and / or weeds. As a result, additional pesticides may have to be sprayed to control the diseases, pests, and / or weeds that were not controlled, which may increase the producer's effort, costs, and environmental burden. Furthermore, in fields where diseases, pests, and / or weeds cannot be properly controlled, the growth of crop plants may be insufficient, leading to reduced yields. In other words, applying the appropriate pesticides at the right time and in the required amounts is essential for efficient farming.

[0006] Note that there has been a known technology in which an unmanned aerial vehicle operates a unit that sprays a liquid agent at a location determined by a processing unit based on the results of analyzing images of farmland captured by a camera (Patent Document 7). However, this technology does not select an appropriate agent depending on the occurrence of diseases, pests, and / or weeds in paddy fields.

[0007] From the perspective of efficient pesticide application, selecting multiple pesticide formulations containing a single active ingredient, mixing them in appropriate amounts, and applying them in a single application is an effective strategy. However, pesticide formulations come in a variety of forms, including granules, emulsifiable concentrates, flowable formulations, and water dispersible granules, and mixing these forms poses several problems. For example, (1) even pesticide formulations that can be diluted and sprayed (e.g., emulsifiable concentrates, flowable formulations, and water dispersible granules) are optimized for each active ingredient and have different compositions. Therefore, applying a mixture prepared by mixing them can release excess chemicals (auxiliary ingredients) into the environment. Furthermore, (2) mixing multiple selected pesticide formulations can sometimes make spraying difficult due to changes in the physical and chemical properties of the diluted solution (e.g., unexpected instability of the diluted solution, active ingredient degradation in the diluted solution, clogging of spray nozzles due to solid precipitation and aggregation, etc.). To avoid these problems, it is necessary to develop a formulation that uses standardized auxiliary ingredients for active ingredients with various physical and chemical properties.

[0008] Furthermore, from the perspective of efficient application of pesticides to fields, for example, when the field is a paddy field, it is effective to apply the pesticide locally and then use surface water to diffuse the active ingredient throughout the paddy field. Several technologies for diffusing active ingredients have already been proposed. For example, as paddy rice herbicides, flowable formulations with excellent diffusibility that can be applied from the ridge without dilution (Patent Document 1) and jumbo formulations in which floating granules packaged in a water-soluble film are thrown from the ridge have been developed and sold. Furthermore, formulations used for these localized applications may contain dispersants to ensure sufficient diffusion of the active ingredient in water after application (Patent Document 4, Patent Document 5, and Patent Document 6). Another proposed application method is the irrigation application of water dispersible granules (Patent Document 3). However, many of the pesticides used in these applications are mixtures prepared by mixing multiple pesticide formulations, which is undesirable from the perspective of the aforementioned environmental impact. In addition, insecticides such as surf agents have been developed that are dropped onto the water surface to spread an oil film and thereby disperse the active ingredient, but because they use organic solvents, these insecticides are also undesirable from the perspective of environmental impact.

[0009] Furthermore, when selecting multiple pesticide formulations containing a single active ingredient, mixing them in appropriate amounts, and applying them, it is important to ensure the pesticide formulation's diffusibility so that the active ingredient can be diffused throughout the paddy field using surface water. [Prior art documents] [Non-patent literature]

[0010] [Non-Patent Document 1] Pesticide Formulation Guide / Edited by the Pesticide Science Society of Japan, Pesticide Formulation and Application Methods Research Group / 1997 [Patent documents]

[0011] [Patent Document 1] Special Publication No. 7-47522 [Patent Document 2] Patent No. 2957751 [Patent Document 3] Japanese Patent Application Laid-Open No. 2010-083869 [Patent Document 4] Japanese Patent Application Laid-Open No. 2011-126786 [Patent Document 5] Japanese Patent Application Laid-Open No. 2002-338403 [Patent Document 6] Japanese Patent Application Laid-Open No. 2012-077097 [Patent Document 7] International Publication No. 2020 / 225278 Summary of the Invention [Problem to be solved by the invention]

[0012] An object of the present invention is to provide a pesticide composition, an information processing device, and a computer program. [Means for solving the problem]

[0013] The pesticide composition according to one embodiment is a liquid pesticide composition comprising the following components (a) and (b1), or a solid pesticide composition comprising the following components (a) and (b2): (a) at least one active ingredient selected from the group consisting of an active ingredient having an insecticidal effect, an active ingredient having a fungicidal effect, and an active ingredient having a herbicidal effect; (b1) 1 to 200 g / L of an anionic dispersant; (b2) 1 to 20% w / w of an anionic dispersant.

[0014] An information processing device according to one embodiment is characterized in that it includes "at least one processor, and the at least one processor is configured to acquire, for a specific area included in a target area where paddy rice is cultivated, type information indicating the type of at least one damage that has occurred in the specific area, and quantity information indicating the amount of each of the at least one damage that has occurred in the specific area, and to determine, based on the type information and the quantity information, a plurality of pesticide compositions that constitute a drug to be sprayed in the specific area, and the amount of each of the plurality of pesticide compositions."

[0015] One embodiment of the method is characterized as "a method executed by at least one processor that executes computer-readable instructions, comprising the steps of: acquiring type information indicating the type of at least one damage that has occurred in a specific area included in a target area where paddy rice is cultivated, and quantity information indicating the amount of each of the at least one damage that has occurred in the specific area; and determining, based on the type information and the quantity information, multiple pesticide compositions that constitute the agent to be sprayed in the specific area, and the amount of each of the multiple pesticide compositions."

[0016] One embodiment of the computer program is characterized in that "when executed by at least one processor, the computer program causes the at least one processor to function in such a way that, for a specific area included in a target area where paddy rice is cultivated, type information indicating the type of at least one damage that has occurred in the specific area and quantity information indicating the amount of each of the at least one damage that has occurred in the specific area are acquired, and based on the type information and the quantity information, multiple pesticide compositions that constitute the agent to be sprayed in the specific area and the amount of each of the multiple pesticide compositions are determined."

[0017] Another aspect of the information processing device is characterized in that it "has at least one processor, and is configured to acquire type information indicating the type of at least one damage that has occurred in a specific area included in a target area where seed rice will be directly sown or where seedlings grown from seedling boxes in which seed rice has been sown will be transplanted, and quantity information indicating the amount of each of the at least one damage that has occurred in the specific area, and to determine, based on the type information and the quantity information, a plurality of pesticide compositions that constitute agents to be applied to the seed rice to be directly sown in the specific area or the seed rice to be sown in seedling boxes for transplanting seedlings in the specific area, and the amount of each of the plurality of pesticide compositions."

[0018] Another embodiment of the method is characterized as "a method executed by at least one processor that executes computer-readable instructions, comprising the steps of: acquiring type information indicating the type of at least one damage that has occurred in a specific area included in a target area where rice seeds will be directly sown or where seedlings grown from seedling boxes in which the rice seeds have been sown will be transplanted; and acquiring quantity information indicating the amount of each of the at least one damage that has occurred in the specific area, and determining, based on the type information and the quantity information, a plurality of pesticide compositions that constitute agents to be applied to the rice seeds to be directly sown in the specific area or the rice seeds to be sown in seedling boxes for transplanting the seedlings in the specific area, and the amount of each of the plurality of pesticide compositions."

[0019] Another embodiment of the computer program is characterized in that, "when executed by at least one processor, it causes the at least one processor to function in the following manner: for a specific area included in a target area where rice seeds are to be directly sown or where seedlings grown from seedling boxes into which rice seeds have been sown are to be transplanted, type information indicating the type of at least one damage that has occurred in the specific area, and quantity information indicating the amount of each of the at least one damage that has occurred in the specific area; and, based on the type information and the quantity information, determine a plurality of pesticide compositions that constitute agents to be applied to the rice seeds to be directly sown in the specific area or the rice seeds to be sown in seedling boxes for transplanting seedlings in the specific area, and the amount of each of the plurality of pesticide compositions." [Effects of the Invention]

[0020] According to the present invention, it is possible to provide a pesticide composition, an information processing device, and a computer program. [Brief explanation of the drawings]

[0021] [Figure 1] FIG. 1 is a block diagram illustrating an example of the overall configuration of a determination system according to an embodiment. [Figure 2]FIG. 2 is a block diagram showing an example of the hardware configuration of the information processing device 10A mounted on the server device 10, the mobile spraying device 20, and the terminal device 30 in the determination system shown in FIG. [Figure 3] FIG. 3 is a flow diagram showing an example of the operation executed by the decision system 1 shown in FIG. [Figure 4] FIG. 4 is a flow chart specifically showing an example of a part (ST100) of the operations executed by the decision system 1 shown in FIG. [Figure 5] FIG. 5 is a schematic diagram conceptually illustrating an example of information generated by the information processing device 10A using a learning model in the determination system shown in FIG. [Figure 6] FIG. 6 is a schematic diagram showing an example of a zone determination method executed by the information processing device 10A in the determination system shown in FIG. [Figure 7] FIG. 7 is a flow chart specifically showing an example of a part (ST200) of the operations executed by the decision system 1 shown in FIG. [Figure 8] FIG. 8 is a block diagram showing an example of the hardware configuration of the mobile spraying device 20 shown in FIG. [Figure 9] FIG. 9 is a flow chart specifically showing an example of a part (ST300) of the operations executed by the decision system 1 shown in FIG. [Figure 10] FIG. 10 is a flow diagram illustrating another example of the operations performed by the decision system 1 shown in FIG. [Figure 11] FIG. 11 is a flow chart specifically showing an example of a part (ST1000) of the operations executed by the decision system 1 shown in FIG. [Figure 12] FIG. 12 is a flow chart specifically showing an example of a part (ST2000) of the operations executed by the decision system 1 shown in FIG. [Figure 13] FIG. 13 is a diagram showing the application location of the pesticide composition by drone spraying. [Figure 14]FIG. 14 is a graph showing the relationship between the amount of anionic dispersant added and heat generation. DETAILED DESCRIPTION OF THE INVENTION

[0022] The following section <1> In this section, we will mainly explain the system for mixing pesticide compositions to prepare the drug, and in Section <2> and <3> Next, we will explain the pesticide composition that can be prepared by this system and the method for controlling diseases, pests and / or weeds that can use this system.

[0023] <1> Decision-making system for determining a plurality of pesticide compositions to be mixed Various embodiments of the present invention will be described below with reference to the accompanying drawings. Note that common components in the drawings are designated by the same reference numerals. It should also be noted that components depicted in one drawing may be omitted in another drawing for the sake of clarity. It should also be noted that the accompanying drawings are not necessarily drawn to scale.

[0024] The various systems, methods, and devices described herein should not be construed as limiting in any way. Indeed, the present disclosure is directed to all novel features and aspects of each of the various disclosed embodiments, combinations of these various embodiments with each other, and combinations of portions of these various embodiments with each other. The various systems, methods, and devices described herein are not limited to specific aspects, specific features, or combinations of such aspects with specific features, nor do the products and methods described herein require that one or more particular advantages be present or problems be solved. Furthermore, various features or aspects of the various embodiments described herein, or portions of such features or aspects, may be used in combination with each other.

[0025] Although the operations of some of the various methods disclosed herein are described in a particular order for convenience, it should be understood that description in this manner encompasses rearranging the order of the operations unless a particular order is required by specific text below. For example, operations described in a sequence may, in some cases, be rearranged or performed simultaneously. Furthermore, for purposes of simplicity, the accompanying drawings do not show the various ways in which the various items and methods described herein can be used in conjunction with other items and methods.

[0026] Any theories of operation, scientific principles, or other theoretical descriptions presented herein in connection with the devices or methods of the present disclosure are provided for the purpose of better understanding and are not intended to limit the scope of the technology, and the devices and methods within the scope of the appended claims are not limited to devices and methods that operate in a manner described by such theories of operation.

[0027] Any of the various methods disclosed herein may be implemented using computer-executable instructions stored on one or more computer-readable media and executed on a computer. The one or more media may be non-transitory computer-readable storage media, such as at least one optical media disk, volatile memory components, or non-volatile memory components. The volatile memory components may include, for example, DRAM or SRAM. The non-volatile memory components may include, for example, hard drives and solid-state drives (SSDs). The computer may include any computer available on the market, including, for example, smartphones and other mobile devices with computing hardware.

