Information processing system and information processing method
Patent Information
- Application Number
- MYPI2025002939
- Authority / Receiving Office
- MY · MY
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2022-11-17
- Filing Date
- 2023-11-17
- Publication Date
- 2024-05-17
AI Technical Summary
Conventional methods for diagnosing plant diseases with pathogenic bacteria only determine the presence or absence of infection, failing to assess the progression stage, which limits effective treatment measures.
An information processing system and method that utilize a learning model to analyze image and sensing data from plants using spectrum or RGB cameras, determining the disease progression stage and recommending appropriate treatments based on stored treatment information.
Enables precise determination of treatment strategies based on the progression stage of plant diseases, improving the effectiveness of disease management and prevention.
Smart Images

Figure PI2025002939_DRAWING_001 
Figure PI2025002939_DRAWING_002 
Figure PI2025002939_DRAWING_003
Abstract
Description
Information processing system and information processing method
[0001] The present invention relates to an information processing system and an information processing method for monitoring the progress of a disease in a plant.
[0002]
[0003] Conventionally, there have been techniques for diagnosing disease (infection with pathogenic fungi) in trees based on photographed images of the trees. For example, Patent Document 1 below discloses a method for acquiring an image of a tree (hevea brasiliensis) to be diagnosed, inputting the image of the tree to be diagnosed into a diagnostic model that has undergone machine learning to diagnose trees with diseased leaves as diseased trees from tree images, using the image of the tree, information indicating the range in which the leaves are captured in the image, and information indicating whether the range contains characteristics of disease (infection with white root rot) as training data, and diagnosing the tree with diseased leaves as diseased trees, and outputting the diagnosis results.
[0003] Japanese Patent Application Laid-Open No. 2021-145557
[0004] However, the technology of Patent Document 1 only diagnoses whether or not a tree is diseased, but is unable to diagnose the extent of infection with pathogenic fungi, and therefore is unable to take measures according to the degree of infection.
[0005] The object of the present invention is to determine treatment depending on the stage of disease development in plants.
[0006] An information processing system according to one embodiment of the present invention is an information processing system for monitoring plants, comprising a memory unit and a control unit. The information processing system may be realized by a single device or by multiple devices. The memory unit stores treatment information regarding treatments to be taken according to multiple disease progression stages of the plant, and a learning model that has been trained on characteristics of each sensing data obtained by remotely sensing the plant at the multiple disease progression stages. The control unit acquires image data based on first sensing information obtained by capturing images of the plant using a first sensing means, such as a spectral camera or an RGB camera, of the remote sensing device, and determines the disease progression stage of the plant in the image data based on the acquired image data and the learning model. The control unit also determines the treatment to be taken according to the determined disease progression stage based on the determined disease progression stage and the stored treatment information, and generates treatment proposal information that proposes the treatment.
[0007] An information processing method according to one embodiment of the present invention is an information processing method for monitoring plants, comprising: storing treatment information relating to treatments to be taken according to multiple stages of disease progression in the plant, and a learning model that has been trained on the characteristics of each sensing data obtained by remotely sensing the plant at multiple stages of disease progression; acquiring image data based on first sensing information obtained by capturing an image of the plant with a spectral camera or RGB camera, which is a first sensing means possessed by a remote sensing device; determining the stage of disease progression in the plant in the image data based on the acquired image data and the learning model; and determining the treatment to be taken according to the determined stage of disease progression based on the determined stage of disease progression and the stored treatment information, and generating treatment proposal information that proposes the treatment.
[0008] According to the present invention, it is possible to determine treatment depending on the stage of disease progression in the plant.
[0009] 1 is a diagram showing the configuration of a tree infection monitoring system according to one embodiment of the present invention. FIG. 2 is a diagram showing the hardware configuration of a monitoring server according to one embodiment of the present invention. FIG. 3 is a diagram showing the configuration of a database held by a monitoring server according to one embodiment of the present invention. FIG. 4 is a flowchart showing the flow of a learning model generation process by a monitoring server according to one embodiment of the present invention. FIG. 5 is a flowchart showing the flow of a tree treatment method determination process by a monitoring server according to one embodiment of the present invention. FIG. 6 is a table showing examples of tree treatment methods according to the infection progression stage determined in one embodiment of the present invention. FIG. 7 is a diagram showing an example of a progression stage map generated in one embodiment of the present invention. FIG. 8 is a diagram showing an example of treatment suggestion information displayed when one area of the progression stage map generated in one embodiment of the present invention is selected by a user. FIG. 9 is a diagram showing an example of a treatment treatment map generated in one embodiment of the present invention. FIG. 10 is a table showing examples of tree treatment methods according to the infection progression stage determined in another embodiment of the present invention. FIG. 11 is a diagram showing an example of a progression stage map generated in another embodiment of the present invention.