[0028] Any such computer-executable instructions for implementing the techniques disclosed herein, along with any data generated and used during the implementation of various embodiments disclosed herein, may be stored on one or more computer-readable media (e.g., non-transitory computer-readable storage media). Such computer-executable instructions may, for example, be part of a separate software application, or may be part of a software application accessed or downloaded via a web browser or other software application (such as a remote computing application). Such software may, for example, be executed on a single local computer (e.g., as a process running on any suitable commercially available computer) or in a networked environment (e.g., the Internet, a wide area network, a local area network, a client-server network (such as a cloud computing network), or other such network) using one or more networked computers.

[0029] For clarity, only certain selected aspects of various software-based implementations are described. Other details that are well known in the art are omitted. For example, the techniques disclosed herein are not limited to a particular computer language or program. For example, the techniques disclosed herein may be implemented by software written in C, C++, Java, or any other suitable programming language. Similarly, the techniques disclosed herein are not limited to a particular computer or type of hardware. Specific details of suitable computers and hardware are well known and need not be described in detail herein.

[0030] Moreover, any of the various such software-based embodiments (e.g., including computer-executable instructions for causing a computer to perform any of the various methods disclosed herein) may be uploaded, downloaded, or remotely accessed by suitable communications means, including, for example, the Internet, the World Wide Web, an intranet, a software application, cable (including fiber optic cable), magnetic communication, electromagnetic communication (including RF communication, microwave communication, infrared communication), electronic communication, or other such communications means.

[0031] 1. Overview of the decision system Fig. 1 is a block diagram showing an example of the overall configuration of a determination system according to one embodiment. As shown in Fig. 1, the determination system 1 may include, for example, at least one server device 10, at least one mobile spraying device 20, and at least one terminal device 30. These devices 10, 20, and 30 can be connected to each other via a network 2.

[0032] 1 shows only one server device 10 as an example of the at least one server device 10, but it is also possible to use multiple server devices 10. Also, in FIG. 1, only one mobile spraying device 20 is shown as an example of the at least one mobile spraying device 20, but it is also possible to use multiple mobile spraying devices 20. Similarly, in FIG. 1, only one terminal device 30 is shown as an example of the at least one terminal device 30, but it is also possible to use multiple terminal devices 30.

[0033] Network 2 may include, but is not limited to, a cellular network, a wireless network, a landline network, the Internet, an intranet, a local area network (LAN), a wide area network (WAN), and / or an Ethernet network. The wireless network may include, but is not limited to, RF connections via Bluetooth, WiFi (such as IEEE 802.11a / b / n), WiMax, cellular, satellite, laser, infrared, etc.

[0034] The server device 10 is a device equipped with an information processing device (not shown). Such a server device 10 may be, for example, a web server, a cloud server, a personal computer, a workstation, or a supercomputer.

[0035] Mobile spraying device 20 is equipped with an information processing device (not shown) and is controlled by this information processing device to spray a chemical. Such mobile spraying device 20 may be a traveling type spraying device (e.g., a spraying device that travels on the ground or water surface) or an airborne type spraying device (e.g., a drone-type spraying device).

[0036] The terminal device 30 is a device equipped with an information processing device (not shown). Such a terminal device 30 may be a smartphone, a feature phone, a mobile phone, a portable information terminal, a personal computer, a tablet, or the like.

[0037] 1, the server device 10 can receive and acquire, for a specific area (e.g., a specific area in a field) included in a certain field (target area), type information indicating the type of at least one damage that has occurred in this specific area and quantity information indicating the amount of each of the at least one damage that has occurred in this specific area from various devices. Such various devices may include the mobile spraying device 20, the terminal device 30, and / or other devices (not shown) that can be connected to the network 2.

[0038] Furthermore, based on the type information and amount information received in this manner, the server device 10 can determine the multiple pesticide compositions that constitute the agent to be sprayed in the specific area and the amount of each of these multiple pesticide compositions. Such determination can be performed by an information processing device (not shown) installed in the server device 10. Note that, instead of or in addition to the information processing device installed in the server device 10, such determination can also be performed by an information processing device (not shown) installed in the mobile spraying device 20, an information processing device (not shown) installed in the terminal device 30, and / or an information processing device (not shown) installed in another device connectable to the network 2. To achieve this, the device making the determination can receive type information and amount information from other devices (10, 20, 30, etc.) via the network 2. Alternatively, the device making the determination can receive type information and amount information via a user interface or the like provided in the device.

[0039] Furthermore, as an option, the mobile spraying device 20 can prepare a pesticide by mixing the amounts determined as above for each of the multiple pesticide compositions determined as above. Alternatively, the mobile spraying device 20 can obtain the prepared pesticide itself from a preparation device (not shown) that can similarly perform such preparation. This allows the mobile spraying device 20 to move to the specific area (a certain area in the field) in the target area (such as a field) and spray the pesticide on the rice plants present in this specific area.

[0040] 2. Configuration of the information processing device installed in the server device 10, etc. An information processing device 10A can be installed in the server device 10, the mobile spraying device 20, and the terminal device 30. Figure 2 is a block diagram showing an example of the hardware configuration of the information processing device 10A installed in the server device 10, the mobile spraying device 20, and the terminal device 30 in a determination system according to one embodiment.

[0041] 2, the information processing device 10A mainly includes a central processing unit 11, a main memory device 12, an input / output interface device 13, an input device 14, an auxiliary memory device 15, and an output device 16. These devices are connected to each other by a data bus and / or a control bus.

[0042] The central processing unit 11 is a single processor called a "CPU (Central Processing Unit)." The central processing unit 11 performs operations on instructions and data stored in the main memory device 12, and can store the results of these operations in the main memory device 12. Furthermore, the central processing unit 11 can control an input device 14, an auxiliary memory device 15, an output device 16, and the like via an input / output interface device 13. The information processing device 10A can include one or more such central processing units 11.

[0043] The main memory device 12 is referred to as "memory" and can store instructions and data received from the input device 14, the auxiliary memory device 15, and the network 400 via the input / output interface device 13, as well as the results of calculations by the central processing unit 11. The main memory device 12 can include, but is not limited to, RAM (Random Access Memory), ROM (Read Only Memory), and / or flash memory.

[0044] The auxiliary storage device 15 is a storage device with a larger capacity than the main storage device 12. The auxiliary storage device can store instructions and data (computer programs) that constitute an operating system, specific applications, etc. This specific application can be executed by the information processing device 10A mounted on the server device 10 (or the mobile spraying device 20, the terminal device 30), causing the information processing device 10 as a whole to function as the server device 10 (or the mobile spraying device 20, the terminal device 30). Furthermore, the auxiliary storage device 15 can transmit these instructions and data (computer programs) to the main storage device 12 via the input / output interface device 13 by being controlled by the central processing unit 11. The auxiliary storage device 15 can include, but is not limited to, a magnetic disk device and / or an optical disk device.

[0045] The input device 14 is a device that inputs data from the outside and can include, but is not limited to, at least one sensor, a touch panel, a button, a keyboard, and / or a mouse. The at least one sensor can include, for example, an RGB camera, a multispectral camera, a hyperspectral camera, and / or a wide wavelength spectrum camera.

[0046] The output device 16 may include, but is not limited to, a display device, a touch panel, and / or a printer device.

[0047] In this hardware configuration, the central processing unit 11 can sequentially load instructions and data (computer programs) constituting a specific application stored in the auxiliary storage device 15 into the main storage device 12 and perform operations on the loaded instructions and data. This allows the central processing unit 11 to control the input device 14 and / or output device 16 via the input / output interface device 13, or to send and receive various information to and from other devices (e.g., the server device 10, the mobile spraying device 20, and / or the terminal device 30, etc.) via the input / output interface device 13 and the network 2.

[0048] The information processing device 10A may include, instead of the central processing unit 11 or in addition to the central processing unit 11, one or more microprocessors and / or a graphics processing unit (GPU) or the like.

[0049] 3. Operation of the decision system (part 1) Next, a specific example of the operation executed by the determination system 1 having the above configuration will be described with reference to Fig. 3. Fig. 3 is a flow diagram showing an example of the operation executed by the determination system 1 shown in Fig. 1.

[0050] 3, the operations performed by the determination system 1 can broadly include a step (hereinafter referred to as "ST") 100 of acquiring information about a target area (or a specific area included in the target area), and a step ST200 of determining a plurality of pesticide compositions constituting a pharmaceutical and the amount of each pesticide composition based on the information acquired in ST100. Furthermore, as an option, the operations performed by the determination system 1 can include a step ST300 of spraying a pharmaceutical prepared by mixing determined amounts of each of the plurality of pesticide compositions determined in ST200 onto the target area (or a specific area included in the target area). Specific examples of ST100, ST200, and ST300 will be explained below in turn.

[0051] 3-1. Operations performed by ST100 FIG. 4 is a flow chart specifically showing an example of a part (ST100) of the operations executed by the decision system 1 shown in FIG. Here, the target area refers to the area where rice is cultivated (including, for example, the area where seedlings are transplanted and / or the area where seed rice is directly sown), and may be one field, a portion of one field, and / or multiple fields. Furthermore, damage may include, but is not limited to, diseases that have occurred in the target area (or rice grown in the target area), pests that have occurred in the target area (or rice grown in the target area), and / or weeds that have occurred in the target area.

[0052] First, in ST102, the information processing device 10A can acquire an image of the target area. For example, such an image may be an image acquired by a mobile spraying device 20 or the like capturing an image of the target area from above the target area (such an image may be transmitted to the information processing device 10A via the network 2). For another example, such an image may be an image acquired by a user located in a building, helicopter, airplane, or the like capturing an image of the target area using a camera that is an input device 14 such as a terminal device 30 (such an image may be transmitted to the information processing device 10A via the network 2). For yet another example, such an image may be an image acquired by the information processing device 10A via the network 2 from some device (such as a server device that provides images of aerial photographs of the entire country).

[0053] In ST104, the information processing device 10A inputs the image acquired in ST102 into a trained learning model, thereby being able to identify which pixels contained in the image have any harm occurring or have no harm occurring.

[0054] Such a trained learning model may be a learning model capable of performing machine learning (particularly deep learning) that includes an input layer, an output layer, and multiple intermediate layers arranged between the input layer and the output layer. Each of these learning models has previously trained using a large number (e.g., tens of thousands or more) of training data, including an image (input data) of rice plants suffering from a certain type of damage, information (output data) indicating which pixel (one or more pixels) in the image the damage is occurring in. Such a learning model can perform training by optimizing the large number of coefficients used in the learning model so as to reduce the error between the data output when input data in each training data is input and the output data in that training data.

[0055] In a first example, the information processing device 10A can have such a trained learning model. In a second example, the information processing device 10A can receive (acquire) such a trained learning model stored in a device other than the device in which the information processing device 10A is installed, via the network 2. In a third example, the information processing device 10A can receive (acquire) output data of such a trained learning model stored in a device other than the device in which the information processing device 10A is installed, by transmitting an image via the network 2 to the trained learning model.

[0056] By inputting an image into such a trained learning model, the information processing device 10A can acquire, from this learning model, information indicating which pixel in the image has which defect.

[0057] Fig. 5 is a schematic diagram conceptually illustrating an example of information generated by the information processing device 10A using a learning model in the determination system shown in Fig. 1. For the sake of simplicity, Fig. 5 illustrates an example of information generated for each of a small portion of unit areas in a target area by inputting an image of the target area into the learning model.

[0058] 5 shows only six unit areas 120A to 120I as an example. Each unit area can be represented by any number of pixels, but in this example, it can be represented by 25 pixels (5 pixels x 5 pixels). In this example, each unit area is square, but in another example, it can be any shape, such as a polygon, circle, ellipse, trapezoid, or diamond.

[0059] For each unit area, each pixel may be associated with information specific to the damage that has occurred at the position corresponding to that pixel. For example, focusing on unit area 120A, the position corresponding to pixel 120A1 is associated with information "1" specific to the first damage that has occurred at that position. Here, as an example, each position corresponding to a total of 10 pixels may be associated with information "1" specific to the first damage that has occurred at that position. The first damage may be any of a disease that has occurred in the target area (or paddy rice grown in the target area), a pest that has occurred in the target area (or paddy rice grown in the target area), and weeds that have occurred in the target area.

[0060] In addition, the position corresponding to pixel 120A2 may be associated, in one example, with information "null" that is specific to the event that no harm has occurred at this position, or, in another example, with information such as "0" that is specific to the event that no harm has occurred at this position.

[0061] Furthermore, for example, when focusing on unit area 120G, the position corresponding to pixel 120G1 is associated with information "4" specific to the fourth damage that has occurred at this position. Here, as an example, information "4" specific to the fourth damage that has occurred at this position can be associated with each position corresponding to a total of five pixels. The fourth damage is any of a disease that has occurred in the target area (or paddy rice grown in the target area), a pest that has occurred in the target area (or paddy rice grown in the target area), and weeds that have occurred in the target area.

[0062] Furthermore, the position corresponding to pixel 120G2 is associated with information "5" specific to the fifth damage that has occurred at this position. Here, as an example, information "5" specific to the fifth damage that has occurred at this position may be associated with each of the positions corresponding to a total of three pixels. The fourth damage is any of a disease that has occurred in the target area (or paddy rice grown in the target area), a pest that has occurred in the target area (or paddy rice grown in the target area), and weeds that have occurred in the target area.

[0063] Furthermore, the position corresponding to pixel 120G3 is associated with information "6" specific to the sixth damage that has occurred at this position. Here, as an example, the information "6" specific to the sixth damage that has occurred at this position may be associated with each position corresponding to a total of two pixels. The sixth damage is any one of disease that has occurred in the target area (or paddy rice grown in the target area), pests that have occurred in the target area (or paddy rice grown in the target area), and weeds that have occurred in the target area.

[0064] Furthermore, the location corresponding to pixel 120G4 may be associated, in one example, with information "null" specific to the event that no harm has occurred at this location, while in another example, with information such as "0" specific to the event that no harm has occurred at this location.

[0065] The other unit areas will not be described in detail, but the concept is similar to that described for unit areas 120A and 120G.

[0066] Returning to Figure 4, in ST106, the information processing device 10A can use the information output from the learning model in ST104 (for example, the information illustrated in Figure 5) to acquire and store (for example, in the auxiliary storage device 15) "unit type information" indicating at least one type of harm that has occurred in each of the multiple unit areas included in the target area.

[0067] 4, the information processing device 10A can acquire, as unit type information, information "1" indicating the type of harm that occurred in the unit area 120A (information "1" corresponding to the first harm) for the unit area 120A. The information processing device 10A can also acquire, as unit type information, information "1" indicating the type of harm that occurred in each of the unit areas 120B and 120C.

[0068] Furthermore, for the unit area 120D, the information processing device 10A can acquire, as unit type information, information "2" and information "3" (information "2" corresponding to the second harm and information "3" corresponding to the third harm) indicating the type of harm that occurred in this unit area 120D. For each of the unit areas 120E and 120F, the information processing device 10A can also acquire, as unit type information, information "2" and information "3" indicating the type of harm that occurred in this unit area.

[0069] Furthermore, for the unit region 120G, the information processing device 10A can acquire, as unit type information, information "4," information "5," and information "6" (information "4" corresponding to the fourth harm, information "5" corresponding to the fifth harm, and information "6" corresponding to the sixth harm) indicating the type of harm that occurred in this unit region 120G. For the unit region 120H, the information processing device 10A can also acquire, as unit type information, information "4," information "5," and information "6" indicating the type of harm that occurred in this unit region 120H.

[0070] Furthermore, for the unit area 120I, the information processing device 10A can acquire, as unit type information, information "null" (or "0" or the like) indicating an event in which no harm has occurred in this unit area 120I.

[0071] In one example, the information processing device 10A can acquire unit type information for all unit areas included in the target area, but in another example, it can acquire unit type information for each of some of the unit areas (one or more unit areas) included in the target area.

[0072] In yet another example, the information processing device 10A can acquire, for a specific region included in the target region, “type information” indicating at least one type of damage that has occurred in the specific region. For example, if the target region is a single farm field and the specific region is a portion of the single farm field, the information processing device 10A can acquire, for the portion of the region, type information indicating at least one type of damage that has occurred in the portion of the region. In this case, the information processing device 10A can acquire the type information by searching for information associated with each pixel that constitutes the portion of the region in the information acquired from the learning model in ST104. For example, if the specific region is an area composed of unit region 120D and unit region 120G, the information processing device 10A can acquire, as type information, information “2,” information “3,” information “4,” information “5,” and information “6” indicating the types of damage that have occurred in these unit regions.

[0073] The information identifying the target region and the information identifying the specific region in this target region may be input by the user via the input device 14 of the information processing device 10A and acquired by the information processing device 10A, for example.

[0074] 4, next, in ST108, the information processing device 10A can determine, for each of a plurality of unit areas included in the target area, the amount of at least one harm that has occurred in this unit area, using the information output from the learning model in ST104 (for example, the information exemplified in FIG. 5). As a result, the information processing device 10A can acquire, for each of a plurality of unit areas included in the target area, unit amount information that indicates the amount of at least one harm that has occurred in this unit area, and store it (for example, in the auxiliary storage device 15).

[0075] 4, the information processing device 10A can acquire, for the unit region 120A, information "1-10" indicating that there are 10 pieces of information "1" indicating the type of harm that has occurred in this unit region 120A (information indicating that there are 10 pieces of information "1" corresponding to the first harm / information indicating that the density of information "1" is "10") as unit amount information. For each of the unit regions 120B and 120C, the information processing device 10A can also acquire, as unit amount information, information "1-10" indicating that there are 10 pieces of information "1" indicating the type of harm that has occurred in this unit region 120A.

[0076] Furthermore, for the unit region 120D, the information processing device 10A can acquire, as unit amount information, information "2-4" indicating that there are four pieces of information "2" indicating the type of harm that has occurred in this unit region 120D (information indicating that there are four pieces of information "2" corresponding to the second harm / information indicating that the density of the information "2" is "4") and information "3-6" indicating that there are six pieces of information "3" indicating the type of harm that has occurred in this unit region 120D (information indicating that there are six pieces of information "3" corresponding to the third harm / information indicating that the density of the information "3" is "6"). For each of the unit regions 120E and 120F, the information processing device 10A can acquire, as unit amount information, information "2-4" indicating that there are four pieces of information "2" indicating the type of harm that has occurred in this unit region and information "3-6" indicating that there are six pieces of information "3" indicating the type of harm that has occurred in this unit region 120D.

[0077] Furthermore, for the unit region 120G, the information processing device 10A can acquire, as unit amount information, information "4-5" indicating that there are five pieces of information "4" indicating the type of harm that has occurred in this unit region 120G, information "5-3" indicating that there are three pieces of information "5" indicating the type of harm that has occurred in this unit region 120G, and information "6-2" indicating that there are two pieces of information "6" indicating the type of harm that has occurred in this unit region 120G. For the unit region 120H, the information processing device 10A can also acquire, as unit amount information, information "4-5" indicating that there are five pieces of information "4" indicating the type of harm that has occurred in this unit region 120H, information "5-3" indicating that there are three pieces of information "5" indicating the type of harm that has occurred in this unit region 120H, and information "6-2" indicating that there are two pieces of information "6" indicating the type of harm that has occurred in this unit region 120H.

[0078] In one example, the information processing device 10A can acquire unit quantity class information for all unit areas included in the target area, but in another example, it can acquire unit quantity information for each of some of the unit areas (one or more unit areas) included in the target area.

[0079] Furthermore, in another example, the information processing device 10A can acquire "quantity information" for a specific region included in the target region, the quantity indicating the amount of at least one damage that has occurred in the specific region. For example, if the target region is a single farm field and the specific region is a portion of the single farm field, the information processing device 10A can acquire quantity information for the portion of the region, the quantity indicating the amount of at least one damage that has occurred in the portion of the region. In this case, the information processing device 10A can acquire the type information by searching for information associated with each pixel that constitutes the portion of the region in the information acquired from the learning model in ST104. For example, if the specific region is an area composed of unit region 120D and unit region 120G, the information processing device 10A can acquire the following information as quantity information: - Information "2-4" indicating that there are four pieces of information "2" indicating the type of damage that occurred in these unit areas - Information "3-6" indicating that there are six pieces of information "3" indicating the type of damage that has occurred in these unit areas - Information "4-5" indicating that there are five pieces of information "4" indicating the type of damage that has occurred in these unit areas -Information "5-3" indicating that there are three pieces of information "5" indicating the type of damage that has occurred in these unit areas -Information "6-2" indicating that there are two pieces of information "6" indicating the type of damage that has occurred in these unit areas

[0080] 4 again, up to this point, an example has been described in which information processing device 10A, as one embodiment, acquires unit type information (or type information) in ST104 and ST106 via ST102, and acquires unit amount information (or amount information) in ST108. However, in another embodiment, information processing device 10A can acquire unit type information (or type information) in ST106, and acquire unit amount information (or amount information) in ST108, without going through ST102 and ST104. Specifically, for example, information processing device 10A can acquire unit type information (or type information) from some device via network 2 in ST106, and acquire unit amount information (or amount information) from some device via network 2 in ST108. In this case, the device in question may be another information processing device capable of acquiring (generating) this information by executing the processes shown in ST102 to ST108, or may be another information processing device capable of acquiring (generating) this information by executing any other process different from the processes shown in ST102 to ST108.

[0081] Next, optionally, in ST110, information processing device 10A can determine at least one zone from within the target region using the unit type information and unit amount information (or type information and amount information). Here, a zone can be said to be a plurality of unit areas that are common or substantially common in the type of damage that has occurred and the amount of damage that has occurred.

[0082] A method for determining a zone executed by the information processing device 10A will be described with reference to Fig. 6. Fig. 6 is a schematic diagram showing an example of a method for determining a zone executed by the information processing device 10A in the determination system shown in Fig. 1. Fig. 6 is expressed in correspondence with Fig. 5 referred to above.

[0083] First, let us focus on unit area 120A. Ten pieces of information "1" exist in unit area 120A. That is, the unit type information for unit area 120A contains information "1," and the unit quantity information for unit area 120A contains information "1-10." By searching for unit type information and unit quantity information for other unit areas, information processing device 10A can identify that the unit type information for each of unit area 120B and unit area 120C both contains only information "1," and the unit quantity information for each of unit area 120B and unit area 120C both contains only information "1-10." That is, information processing device 10A can identify that unit areas 120A, 120B, and 120C are common to each other in terms of both the type of harm caused and the amount of harm caused. As a result, information processing device 10A can determine an area including all of unit areas 120A, 120B, and 120C as first zone 130.

[0084] Next, attention is focused on unit region 120D. In unit region 120D, there are four pieces of information "2" and six pieces of information "3." That is, the unit type information for unit region 120D includes information "2" and information "3," and the unit quantity information for unit region 120D includes information "2-4" and information "3-6." By searching for unit type information and unit quantity information for other unit regions, information processing device 10A can identify that the unit type information for unit region 120E and unit region 120F each includes only information "2" and information "3," and that the unit quantity information for unit region 120E and unit region 120F each includes only information "2-4" and information "3-6." That is, information processing device 10A can identify that unit regions 120D, 120E, and 120F are common to each other in terms of both the type of harm caused and the amount of harm caused. As a result, the information processing device 10A can determine the area including the entire unit areas 120D, 120E, and 120F as the second zone 131.

[0085] Next, focus on unit region 120G. In unit region 120G, there are five pieces of information "4," three pieces of information "5," and two pieces of information "6." That is, the unit type information for unit region 120G includes information "4," information "5," and information "6," and the unit quantity information for unit region 120G includes information "4-5," information "5-3," and information "6-2." By searching unit type information and unit quantity information for other unit regions, information processing device 10A can identify that the unit type information for unit region 120H includes only information "4," information "5," and information "6," and that the unit quantity information for unit region 120H includes only information "4-5," information "5-3," and information "6-2." That is, information processing device 10A can identify that unit regions 120G and 120H are common to each other in terms of both the type of harm caused and the amount of harm caused. As a result, the information processing device 10A can determine the area including the entire unit areas 120G and 120H as the third zone 132.

[0086] 6, the information processing device 10A can identify multiple areas that are "common" in the type of harm that occurred and the amount of harm that occurred, and determine an area that includes all of the multiple areas identified in this way as one zone. However, in another embodiment, the information processing device 10A can identify multiple areas that are "substantially common" in the type of harm that occurred and the amount of harm that occurred, and determine an area that includes all of the multiple areas identified in this way as one zone.

[0087] Here, "substantially common" does not only mean that the type of harm occurring in a certain unit area and the type of harm occurring in another unit area are completely identical, but also that the type of harm occurring in the certain unit area and the type of harm occurring in the other unit area are similar (it is possible to regard both types as substantially the same).