[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
[0011] [System Configuration] As shown in FIG. 1, this tree infection monitoring system includes a monitoring server 100 on the Internet 50 , a user terminal 200 , and a proposer terminal 300 .
[0012] This system monitors the infection status of Ganoderma fungi on plantation trees (palm trees), determines the stage of infection (infection stage) of the infection for infected palm trees, and proposes treatment (cure or prevention) methods for Ganoderma depending on the infection stage. Palm trees include oil palms, coconut palms, etc.
[0013] The monitoring server 100 is a server (information processing device) operated by the provider of the monitoring service. The monitoring server 100 is connected to the user terminal 200 and the proposer terminal 300 via the Internet 50.
[0014] The user terminal 200 is a terminal used by the owner or manager (user) of the farm, and may be, for example, a smartphone, a mobile phone, a tablet PC (Personal Computer), a notebook PC, or a desktop PC.
[0015] The owner or manager (user) of a farm requests a monitoring service provider to provide monitoring services for his or her farm, and the user terminal 200 receives the monitoring results from the monitoring server 100, i.e., information on the infection stage of Ganoderma in each palm tree on the farm and information on the corresponding treatment (treatment or prevention).
[0016] In response to a monitoring request from the user terminal 200, the monitoring server 100 periodically (for example, twice a year) captures images of the target area (all or part) of the user's farm from the air or the ground using a remote sensing device.
[0017] As shown in the figure, the remote sensing device is at least one of three types: a ground mobile device, a low-altitude flying vehicle, and a low-orbit microsatellite, and is equipped with sensing means. Preferably, a low-altitude flying vehicle is used.
[0018] A remote sensing device is a small device that has basic mobile device means such as communication functions, GPS (Global Positioning System), and acceleration sensors in addition to sensing means such as an optical RGB camera, spectral camera, radar, spectral sensor, and temperature sensor described below, and is a terminal that can photograph objects such as plants at specific wavelengths.
[0019] Low-altitude flying objects are unmanned aerial vehicles (mainly drones) including multicopters and fixed-wing aircraft, but also manned aircraft such as helicopters and gliders, and are equipped with the aforementioned sensing means / devices and basic means / devices for movement.
[0020] A low-orbit microsatellite is an artificial satellite weighing less than 100 kg that orbits at high speed in a low orbit of about 400-600 km above the Earth and is equipped with the above-mentioned sensing means / device and basic mobile means / device.
[0021] The ground mobile device is a device that can move on or near the ground, such as a mobile robot or an unmanned vehicle, and is equipped with the above-mentioned sensing means / device and mobile device basic means / device.
[0022] The remote sensing device includes an optical RGB camera or a spectral camera as a sensing means / device to obtain first sensing data, including reflected wavelengths, based on which an image or image information can be synthesized.
[0023] The remote sensing device preferably further includes a second sensing means or device, such as a radar-based sensor (e.g., LiDAR, SAR), a temperature sensor (e.g., infrared temperature sensor), or the like, and preferably acquires second sensing data using the second sensing means or device. The first sensing data acquisition means / device and the second sensing data acquisition means / device may be used in combination, which is preferable in terms of improving the accuracy of detecting early infection stages of disease. Here, each sensing means may be an integrated device or may be separate. In this embodiment, measurements are performed using one of the spectral camera and RGB camera (preferably a spectral camera) and at least one other type of sensor (e.g., a temperature sensor or radar-based sensor).
[0024] By using second sensing data such as a temperature sensor or radar, highly accurate information on temperature or plant morphology can be acquired and received, and by using this information in combination with the image information acquired and received by the first sensing, the accuracy of determining the infection stage can be improved. For example, by detecting changes in palm tree temperature due to Ganoderma infection, or slight changes in palm tree morphology due to infection, such as slight changes in leaf angle or bending, it is possible to determine the presence or absence of infection and the infection stage.
[0025] The monitoring server 100 acquires sensing data (spectral images and / or RGB images, and other sensing data) from a remote sensing device, and determines the infection stage of Ganoderma for each palm tree in the data based on the acquired image data and other sensing data, and a learning model that has learned the characteristics of each sensing data obtained by remotely sensing palm trees at multiple infection stages of Ganoderma.
[0026] The monitoring server 100 then determines the treatment appropriate to the infection stage determined above based on the treatment information regarding the treatment to be taken according to each infection stage received from the proposer terminal 300, notifies the infection status information, and generates treatment proposal information (infection stage map) that suggests the treatment to the user, and sends it to the user terminal 200.
[0027] The proposer terminal 300 is a terminal used by a business operator proposing the above-mentioned Ganoderma infection countermeasure method, and is, for example, a smartphone, mobile phone, tablet PC, notebook PC, desktop PC, etc. The proposer terminal 300 proposes and calculates information regarding the treatment to be taken depending on the infection stage of Ganoderma, typically information regarding treatment actions such as spraying, injecting into trees, irrigation, adding fertilizer, adding microorganisms, etc., information regarding additives used in these treatments such as pesticides, adjuvants, fertilizers, fungicides, microorganisms, etc., more specifically, information regarding the type of additive, amount added (per tree), addition concentration, etc., and treatment frequency, and transmits this information to the monitoring server 100.