[0088] Furthermore, "substantially the same" does not only mean that the amount (density) of harm occurring in a certain unit area and the amount (density) of harm occurring in another unit area are completely identical, but also means that the error between the amount (density) of harm occurring in the certain unit area and the amount (density) of harm occurring in another unit area is equal to or less than any predetermined value (the two amounts can be considered to be substantially the same).

[0089] Returning to FIG. 4, next, in ST112, the information processing device 10A can generate and store (in the auxiliary storage device 15, for example) zone information indicating each zone (one or more zones) determined in ST110.

[0090] 6, the information processing device 10A can generate and store, for the first zone 130, zone information indicating that the first zone 130 is an area including unit areas 120A, 120B, and 130C. Furthermore, for the second zone 131, the information processing device 10A can generate and store zone information indicating that the second zone 131 is an area including unit areas 120D, 120E, and 130F. Similarly, for the third zone 132, the information processing device 10A can generate and store zone information indicating that the third zone 132 is an area including unit areas 120G and 120H.

[0091] As will be described later, by using such zone information, the mobile spraying device 20 can spray a pesticide into a zone that is a mixture of multiple pesticide compositions determined to correspond to the zone, in the amounts of each pesticide composition determined to correspond to the zone.

[0092] Note that the information processing device 10A can also execute the above-described processes in ST110 and ST112 (FIG. 4) in ST200 (FIG. 3).

[0093] 3-2. Operations performed by ST200 Fig. 7 is a flow chart specifically illustrating an example of a part (ST200) of the operations executed by the determination system 1 shown in Fig. 3. The information processing device 10A can execute the operations shown in Fig. 7 for each specific area included in the target area or each zone included in the target area. Of course, the information processing device 10A can also execute the operations shown in Fig. 7 for any one specific area included in the target area or any one zone included in the target area.

[0094] First, in ST202, the information processing device 10A can acquire type information indicating at least one damage that has occurred in a specific area (for example, one farm field, or a partial area included in one farm field, etc.) and quantity information indicating the amount of each of the at least one damage that has occurred in the specific area. Such information has already been acquired and written by the information processing device 10A in ST106 and ST108.

[0095] Here, when the zone determined in ST110 (see FIG. 4) is used as the specific area, the information processing device 10A can acquire the zone information stored in ST112 (see FIG. 4) corresponding to that zone, and acquire the unit type information and unit amount information corresponding to each unit area identified by this zone information.

[0096] Next, in ST204, the information processing device 10A can acquire various pieces of information to be input to the trained first learning model or the trained first decision tree model. Specifically, the information processing device 10A can acquire the following information (1) to information (5).

[0097] (1) Soil information on the type of soil in a specific area (zone) (2) Information on past harms that have occurred in a specific area (zone) (3) Seed burial information on seeds buried in the soil of a specific area (zone) (4) Information on the prevalence of pests resistant to drugs in specific areas (zones) (5) Weather forecast information regarding weather data predicted for the rice cultivation period in a specific area (zone).

[0098] The information (1), i.e., soil information, may be, for example, information indicating (identifying) the type of soil (sand, clay, etc.) in a specific area (zone). This information (1) may be used in consideration of the possibility that the type of soil in a specific area (zone) may affect the damage that occurs in the soil, the speed and timing of the development of this damage, etc.

[0099] Information (2), i.e., harm occurrence information, may be information that indicates (identifies) harm that has occurred in the past in a specific area (zone). This information (2) may be used in consideration that harm that has already occurred in the past in a specific area (zone) is likely to occur in the same specific area (zone) in the present.

[0100] Information (3), i.e., seed burial information, can be information that indicates (identifies) seeds (weed seeds) buried in a specific area (zone). When a certain seed is buried in a specific area (zone), this information (3) can be used in consideration of the high possibility that a weed corresponding to that seed is still emerging in the same specific area (zone).

[0101] Information (4), i.e., pest prevalence information, can be information that indicates (identifies) which pests that are resistant to chemicals (fungicides, insecticides, and / or herbicides) are currently prevalent in a specific area (zone). If a pest with a certain resistance is currently prevalent in a region that includes a specific area (zone), this information (5) can be used, taking into account the high possibility that such a pest with resistance will still occur in the same specific area (zone).

[0102] Information (5), i.e., weather forecast information, may be information indicating (identifying) weather data (temperature, humidity, weather, and / or precipitation, etc.) predicted for the period during which rice is cultivated in a specific area (zone). This weather data for the period during which rice is cultivated in a specific area (zone) may be used in consideration of the possibility that damage occurring in the same specific area (zone), the speed and timing of the development of this damage, etc.

[0103] The above information may be information that the user inputs using the input device 14 (keyboard, mouse, touch panel, etc.) of the information processing device 10A by utilizing a menu screen displayed on the output device (display, etc.) 16 of the information processing device 10A.

[0104] Next, in ST206, the information processing device 10A can acquire the following information (6) to information (8) by inputting the information (type information and quantity information) acquired in ST202 and at least one of the information acquired in ST204 (i.e., at least one of information (1) to information (5)) into the trained first learning model or the trained first decision tree model.

[0105] (6) Information on the type and amount of damage that may occur in a specific area (zone) (7) Information on the predicted growth period of this pest (8) Information on phytotoxicity occurring in specific areas (zones)

[0106] Information (6), i.e., harm occurrence prediction information, can literally be information that indicates (identifies) the type of harm that will occur in a specific area (zone) and the amount of this harm.

[0107] Information (7), i.e., pest growth prediction information, may be information that indicates (identifies) the timing (how early, at what time) of the occurrence of the pest indicated in information (6).

[0108] The information (8), that is, the drug damage information, can be information that indicates (identifies) what kind of drug damage (side effects, etc.) will occur in a specific area (zone).

[0109] The first learning model (e.g., a machine learning / deep learning model having an input layer, multiple intermediate layers, and an output layer) and the first decision tree model each undergo training in advance using a large number (e.g., tens of thousands or more) of training data including type information, quantity information, and information (1) to information (5) (input data) and information (6) to information (8) (output data). Such a first learning model or first decision tree model can perform training by optimizing a large number of coefficients (or a large number of parameters) used in the first learning model or first decision tree model so as to reduce the error between the data output when input data in each training data is input and the output data in that training data.

[0110] In a first example, the information processing device 10A can have such a trained first learning model and a trained first decision tree model. In a second example, the information processing device 10A can receive (acquire) such a trained first learning model and a trained first decision tree model stored in a device other than the device in which the information processing device 10A is installed, via the network 2. In a third example, the information processing device 10A can receive (acquire) information (6) to information (8), which are output data of the first learning model or the first decision tree model, by transmitting necessary information (e.g., at least one of information (1) to information (5)) via the network 2 to the trained first learning model or the trained first decision tree model stored in a device other than the device in which the information processing device 10A is installed.

[0111] Next, in ST208, the information processing device 10A inputs the information (6) to information (8) acquired in ST206 into a trained second learning model or a trained second decision tree model, thereby acquiring and storing (for example, in the auxiliary storage device 15) the information (9) and information (10) shown below.

[0112] (9) Information indicating (identifying) multiple pesticide compositions that constitute the agent to be sprayed in a specific area (zone). (10) Information indicating (identifying) the amount of each of the multiple pesticide compositions shown in information (9).

[0113] Additionally, as an option, the information processing device 10A can input the information (6) to information (8) acquired in ST206 into a trained second learning model or a trained second decision tree model, thereby acquiring and storing (for example, in the auxiliary storage device 15) at least one of the following information (11) and information (12) in addition to information (9) and information (10):

[0114] (11) Optimal continuous processing information regarding the optimal method for continuously spraying (processing) the agent mixed according to information (9) and information (10). (12) Optimal treatment timing information regarding the most suitable time to spray (treat) the chemicals mixed in accordance with information (9) and information (10).

[0115] Information (11), i.e., optimal continuous treatment information, may be information indicating (identifying) the optimal method for spraying (treating) the agent mixed in accordance with information (9) and information (10) (e.g., a method of spraying the agent continuously every two weeks, etc.).

[0116] Information (12), i.e., optimal treatment time information, may be information indicating (identifying) the optimal time (e.g., March) to spray (treat) the pesticide mixed according to information (9) and information (10).

[0117] Such second learning model and second decision tree model are each trained in advance using a large number (e.g., tens of thousands or more) of training data including information (6) to (8) (input data) and information (9) to (12) (output data). Such second learning model or second decision tree model can be trained by optimizing a large number of coefficients (or a large number of parameters) used in the second learning model or second decision tree model so as to reduce the error between the data output when input data in each training data is input and the output data in that training data.

[0118] In a first example, the information processing device 10A can have such a trained second learning model and a trained second decision tree model. In a second example, the information processing device 10A can receive (acquire) such a trained second learning model and a trained second decision tree model stored in a device other than the device in which the information processing device 10A is installed, via the network 2. In a third example, the information processing device 10A can receive (acquire) information (9) to information (12), which are output data of the second learning model or the second decision tree model, by transmitting necessary information (e.g., information (6) to information (8)) via the network 2 to the trained second learning model or the trained second decision tree model stored in a device other than the device in which the information processing device 10A is installed.

[0119] In ST208, the information processing device 10A can also display the acquired information (9) and information (10) (and further information (11) and / or information (12)) to the user via the output device 16.

[0120] Next, in optional ST210, the information processing device 10A can acquire various pieces of information to be input to the trained third learning model or the trained third decision tree model. Specifically, the information processing device 10A can acquire the following information (13) and "at least one" piece of information (14) to information (16).

[0121] (13) Information on the general effectiveness of the drug against the harm indicated by the above information (6) (Harm Prediction Information) (14) Information on the type and effects of drugs previously used in a specific area (zone) (15) Information on the composition of soil in a specific area (zone). (16) Information about the moisture retention capacity of specific areas (zones)

[0122] Information (13) may be information indicating (identifying) a commonly known effect of a drug on the harm indicated by information (6).

[0123] The information (14) may be information indicating (identifying) the type of drug that has actually been used in the past in a specific area (zone) and the effects that have actually been obtained with that drug in the past. Even if a certain drug is known to generally have a high (or low) effect on a certain harm, there may be cases where the drug was actually used in the past in a specific area (zone) but was not sufficiently effective (or was able to be sufficiently effective). This information (14) may be used to take such cases into consideration.

[0124] The information 15 can be information that indicates (identifies) the composition of the soil of a particular area (zone), which can be used in light of the high likelihood that the composition of the soil of a particular area (zone) will affect the effectiveness of the pesticide composition applied to this target area.

[0125] The information 16 can be information that indicates (identifies) the water-holding capacity of a particular area (zone), which can be used in light of the high likelihood that the water-holding capacity of a particular area (zone) will affect the effectiveness of a pesticide composition applied to this target area.

[0126] The above information may be information that the user inputs using the input device 14 (keyboard, mouse, touch panel, etc.) of the information processing device 10A by utilizing a menu screen displayed on the output device (display, etc.) 16 of the information processing device 10A.

[0127] Next, in optional ST212, the information processing device 10A can acquire and store (e.g., in the auxiliary storage device 15) the following information (9) and information (10) described above by inputting the information acquired in ST210, i.e., information (13) and at least one of information (14) to information (16), into a trained third learning model or a trained third decision tree model. In this case, information (13) may be essential information.

[0128] (9) Information indicating multiple pesticide compositions constituting the agent to be sprayed in a specific area (zone) (10) Information indicating the amount of each of the multiple pesticide compositions shown in information (9).

[0129] Optionally, the information processing device 10A can input the information acquired in ST210, i.e., information (13) and at least one of information (14) to information (16), into a trained third learning model or a trained third decision tree model, thereby acquiring and storing (for example, in the auxiliary storage device 15) at least one of information (11) and information (12) in addition to the above-described information (9) and information (10). In this case, information (13) may be essential information.

[0130] (9) Information indicating multiple pesticide compositions constituting the agent to be sprayed in a specific area (zone) (10) Information indicating the amount of each of the multiple pesticide compositions shown in information (9). (11) Optimal continuous processing information regarding the optimal method for continuously spraying (processing) the agent mixed according to information (9) and information (10). (12) Optimal treatment timing information regarding the most suitable time to spray (treat) the chemicals mixed in accordance with information (9) and information (10).

[0131] The third learning model (e.g., a machine learning / deep learning model having an input layer, multiple intermediate layers, and an output layer) and the third decision tree model each undergo pre-training using information (13) to information (16) (input data) and information (9) to information (12) (output data), as well as a large number (e.g., tens of thousands or more) of training data. Such a third learning model or third decision tree model can be trained by optimizing a large number of coefficients (or a large number of parameters) used in the third learning model or third decision tree model so as to reduce the error between the data output when input data in each training data is input and the output data in that training data.

[0132] In a first example, the information processing device 10A can have such a trained third learning model and a trained third decision tree model. In a second example, the information processing device 10A can receive (acquire) such a trained third learning model and a trained third decision tree model stored in a device other than the device in which the information processing device 10A is installed, via the network 2. In a third example, the information processing device 10A can receive (acquire) information (9) to information (12), which are output data of the third learning model or the third decision tree model, by transmitting necessary information (e.g., information (13) to information (16)) via the network 2 to the trained third learning model or the trained third decision tree model stored in a device other than the device in which the information processing device 10A is installed.

[0133] In ST212, the information processing device 10A can also display the acquired information (9) and information (10) (and further information (11) and / or information (12)) to the user via the output device 16.

[0134] In the above-described ST208, the information processing device 10A can identify the multiple pesticide compositions constituting the agent to be sprayed on the specific area and the amount of each pesticide composition, respectively, based on information (9) and information (10). At this point, the information processing device 10A can achieve its intended purpose. However, by executing the above-described optional ST210 and ST212, the information processing device 10A can input information (13) and at least one of information (14) to information (16) into a trained third learning model or a trained third decision tree model, thereby acquiring information (9) and information (10) based on this information. In other words, the information processing device 10A can acquire the multiple pesticide compositions constituting the agent to be sprayed on the specific area and the amount of each pesticide composition based on information (13) and at least one of information (14) to information (16). As a result, the user can recognize two types of information as options, including information (9) and information (10) (and information (11) and / or information (12)) displayed on the information processing device 10A in ST208, and information (9) and information (10) (and information (11) and / or information (12)) displayed on the information processing device 10A in ST212.

[0135] 3-3. Operations performed in ST300 In this ST300, the mobile spraying device 20 can spray a pesticide prepared by mixing the pesticide compositions determined in ST200 in determined amounts over a specific area (or zone). Before describing the operations executed in ST300, the hardware configuration of the mobile sprinkling device 20 will be briefly described with reference to FIG.

[0136] Fig. 8 is a block diagram showing an example of the hardware configuration of mobile spraying device 20 shown in Fig. 1. As shown in Fig. 8, mobile spraying device 20 includes the above-mentioned information processing device 10A, a drive source 21 and a drive mechanism 22 for moving mobile spraying device 20, and a GPS (Global Positioning System) for acquiring location information relating to the current location of mobile spraying device 20. ing System) unit 23 and a plurality of tanks each containing a unique pesticide composition. The apparatus may include a tank 24, a preparation device 25 that mixes any two or more of the pesticide compositions from the tanks 24 to prepare a pesticide, and one or more nozzles 26 that spray the pesticide supplied from the preparation device 25.

[0137] The driving source 21 may be an engine and / or a motor that is capable of generating and providing to the driving mechanism 22 a driving force for moving the mobile sprinkling device 20 under the control of the information processing device 10A.

[0138] The drive mechanism 22 is any mechanism that converts the driving force supplied from the drive source 21 into propulsion force for the mobile spraying device 20, and may include, but is not limited to, gears, shafts, tires, caterpillars, and / or propellers. The drive mechanism 22 is also any mechanism that controls the direction and / or speed of movement of the mobile spraying device 20 under control of the information processing device 10A, and may include, but is not limited to, a transmission, a reducer, a steering mechanism, flaps, and / or a rudder.

[0139] The GPS unit 23 is controlled by the information processing device 10A and can provide the information processing device 10A with location information relating to the current location of the mobile spraying device 20 using well-known GPS technology.

[0140] Each of the plurality of tanks 24 may contain a pesticide composition unique to that tank from a plurality of pesticide compositions.

[0141] The preparation device 25 can be connected to each of the plurality of tanks 24 via pipes, valves, and sensors that the preparation device 25 has. The preparation device 25, under the control of the information processing device 10A, can prepare a pesticide by obtaining and mixing a plurality of pesticide compositions from any two or more of the plurality of tanks 24.

[0142] Specifically, the preparation device 25 may have valves that are provided in pipes connecting the preparation device 25 to each of the multiple tanks 24 and that can be opened and closed under the control of the information processing device 10A. The preparation device 25 can open a valve designated by the information processing device 10A and measure the amount of pesticide composition passing through that valve using a sensor. The preparation device 25 can return the valve to a closed state when the amount of pesticide composition measured by the sensor reaches the amount designated by the information processing device 10A. In this way, the preparation device 25 can prepare a pesticide by obtaining and mixing the amount of pesticide composition stored in any of the multiple tanks 24 designated (determined) by the information processing device 10A, in the amount designated (determined) by the information processing device 10A.

[0143] Each of the one or more nozzles 26 is coupled to the preparation device 25. Each nozzle 26 is capable of dispersing (spraying) the medicine supplied from the preparation device 25 under the control of the information processing device 10A.

[0144] Next, the operations executed by the mobile spraying device 20 having the above configuration will be described with reference to Fig. 9. Fig. 9 is a flow chart specifically showing an example of a part (ST300) of the operations executed by the determination system 1 shown in Fig. 3.

[0145] First, in ST302, the information processing device 10A can load (acquire) from the auxiliary storage device 15 the multiple pesticide compositions and the amounts of each pesticide composition determined in ST200 for the specific region (or each zone).

[0146] Next, in ST304, information processing device 10A can acquire position information indicating the position of the specific area (or each zone). In one example, this position information can be calculated based on position information (of the target area) assigned to the image captured by the device that generated the image in ST102, and the relative positional relationship of the specific area (each zone) with respect to the target area derived from the image. In another example, this position information can be input by the user via input device 14 of information processing device 10A at any timing, and acquired by information processing device 10A.

[0147] Next, in ST306, information processing device 10A can first use GPS unit 23 to acquire current location information indicating the current location of mobile spraying device 20. Furthermore, information processing device 10A can determine a travel route (information regarding the locations corresponding to multiple points to be passed through sequentially) based on the current location acquired in this manner and the location of the specific area (or each zone) acquired in ST304. Such a travel route can be determined, for example, using technology used in well-known navigation systems. Here, in the case of spraying a chemical agent in multiple zones, information processing device 10A can determine a travel route that passes through all of the multiple unit areas included in each zone. For example, in the example shown in FIG. 6, the travel route can be determined so that for first zone 130, all of unit areas 120A, 120B, and 120C are passed; for second zone 131, all of unit areas 120D, 120E, and 120F are passed; and for third zone 132, all of unit areas 120G and 120H are passed.

[0148] In ST308, information processing device 10A controls drive source 21 and drive mechanism 22 to start movement of mobile spraying device 20. Furthermore, information processing device 10A controls drive source 21 and drive mechanism 22 to move mobile spraying device 20 along the movement route determined in ST306. This can be achieved by controlling drive source 21 and / or drive mechanism 22 so that the current position of mobile spraying device 20 provided by GPS unit 23 is aligned along the movement route determined in ST306.

[0149] In ST310, the information processing device 10A can control the preparation device 25, for example, when approaching a specific area (or each zone), so as to mix a plurality of pesticide compositions determined to correspond to the specific area (or each zone) in the determined amounts of each pesticide composition. This allows the preparation device 25 to perform the above-mentioned operation and mix a plurality of pesticide compositions determined for the specific area (or each zone) in the determined amounts of each pesticide composition, thereby preparing a drug corresponding to the specific area (or each zone).

[0150] Furthermore, when the difference between the current position of the mobile spraying device 20 and the position of the specific area (or each zone) becomes equal to or less than a threshold value, the information processing device 10A can transmit a signal to the nozzle 26 instructing the nozzle to be in an open state. This allows the nozzle 26 to spray (spray) the agent supplied from the preparation device 25.

[0151] Furthermore, information processing device 10A can continue to send a signal to nozzle 26 instructing it to open while the difference between the current position of mobile spraying device 20 and the position of that specific area (or each zone) is equal to or less than a threshold value. This allows mobile spraying device 20 to spray the agent corresponding to that specific area (or each zone) while it is in a position that coincides with the position of that specific area (or each zone).

[0152] Thereafter, when the difference between the current position of the mobile spraying device 20 and the position of the specific area (or each zone) exceeds a threshold value, the information processing device 10A can transmit a signal to the nozzle 26 instructing it to close the nozzle, thereby enabling the nozzle 26 to stop spraying the agent.

[0153] In the case where the mobile spraying device 20 sprays the agent in a plurality of specific areas (or a plurality of zones), ST308 and ST310 may be executed sequentially for each specific area (each zone).

[0154] The operations performed by information processing device 10A on a specific area as described above can be applied to operations performed by information processing device 10A on a zone.

[0155] 4. Operation of the decision system (part 2) So far, we have explained an embodiment (first embodiment) in which multiple pesticide compositions and the amount of each pesticide composition to be contained in a pesticide to be sprayed in a specific area (or zone) included in a target area are determined. However, this first embodiment is similarly applicable to an embodiment (second embodiment) in which multiple pesticide compositions and the amount of each pesticide composition to be contained in a pesticide to be treated (applied) on rice seed to be directly sown in a specific area (or zone) included in the target area (or on rice seed to be sown in a nursery box for transplanting seedlings into the specific area) are determined. In this Section 4, the target area refers to the area where rice seed is directly sown and / or the area where seedlings grown from the nursery box into which the rice seed is sown are transplanted, and may be one field, a partial area within one field, and / or multiple fields.

[0156] In order to avoid redundant explanation, the second aspect will be briefly described below, focusing on only the parts that are different from the first aspect described above, with reference to Fig. 10. Fig. 10 is a flow chart showing another example of the operation executed by the determination system 1 shown in Fig. 1.

[0157] 10 , the operations performed by the determination system 1 can broadly include ST1000, which acquires information about a target area (or a specific area included in the target area), and ST2000, which determines, based on the information acquired in ST1000, a plurality of pesticide compositions constituting the pesticide and the amount of each pesticide composition for treating rice seeds to be directly sown in the target area (or the specific area included in the target area) or rice seeds to be sown in seedling boxes for transplanting seedlings into the target area (or the specific area). Furthermore, as an option, the operations performed by the determination system 1 can include ST3000, which treats rice seeds to be directly sown in the target area (or the specific area included in the target area) or rice seeds to be sown in seedling boxes for transplanting seedlings into the target area (or the specific area) with a pesticide prepared by mixing determined amounts of each of the plurality of pesticide compositions determined in ST2000.

[0158] First, attention is focused on ST1000. Fig. 11 is a flow diagram specifically showing an example of a part (ST1000) of the operations executed by the determination system 1 shown in Fig. 10. The operation shown in Fig. 11 does not include ST102 and ST104 shown in Fig. 4.

[0159] ST1002 corresponds to ST106 shown in FIG. 4. However, in ST1002, for each of a plurality of unit areas included in the target area, the "unit type information" indicating at least one type of harm that has occurred in this unit area may be, for example, past information (such as the previous year) actually collected in relation to the unit area. In one example, such unit type information may be input by a user via the input device 14 of the information processing device 10A and acquired by the information processing device 10A. In another example, such unit type information may be acquired by the information processing device 10A via the network 2 from any other device (such as a server device) that collects and stores such information.

[0160] ST1004 corresponds to ST108 shown in FIG. 4. However, in ST1004, for each of a plurality of unit areas included in the target area, the "unit amount information" indicating the amount of each of at least one harm that occurred in this unit area may be, for example, information actually obtained in the past (the previous year, etc.). In one example, such unit amount information may be input by a user via the input device 14 of the information processing device 10A and acquired by the information processing device 10A. In another example, such unit type information may be acquired by the information processing device 10A via the network 2 from any other device (such as a server device) that collects and stores such information.

[0161] ST1006 and ST1008 correspond to ST110 and ST112 shown in FIG. 4, respectively.

[0162] Returning to Fig. 10, attention will now be focused on ST2000. Fig. 12 is a flow chart specifically showing an example of part (ST2000) of the operations executed by the decision system 1 shown in Fig. 10.

[0163] ST2002 to ST2006 respectively correspond to ST204 to ST208 shown in Fig. 7. Only the differences between ST2002 to ST2006 and those described above with reference to Fig. 7 will be explained below.

[0164] In ST2002, the information processing device 10A can acquire the following information (1) to information (5) to be input to the trained first learning model or the trained first decision tree model.

[0165] (1) Soil information on the type of soil in a specific area (zone) (2) Information on past harms that have occurred in a specific area (zone) (3) Information on the prevalence of pests resistant to drugs in specific areas (zones) (4) Weather forecast information regarding weather data predicted for the rice cultivation period in a specific area (zone).

[0166] These pieces of information (1) to (4) correspond to the information (1), information (2), information (4), and information (5) described above in relation to ST204, respectively.

[0167] Next, in ST2004, the information processing device 10A can acquire the following information (5) to information (7) by inputting at least one of the information acquired in ST2002 (i.e., at least one of information (1) to information (4)) into a trained first learning model or a trained first decision tree model.

[0168] (5) Information on the type and amount of damage that may occur in a specific area (zone). (6) Information on the predicted growth period of this pest (7) Information on phytotoxicity occurring in specific areas (zones)

[0169] These pieces of information (5) to (7) correspond to the pieces of information (6) to (8) described above in relation to ST206, respectively.

[0170] The first learning model (e.g., a machine learning / deep learning model having an input layer, multiple intermediate layers, and an output layer) and the first decision tree model each undergo training in advance using a large number (e.g., tens of thousands or more) of training data including information (1) to (4) (input data) and information (5) to (7) (output data). Such a first learning model or first decision tree model can perform training by optimizing a large number of coefficients (or a large number of parameters) used in the first learning model or first decision tree model so as to reduce the error between the data output when input data in each training data is input and the output data in that training data.

[0171] Next, in ST2006, the information processing device 10A inputs the information (5) to information (7) acquired in ST2004 into a trained second learning model or a trained second decision tree model, thereby acquiring and storing (for example, in the auxiliary storage device 15) the information (8) and information (9) shown below.

[0172] (8) Information indicating a plurality of pesticide compositions constituting the agent to be applied to the rice seeds to be directly sown in a specific area (or information indicating a plurality of pesticide compositions constituting the agent to be applied to the rice seeds to be sown in a nursery box for transplanting seedlings in a specific area) (9) Information indicating the amount of each of the multiple pesticide compositions shown in information (8).

[0173] Such second learning model and second decision tree model are each trained in advance using a large number (e.g., tens of thousands or more) of training data including information (5) to (7) (input data) and information (8) to (9) (output data). Such second learning model or second decision tree model can be trained by optimizing a large number of coefficients (or a large number of parameters) used in the second learning model or second decision tree model so as to reduce the error between the data output when input data in each training data is input and the output data in that training data.

[0174] Returning to FIG. 10 again, in ST3000, a pesticide can be prepared by mixing the multiple pesticide compositions determined for a specific area (zone) in the ST2000 in the amounts of each pesticide composition determined. In one example, the pesticide prepared in this manner is applied to rice seeds to be directly sown in the specific area (zone). The rice seeds treated in this manner are then sown in the specific area (zone). In another example, the pesticide prepared in this manner is applied to rice seeds, and the treated rice seeds are then sown in a seedling box. The seedlings grown in the seedling box are then transplanted to the specific area (zone).

[0175] The operations performed by information processing device 10A on a specific area as described above can be applied to operations performed by information processing device 10A on a zone.

[0176] As described above, according to the various embodiments disclosed in the present application, for a chemical to be sprayed in a specific area (zone) included in a target area where paddy rice is cultivated, the multiple pesticide compositions constituting the chemical and the amount of each pesticide composition can be determined at any timing. Furthermore, for a chemical to be applied to rice seeds to be directly sown in a specific area (zone) included in a target area or to rice seeds to be sown in seedling boxes for transplanting seedlings into this specific area (zone), the multiple pesticide compositions constituting the chemical and the amount of each pesticide composition can be determined at any timing.

[0177] 5. Variations In the example described with reference to FIG. 4, ST102 to ST112 can be executed by the same information processing device 10A. However, ST102 to ST112 can be shared and executed by an information processing device 10A mounted on the mobile spraying device 20, an information processing device 10A mounted on the terminal device 30, an information processing device 10A mounted on the server device 10, and / or any other device connectable to network 2 (such as another server device). In this case, each device that shares and executes ST102 to ST112 can transmit necessary information via network 2 to the device that will execute the next process. This is similarly applicable to the example described with reference to FIG. 11.

[0178] In the example described with reference to FIG. 7, ST202 to ST212 can be executed by the same information processing device 10A. However, ST202 to ST212 can be shared and executed by an information processing device 10A mounted on the mobile spraying device 20, an information processing device 10A mounted on the terminal device 30, an information processing device 10A mounted on the server device 10, and / or any other device connectable to network 2 (such as another server device). In this case, each device that shares and executes ST202 to ST212 can transmit necessary information via network 2 to the device that will execute the next process. This is similarly applicable to the example described with reference to FIG. 12.

[0179] The above has been described as a case where a drug is prepared by mixing multiple pesticide compositions. However, the technology disclosed in the present application is also applicable to a case where only one pesticide composition is selected from multiple pesticide compositions and a drug is prepared using the selected one pesticide composition in a determined amount. In this case, for example, the information (9) and information (10) used in the learning model described in ST208 can be changed as follows. (9) Information indicating one pesticide composition constituting the agent to be sprayed in a specific area (zone). (10) Information indicating the amount of one pesticide composition shown in information (9). Similarly, "information indicating multiple pesticide compositions" used in various learning models can be replaced with "information indicating one pesticide composition," and "information indicating the amount of each of multiple pesticide compositions" can be replaced with "information indicating the amount of one pesticide composition."

[0180] The inventors confirmed that there is a correlation between the data input to each learning model described in this application and the data output from this learning model, and as a result, adopted a configuration in which such data is used for each learning data.

[0181] As would be easily understood by a person skilled in the art having the benefit of this disclosure, the various examples described above can be appropriately combined with one another in various patterns unless a contradiction occurs.

[0182] <2> Pesticide composition The pesticide composition according to the present invention is a liquid pesticide composition containing the following components (a) and (b1), or a solid pesticide composition containing the following components (a) and (b2): (a) at least one active ingredient selected from the group consisting of an active ingredient having a herbicidal effect, an active ingredient having an insecticidal effect, and an active ingredient having a fungicidal effect; (b1) 1 to 200 g / L of anionic dispersant; (b2) 1 to 20% w / w of an anionic dispersant.

[0183] The pesticide composition of the present invention may contain one or more of the active ingredients described above, and may also contain one or more anionic dispersants.

[0184] As described above, by mixing the anionic dispersant at a specific content, sufficient dispersibility of the pesticide composition in the field can be ensured, which makes it possible to reduce phytotoxicity caused by local application and the environmental burden caused by the active ingredient and auxiliary ingredients.

[0185] When the pesticide composition is a liquid pesticide composition, the formulation is, for example, an aqueous suspension or an oily suspension. Specific formulations of liquid pesticide compositions include, for example, suspension concentrates (SCs).

[0186] When the pesticide composition is a solid pesticide composition, the formulation is, for example, a wettable powder or a granule. Specific formulations of the solid pesticide composition include, for example, a water dispersible granule (WG) and a floating granule. In one embodiment, the solid pesticide composition is a self-dispersible pesticide. It is a wettable powder that has at least one of diffusibility and floating properties. By having such properties, it is possible to uniformly control pests and weeds in the field where the pesticide is sprayed.

[0187] (a) Active ingredient The pesticide composition of the present invention contains at least one selected from the group consisting of (a1) an active ingredient having a herbicidal effect, (a2) an active ingredient having an insecticidal effect, and (a3) ​​an active ingredient having a fungicidal effect.

[0188] (a1) Active ingredient having herbicidal effect Examples of active ingredients having herbicidal effects include, but are not limited to, tefuryltrione, triafamone, fentrazamide, clomeprop, oxadiazon, ipfencarbazone, cafenstrole, indanofan, fenoxasulfone, mefenacet, butachlor, pretilachlor, fenquinotrione, benzobicyclon, sulcotrione, mesotrione, pyrazolate, benzofenap, pyrimisulfan, ethoxysulfuron, bensulfuron methyl, propyrisulfuron, metazosulfuron, penoxsulam, oxadiargyl, pyraclonil, pentoxazone, florpyrauxifenbenzyl, fenoxaprop ethyl, cyhalofop butyl, metamifop, cyclopyrimorate, dimethamethrin, simetryn, bentazon, oxaziclomefone, bromobutide, and tetflupyrolimet. The active ingredient having herbicidal effect is at least one selected from the group consisting of these compounds.

[0189] The pesticide composition of the present invention may contain a plurality of active ingredients (a1) having herbicidal effects. In a preferred embodiment, the pesticide composition of the present invention contains one or more active ingredients (a1) having herbicidal effects selected from the group consisting of tefuryltrione, triafamone, fentrazamide, clomeprop, oxadiazon, and ethoxysulfuron.

[0190] (a2) Active ingredient with insecticidal effect Examples of active ingredients having insecticidal effects include, but are not limited to, imidacloprid, thiacloprid, dinotefuran, flupyradifurone, flupirimine, nitenpyram, clothianidin, sulfoxaflor, ethiprole, fipronil, spinosad, tetraniliprole, chlorantraniliprole, cyantraniliprole, pymetrozine, triflumezopyrim, benzpyrimoxane, and oxazosulfil. The active ingredient having insecticidal effects is at least one selected from the group consisting of these compounds. The pesticide composition of the present invention may contain multiple active ingredients (a2) having insecticidal effects.

[0191] (a3) Active ingredient with bactericidal effect Examples of active ingredients having fungicidal effects include, but are not limited to, isotianil, probenazole, tricyclazole, diclobenthiazox, penflufen, thifluzamide, impirfluxam, kasugamycin, validamycin, fthalide, metominostrobin, azoxystrobin, and pencycuron. The active ingredient having fungicidal effects is at least one selected from the group consisting of these compounds. The pesticide composition of the present invention may contain multiple active ingredients (a3) ​​having fungicidal effects.

[0192] When the pesticide composition of the present invention contains a plurality of active ingredients selected from the group consisting of the above-mentioned components (a1), (a2) and (a3), the content specified for the active ingredients is the total content of each active ingredient.

[0193] When the pesticide composition of the present invention is a liquid pesticide composition, it contains, for example, 10 to 700 g / L, preferably 50 to 600 g / L, and more preferably 200 to 600 g / L of the active ingredient (a).

[0194] When the pesticide composition of the present invention is a solid pesticide composition, it contains, for example, 10 to 70% w / w, preferably 50 to 60% w / w, and more preferably 20 to 60% w / w of the active ingredient (a).

[0195] (b) Anionic dispersants Examples of anionic dispersants include, but are not limited to, alkylnaphthalenesulfonic acid derivatives and ligninsulfonic acid derivatives. The anionic dispersants may be synthetic or commercially available. Examples of commercially available alkylnaphthalenesulfonic acid derivatives of anionic dispersants include Demol SNB, Newkalgen PS-P, Newkalgen WG-101, Newkalgen BX-C, AEROSOL OS, MORWET D425, MORWET IP, SUPRAGIL WP, SUPRAGIL MNS / 90, and TERSPERSE 2020. Examples of commercially available lignin dispersants of anionic dispersants include Demol SNB, Newkalgen PS-P, Newkalgen WG-101, Newkalgen BX-C, AEROSOL OS, MORWET D425, MORWET IP, SUPRAGIL WP, SUPRAGIL MNS / 90, and TERSPERSE 2020. Examples of sulfonic acid derivatives include Newkalgen RX-B, Newkalgen WG-4, Borresperse CA, Borresperse NA, and Marasperse. CBOS-4, POLYFON H, POLYFON T, POLYFON O, UFOX ANE 3A, VANISPERSE CB, etc. The anionic dispersant is at least one selected from the group consisting of the above-mentioned compounds.

[0196] When the pesticide composition of the present invention contains a plurality of anionic dispersants, the content of the anionic dispersants specified is the total content of the respective anionic dispersants.

[0197] When the pesticide composition of the present invention is a liquid pesticide composition, it contains, for example, 1 to 200 g / L, preferably 5 to 150 g / L, and more preferably 50 to 150 g / L of an anionic dispersant.

[0198] When the pesticide composition of the present invention is a solid pesticide composition, it contains, for example, 1 to 20% w / w, preferably 5 to 20% w / w, more preferably 5 to 15% w / w of an anionic dispersant.

[0199] In a preferred embodiment, the pesticide composition of the present invention is designed so that the application rate of the anionic dispersant when applied to a field is 9 to 300 g / ha. By using such an application rate, when the spray solution after single use or mixed use is sprayed in a field, particularly in a paddy field, it is possible to impart sufficient diffusion performance to the pesticide composition while avoiding phytotoxicity to rice. In a more preferred embodiment, the pesticide composition of the present invention is designed so that the application rate of the anionic dispersant when applied to a field is 9 to 150 g / ha. In an even more preferred embodiment, the pesticide composition of the present invention is designed so that the application rate of the anionic dispersant when applied to a field is 9 to 130 g / ha.

[0200] The pesticide composition of the present invention can ensure sufficient dispersibility even when applied locally to paddy fields, and can reduce the environmental load caused by the active ingredient and / or auxiliary ingredients. Furthermore, because the auxiliary ingredients used are the same, even when multiple pesticide compositions of the present invention are mixed and applied, the environmental load caused by using more auxiliary ingredients than necessary can be reduced.

[0201] The pesticide composition of the present invention may further contain auxiliary ingredients, such as surfactants, antifreeze agents, thickeners, preservatives, antifoaming agents, safeners, and carriers. These auxiliary ingredients may be synthetic or commercially available.

[0202] The surfactant may be, for example, a nonionic surfactant. The pesticide composition of the present invention may contain one or more nonionic surfactants. Examples of nonionic surfactants include polyoxyethylene-polyoxypropylene block polymers, polyalkylene oxide block polymers, polyoxyalkylene alkylphenyl ethers, polyoxyalkylene fatty acid esters, polyoxyalkylene tristyrylphenyl ethers, polyoxyalkylene alkylamines, polyoxyethylene alkanediols, acetylene glycol, polyoxyethylene acetylene glycol, sorbitan fatty acid esters, sucrose fatty acid esters, polyoxyalkylene sorbitan esters, and glycerin fatty acid esters. Specific examples of commercially available nonionic surfactants include Surfynol 104, Surfynol 420, Surfynol 440, Newkalgen TG-310, ATLAS G 5000, DOWFAX 100N50, GENAPOL 10500, PLURONIC (registered trademark) F127, PLURONIC (registered trademark) L62, PLURONIC (registered trademark) P105, and PLURONIC (registered trademark) PE 6. 200, PLURONIC (registered trademark) PE 10500, SYNPERONIC PE / F 127, SOPROPHOR BSU, SOPROPHOR S / 40P, STEP-FLOW 26, SYNPERONIC PE / L 62, TERMUL 5429, ULTR ARIC PE 62, ULTRARIC PE 105, Toho Chemical Co., Ltd. SORPOL T Examples include the series products, various SPAN products, Mitsubishi Chemical Corporation's Ryoto Sugar Ester product, Daiichi Kogyo Seiyaku Co., Ltd.'s DK Ester products, and Kao Corporation's Leodor SP series products.

[0203] Examples of antifreeze agents include urea, glycerin, polyglycerin and polyglycerin derivatives, ethylene glycol, propanediol, and propylene glycol.

[0204] Specific examples of thickeners include Gohsenol GL-05, Diutan Gum, Leocrysta, FBP-34, Welan Gum, Kelzan, Kelzan BT, and KU. NIPIA F, KUNIPIA G, RHODOPOL G, RHODOPOL 23, RHODOPOL 50 MC, SATIAXANE CX911, VAN GEL B, V EEGUM R, VOLCLAY HPM-20, WELAN GUM BG3810, and EXILVA.

[0205] Specific examples of preservatives include Biohope, Acticide B 20, Bronopol, Kathon CG / ICP, Preventol Bit 20 N, and Preservatives. VENTOL D 2, PREVENTOL D 7PROXEL GXL and PROXE Examples include L GXL(S).

[0206] Antifoaming agents include silicone oil and calcium stearate. Specific examples of commercially available antifoaming agents include Antifoam E-20 and Antifoam 8. 830 FOOD GRADE, SAG10, SAG30, SAG1 1572, SILC Examples include OLAPSE 426R, SILCOLAPSE 432, SILCOLAPSE 454, SILCOLAPSE 482, SILFOAM SE2, SILFAR SE 4 and SILFOAM SRE.

[0207] Examples of carriers include talc, kaolin clay, silica, clay, chalk, quartz, attapulgite, montmorillonite, diatomaceous earth, calcium carbonate, resin, wax, water, alcohol, organic solvent, mineral oil, and vegetable oil, and are selected appropriately depending on the formulation of the pesticide composition.

[0208] When the pesticide composition of the present invention is a solid pesticide composition, it may further contain the following ingredients: for example, citric acid, malic acid, hydrochloric acid, sulfuric acid, sodium hydroxide, aqueous ammonia, ammonium sulfate, binders (e.g., starch, CMC, PVA, polyurethane, PVP, etc.), granulating / disintegration improving agents (e.g., anionic surfactants), disintegration improving agents (e.g., anionic surfactants, CMC), disintegration extending agents (e.g., polyacrylates), etc.

[0209] Another aspect of the present invention relates to a method for preparing an agrochemical composition, the method comprising mixing an active ingredient (a) and an anionic dispersant (b). Specifically, the method for preparing the agrochemical composition comprises: a first mixing step of mixing the active ingredient (a) and the anionic dispersant (b); A second mixing step is included in which auxiliary ingredients are added to the mixture obtained in the first mixing step and mixed.

[0210] When the pesticide composition of the present invention is a liquid pesticide composition, it can be prepared, for example, by mixing and grinding the active ingredient, an anionic dispersant, a carrier (e.g., water), and other ingredients such as an antifoaming agent, and then adding a separate mixture of water, a thickener, and a preservative.

[0211] When the pesticide composition of the present invention is a solid pesticide composition, it can be prepared, for example, by mixing the active ingredient, anionic dispersant, carrier (e.g., talc and / or kaolin clay) and other ingredients, dry-grinding them, adding a separate mixture of a liquid wetting agent and water to the ground product, mixing and kneading, granulating, drying and sieving. For example, when the solid pesticide composition is a water dispersible granule (WG), it can be prepared by conventional methods such as spray drying, fluidized bed granulation, pan granulation, mixing using a high-speed mixer, and extrusion without using a solid inert substance.

[0212] In another embodiment, the method for producing the pesticide composition of the present invention comprises the steps of: a mixing step of mixing the active ingredient (a) with the anionic dispersant (b); a grinding step in which grinding is performed after the mixing step; and a further mixing step of adding and mixing anionic dispersant (b) after the grinding step. By adding and mixing the anionic dispersant in portions before and after grinding in this manner, heat generation due to grinding in the grinding liquid can be avoided and the product temperature can be lowered. It is preferable to keep the heat generation due to grinding to less than 10°C, for example. This production method can be used for both SC and WG. Preferably, this production method can be used for SC.

[0213] The pesticide composition according to the present invention may be a mixture prepared by mixing a plurality of pesticide compositions encompassed by the present invention. Therefore, yet another aspect of the present invention relates to a method for producing a pesticide composition, which comprises mixing a plurality of the above-mentioned pesticide compositions. When a plurality of pesticide compositions according to the present invention are mixed, the auxiliary components used are the same, and therefore the environmental burden caused by using more auxiliary components than necessary can be reduced.

[0214] <3> Methods for controlling diseases, pests and / or weeds Another aspect of the present invention relates to a method for controlling diseases, pests, and / or weeds using the aforementioned pesticide composition. More specifically, the method relates to a method for controlling diseases, pests, and / or weeds, comprising applying one or more of the aforementioned pesticide compositions to a field. Here, the anionic dispersant is applied to the field at an application rate of, for example, 9 to 300 g / ha, preferably 9 to 150 g / ha, and more preferably 9 to 130 g / ha. By applying the amount within this range, sufficient diffusion performance can be imparted to the pesticide composition when the spray solution after single use or mixed use is sprayed on a field, particularly a paddy field.

[0215] Examples of rice diseases include, but are not limited to, rice blast (Pyricularia oryzae).

[0216] Examples of rice pests include, but are not limited to: Hemiptera: green rice leafhopper (Nephotettix cincticeps), brown planthopper (Nilaparvata lugens), small brown planthopper (Laodelphax striatellus (Fallen)), white-backed planthopper (Sogatella furcifera (Horvath)), etc. Coleoptera: Rice leaf beetle (Oulema oryzae (Kuwayama)), rice water weevil (Lissorhoptrus oryzophilus Kuschel), etc.

[0217] Weeds refer to plants that are undesirable for the growth of crop plants growing in a field, and "weed control" refers to controlling undesirable plants or regulating the growth of such plants. In a method for controlling weeds, one or more pesticide compositions of the present invention are applied to an area where weeds (e.g., harmful plants such as monocotyledonous or dicotyledonous weeds or undesirable crop plants), seeds (e.g., grains, seeds, or vegetative propagules such as tubers or shoot parts bearing sprouts) or crop plants grow (e.g., an area under cultivation).

[0218] Examples of paddy field weeds include, but are not limited to: Dicotyledons of the following genera: Polygonum, Rorippa, Rotala, Lindernia, Bidens, Dopatrium, Eclipta, Elatine, Gratiola, Lindernia, Ludwigia, Oenanthe ), Ranunculus, Deinostema, etc. Monocotyledons of the following genera: Echinochloa, Panicum, Poa, Cyperus, Monochoria, Fimbristylis, Sagittaria, Eleocharis, Scirpus, Alisma, Aneilema, Blyxa, Eriocaulon, Potamogeton, etc.

[0219] More specifically, the pesticide composition of the present invention can be used against, for example, the following typical paddy field weeds: Dicotyledonous plants: Rotala indica Koehne, Lindernia procumbens Philcox, Ludwigia prostrata Roxburgh, Potamogeton distinctus A. Benn, Elatine triandra Schk, Oenanthe javanica, Monocotyledonous plants: Barnyard grass (Echinochloa oryzicola Vasing), Monochoria vaginalis Presl, Eleocharis acicularis L., Eleocharis Kuroguwai Ohwi, Cyperus difformis L., Cyperus serotinus Rottboel, Sagittaria pygmaea Miq, Arrowhead weed (Alisma canaliculatum A. Br. et Bouche), and Scirpus juncoides Roxburgh.

[0220] In the control method of the present invention, application may be carried out by any means, for example, manual spraying or automatic spraying by manned or unmanned aircraft or vehicles.

[0221] In the control method of the present invention, the field is preferably a paddy field. application , more preferably, localized on the water surface of the rice paddy application is.

[0222] In the control method of the present invention, the anionic dispersant is preferably applied to the field at an application rate of 9 to 250 g / ha. By applying the amount within this range, the pesticide composition can be more efficiently dispersed in the field.

[0223] In the control method according to the present invention, the active ingredient is applied to the field at an application rate of, for example, 5 to 1000 g / ha, preferably 15 to 500 g / ha. In one embodiment, the active ingredient is applied to the field at an application rate of 50 to 650 g / ha. The application rate of the active ingredient can be appropriately selected from the above range depending on, for example, the climate, the type of disease, pests and / or weeds in the field, and the application time.

[0224] In the control method of the present invention, the pesticide composition to be applied is automatically selected or selected (for example, by a user) based on, for example, predicted occurrence of diseases, pests, and / or weeds in the field and / or their occurrence status. The pesticide composition to be automatically selected or selected may be one or more. The automatic selection may be performed by the aforementioned information processing device.

[0225] In one embodiment, in the control method according to the present invention, the pesticide composition is automatically metered and applied to the field. In another embodiment, multiple pesticide compositions are automatically metered and / or automatically mixed and applied to the field. The automatic metering and / or automatic mixing may be performed by the information processing device 10A and / or the mobile spraying device 20. In another embodiment in which the information processing device 10A and / or the mobile spraying device 20 are not used, the pesticide composition is metered by, for example, a user using an implement or device and applied to the field, or multiple pesticide compositions are metered and / or mixed by, for example, a user using an implement or device and applied to the field.

[0226] Another aspect of the present invention is a kit for use in the aforementioned control method, the kit comprising the aforementioned pesticide composition and a container. Preferably, the kit comprises: - a plurality of pesticide compositions as described above; Spatially separated multiple containers and a plurality of pesticide compositions are stored in each of the plurality of containers. An example of a kit according to the present invention is the plurality of tanks 24 shown in FIG. 8. With such a kit, an appropriate pesticide composition can be appropriately selected from a plurality of candidate pesticide compositions according to the conditions of the field and applied. Such a kit may be used in combination with the above-mentioned information processing device 10A, or may be incorporated into the above-mentioned mobile spraying device 20.

[0227] The present invention further relates to the use of the aforementioned pesticide composition for controlling diseases, pests and / or weeds. In one embodiment of the use of the pesticide composition of the present invention, the diseases, pests and / or weeds occur in paddy fields.

[0228] As described above, according to the present invention, it is possible to provide a pesticide composition that exhibits excellent diffusibility, particularly excellent diffusibility in water, and therefore, sufficient effects are exhibited even when applied locally, and phytotoxicity to crop plants can be avoided.

[0229] Furthermore, based on the forecast of disease, pest, and / or weed occurrence and their occurrence status, two or more pesticide compositions can be mixed in appropriate amounts depending on the disease, pest, and / or weed to be controlled, to obtain a stable dilution, which can then be applied to a field (e.g., a rice paddy). By applying the chemical selected for control depending on the field conditions, it is possible to avoid spraying excessive chemicals into the environment. Furthermore, by combining multiple pesticide compositions, it is possible to provide pesticide compositions with stable physical and / or chemical properties in the diluted solution, as well as pesticide compositions in the form of mixtures having different physical and chemical properties that can simultaneously control different diseases, pests, and / or weeds. Due to their excellent diffusibility, such pesticide compositions, when mixed and applied, allow the active ingredients to diffuse evenly and widely throughout a field (e.g., a rice paddy) and exert insecticidal, fungicidal, and / or herbicidal effects, thereby enabling disease, pest, and / or weed control while avoiding the spraying of excessive active ingredients that would be a burden on the environment. Furthermore, by applying multiple pesticide compositions within the scope of the present invention, the auxiliary components are common among the pesticide compositions, so that it is possible to avoid using more auxiliary components than necessary, which also reduces the environmental load.

[0230] Furthermore, by combining it with an information processing device and / or a spraying device, it becomes possible to select the type and application amount of pesticide that is more suitable for the disease, pest and / or weed to be controlled based on the prediction of the occurrence of the disease, pest and / or weed and its occurrence status, thereby achieving even more efficient control.

[0231] This application is based on Japanese Patent Application No. 2021-189672, entitled "Pesticide Composition, Information Processing Device, and Computer Program," filed on November 22, 2021, and claims the benefit of priority from this Japanese patent application. The entire contents of this Japanese patent application are incorporated herein by reference. [Example]

[0232] Based on the ingredients and compositions shown in the table below, water, active ingredients, dispersants, antifoaming agents, and other ingredients were mixed and pulverized, and then a separate mixture of water, thickeners, and preservatives was added and mixed to prepare the pesticide compositions of the Examples and Comparative Examples. In the Examples, the "application amount" corresponds to the "application amount." The abbreviations in the table are as follows:

[0233] The suspension concentrate (SC) was prepared by mixing and grinding the active ingredient, anionic dispersant, water, and other ingredients such as antifoaming agent, and then adding a separate mixture of water, thickener, and preservative. ) was prepared by mixing the active ingredient, anionic dispersant, carrier and other ingredients, dry-milling the mixture, adding a separate mixture of liquid wetting agent and water to the milled mixture, mixing and kneading it, granulating, drying and sieving.

[0234] [Table 1-1] [Table 1-2] [Table 1-3] [Table 1-4]

[0235] Test Example 1A The pesticide compositions of Examples 7 and 8 and the pesticide compositions of Comparative Examples 1 and 2 were applied locally to paddy fields at the formulation application rates and dosages shown in the table below. Local application was carried out on one short side of a 2 m x 13 m plot.

[0236] [Table 2-1]

[0237] When the anionic dispersant was applied at a rate of 4 g / ha (Comparative Example 1) and 0.8 g / ha (Comparative Example 2), localized application (treatment of one short side of a 2 m x 13 m plot) caused greater phytotoxicity at the application site than when applied at a rate of 55 g / ha (Example 7) and 9.167 g / ha (Example 8). These results indicate that anionic dispersants in a water dispersible granule formulation require at least 9 g / ha or more.

[0238] Test Example 1B Next, Example 8 and Comparative Example 2, which contain the same active ingredient triafamone but have different dosage forms and dispersant contents, were used to test the difference in effect due to dispersibility.

[0239] [Table 2-2]

[0240] As shown by the above results, when triafamone was contained as the active ingredient, the pesticide composition containing 0.8 g / ha of dispersant (Comparative Example 2) was insufficiently effective at locations distant from the application site, whereas the pesticide composition containing 9.167 g / ha of anionic dispersant (Example 8) was able to ensure sufficient effectiveness even at locations distant from the application site.

[0241] Test Example 1C In this test example, Example 2 and Comparative Example 2 were used to test the difference in phytotoxicity due to dispersibility. Example 2 and Comparative Example 2 were applied locally to paddy fields at the formulation application rates and dosages shown in the table below. Local application was performed on one side of a 1 m x 15 m plot. Both Example 2 and Comparative Example 2 are flowable formulations containing the same active ingredient, triafamone, but Example 2 has a higher dispersant concentration than Comparative Example 2.

[0242] [Table 2-3]

[0243] When the anionic dispersant was applied to the field at a rate of 0.8 g / ha (Comparative Example 2), phytotoxicity at the application site in localized application (applied to one short side of a 1 m x 15 m plot) was greater than when the rate was 10 g / ha (Example 2). This result indicated that anionic dispersant must be applied to the field at a rate of at least 10 g / ha.

[0244] Test Example 1D In this test example, the difference in phytotoxicity due to dispersibility when multiple formulations were mixed was examined. A mixture of Example 2 and Example 4, and a mixture of Comparative Examples 2 and 3, were locally applied to paddy fields at the formulation application rates and dosages shown in the table below. Localized application was performed on one short side of a 1 m x 15 m plot. Both Example 2 and Comparative Example 2 are flowable formulations containing the active ingredient triafamone, but Example 2 has a higher dispersant concentration than Comparative Example 2. Both Example 4 and Comparative Example 3 are flowable formulations containing the active ingredient fentrazamide, but Example 4 has a higher dispersant concentration than Comparative Example 3.

[0245] [Table 2-4]

[0246] When the total amount of anionic dispersant applied to the field was 8.8 g / ha (Comparative Example 2 + Comparative Example 3), the phytotoxicity at the application site in localized application (treatment of one short side of a 1 m x 15 m plot) was greater than when the total amount was 70 g / ha (Example 2 + Example 4). This result showed that the total amount of anionic dispersant applied to the field was insufficient at 8.8 g / ha.

[0247] Test Example 2 Several pesticide compositions were sprayed onto fields using a pesticide spraying drone, and their effectiveness against barnyard grass, Monochoria vaginalis, and Scirpus bulrush was tested six weeks after treatment. Similarly, comparisons were made with manual spraying of pesticides by humans. The pesticide compositions, formulation amounts, application rates, and dispersant amounts used are shown in the table below. Figure 13 also shows the drone spraying flight route with an arrow (→), and the effectiveness survey points are numbered 1 through 12.

[0248] [Table 3]

[0249] The above results demonstrate that drone spraying is just as effective as hand spraying. Furthermore, similar formulations can be applied to different active ingredients, and there were no problems even when multiple pesticide compositions containing at least one of these active ingredients were mixed. These results demonstrate that sufficient effectiveness can be achieved through a variety of application methods and topical application.

[0250] Test Example 3 When mixing the active ingredient, anionic dispersant, water, and other ingredients such as an antifoaming agent, the relationship between the amount of anionic dispersant added and the temperature of the grinding liquid during the grinding process was investigated. Specifically, for the pesticide composition of Example 3, the anionic dispersant was added in portions before and after grinding, and the relationship between the amount added and heat generation was investigated. The results are shown in the table below and Figure 14.

[0251] [Table 4-1]

[0252] [Table 4-2]

[0253] The above results show that adding the dispersant in a predetermined amount in portions can avoid heat generation during grinding and lower the product temperature. Specifically, when the amount of anionic dispersant added in one go is 80 g / L or more, the temperature rises by 10°C or more after 6 minutes of grinding. [Explanation of symbols]

[0254] 1 Decision System 2 Network 10 Server device 20 Mobile spreading equipment 21 Power Source 22 Drive mechanism 23 GPS unit 24 Multiple Tanks 25 Preparation equipment 26 nozzles 30 Terminal Equipment

Claims

1. The following components (a) and (b1): (a) an active ingredient having a herbicidal effect; (b1) 1 to 200 g / L of an anionic dispersant; the anionic dispersant is an alkylnaphthalene sulfonic acid derivative, A liquid pesticide composition designed so that the application amount of the anionic dispersant is 9 to 300 g / ha when applied to a field, the active ingredient having a herbicidal effect is at least one selected from the group consisting of tefuryltrione and triafamone, and optionally containing fentrazamide; It does not contain lignosulfonic acid derivatives as an anionic dispersant, containing a polyoxyethylene-polyoxypropylene block polymer as a nonionic surfactant, or not containing a nonionic surfactant; Liquid pesticide composition.

2. 2. The pesticide composition according to claim 1, wherein the liquid pesticide composition is an aqueous suspension or an oily suspension.

3. 3. The pesticide composition according to claim 1, wherein the active ingredient having a herbicidal effect is at least one selected from the group consisting of triafamone and optionally contained tefuryltrione and fentrazamide.

4. Optionally containing an active ingredient having an insecticidal effect, 3. The pesticide composition according to claim 1, wherein the active ingredient having an insecticidal effect is at least one selected from the group consisting of imidacloprid, thiacloprid, dinotefuran, flupyradifurone, flupirimine, nitenpyram, clothianidin, sulfoxaflor, ethiprole, fipronil, spinosad, tetraniliprole, chlorantraniliprole, cyantraniliprole, pymetrozine, triflumezopyrim, benzpyrimoxane, and oxazosulfil.

5. Optionally containing an active ingredient having a bactericidal effect, 3. The pesticide composition according to claim 1, wherein the active ingredient having a fungicidal effect is at least one selected from the group consisting of isotianil, probenazole, tricyclazole, diclobenthiazox, penflufen, thifluzamide, impirfluxam, kasugamycin, validamycin, fthalide, metominostrobin, azoxystrobin, and pencycuron.

6. A method for controlling diseases, pests and / or weeds, comprising applying one or more pesticide compositions according to claim 1 or 2 to a field, The method for controlling pests, wherein the anionic dispersant is applied to the field at an application rate of 9 to 300 g / ha.

7. The control method according to claim 6, wherein the anionic dispersant is applied to the field at an application rate of 9 to 150 g / ha.

8. The control method according to claim 6, wherein the active ingredient is applied to the field at an application rate of 50 to 650 g / ha.

9. The field is a paddy field, The method for controlling pests according to claim 6, wherein the application is local application to the water surface of a paddy field.

10. the method is performed by at least one processor executing computer-readable instructions; determining one or more of said pesticide compositions and an amount of each of said one or more pesticide compositions; measuring an amount of the one or more pesticide compositions when the one or more pesticide compositions is one pesticide composition, and measuring and / or mixing an amount of the one or more pesticide compositions when the one or more pesticide compositions is a plurality of pesticide compositions; diluting the one or more pesticide compositions; applying the one or more pesticide compositions to the field; The control method according to claim 6, comprising:

11. The control method according to claim 6, wherein the pesticide composition is automatically selected or selected based on predicted occurrence of diseases, pests and / or weeds in the field and / or their occurrence status.

12. A kit for use in the control method according to claim 6, comprising a plurality of containers each containing a plurality of pesticide compositions according to claim 1 or 2, respectively.

13. 3. Use of the pesticide composition according to claim 1 or 2 for controlling diseases, pests and / or weeds.

14. 14. Use of the pesticide composition according to claim 13, wherein the diseases, pests and / or weeds occur in paddy fields.

15. 3. A method for producing the pesticide composition according to claim 1, comprising: a mixing step of mixing the active ingredient and the anionic dispersant; and a grinding step of grinding the pesticide composition after the mixing step. and a further mixing step of further adding and mixing the anionic dispersant after the grinding step.

16. A method for producing an agrochemical composition, comprising mixing a plurality of agrochemical compositions according to claim 1 or 2.

Images