[0028] [Hardware Configuration of Monitoring Server] As shown in FIG. 2, the monitoring server 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an input / output interface 15, and a bus 14 connecting these to each other.
[0029] The CPU 11 accesses the RAM 13 and other memory as needed, and performs various arithmetic processing while controlling all of the blocks in the monitoring server 100. The ROM 12 is a non-volatile memory that permanently stores firmware such as the OS, programs, and various parameters to be executed by the CPU 11. The RAM 13 is used as a working area for the CPU 11, and temporarily stores the OS, various applications currently being executed, and various data currently being processed.
[0030] The input / output interface 15 is connected to a display unit 16, an operation reception unit 17, a storage unit 18, a communication unit 19, and the like.
[0031] The display unit 16 is a display device that uses, for example, an LCD (Liquid Crystal Display), an OELD (Organic ElectroLuminescence Display), a CRT (Cathode Ray Tube), or the like.
[0032] The operation reception unit 17 is, for example, a pointing device such as a mouse, a keyboard, a touch panel, or other input device. When the operation reception unit 17 is a touch panel, the touch panel can be integrated with the display unit 16.
[0033] The storage unit 18 is a non-volatile memory such as a hard disk drive (HDD), a flash memory (SSD; solid state drive), or other solid-state memory. The storage unit 18 stores the OS, various applications, and various data.
[0034] As will be described later, particularly in this embodiment, the storage unit 18 has a farm information database, a monitoring information database, and a treatment information database in addition to programs such as applications that execute each step of the tree treatment method determination process described later. The programs can be realized as a program product.
[0035] The communication unit 19 is, for example, a NIC (Network Interface Card) for Ethernet or various modules for wireless communication such as a wireless LAN, and is responsible for communication processing between the user terminal 200 and the proposer terminal 300 .
[0036] Although not shown in the figure, the basic hardware configuration of the user terminal 200 and the proposer terminal 300 is also approximately the same as the hardware configuration of the monitoring server 100, and each has the basic configuration required to function as a computer, such as a CPU, memory unit, communication unit, etc.
[0037] 3, the monitoring server 100 has a farm information database 31, a monitoring information database 32, and a treatment information database 33 in the storage unit 18. Note that each of these databases may be stored in a storage device or server externally connected to the monitoring server 100, rather than in the storage unit 18.
[0038] The farm information database 31 stores information about the user's farm, i.e., basic information such as the farm name and location, information about growing or past palm trees such as the type of palm tree (seedling variety), tree age, generation, and height, and information about the palm tree growing environment such as the soil type and terrain type (slope) in which the palm trees grow, for each farm, by receiving this information from, for example, the user terminal 200.
[0039] If the farm information database 31 is missing some information about palm trees on a farm, the monitoring server 100 can calculate and supplement the missing information based on sensing information as necessary. For example, height can be calculated from image information, and tree age can be calculated from height and species. Land slope can also be calculated using a radar-based sensor. Examples of missing information include when the information was not entered by the user. This information can be included in the training data of a learning model trained on the characteristics of the image data and other sensing data, as well as remote sensing data of palm trees at multiple Ganoderma infection stages, and can be used to determine the Ganoderma infection stage of each palm tree based on the learning model.
[0040] The monitoring information database 32 stores information regarding the monitoring (sensing and its analysis) of each farm, i.e., the monitoring execution schedule (observation date), execution method (information regarding the remote sensing device, etc.), target area (location information), and monitoring results (image data and other sensing data captured, information regarding the infection stage of Ganoderma in each palm tree determined from the sensing data, and information regarding the infection stage of Ganoderma in palm trees neighboring each palm tree).
[0041] The treatment information database 33 stores information regarding the treatments (cure / prevention) decided / proposed based on the above monitoring results, i.e., information regarding the type and efficacy of the agents (pesticides, adjuvants, fertilizers) used in the treatment, information regarding the method of using the agents (amount and frequency of spraying on the surface of the tree trunk, injecting into the inside of the tree trunk, soil irrigation, etc.), history information on past treatments, information regarding the degree of improvement since the treatment at the first / previous monitoring, infection stage map data sent to the user terminal 200 according to the above infection stage, etc.
[0042] Of these, the information regarding treatment may be stored in the treatment information database 33 by the monitoring server 100 sending information regarding the infection stage of the palm tree determined from the monitoring results to the proposer terminal 200, and receiving treatment information corresponding to the infection stage from the proposer terminal 200.
[0043] These databases are mutually referenced as necessary in the tree treatment method decision process by the tree infection monitoring system described below.
[0044] In addition to these databases, the monitoring server 100 stores a learning model for determining the infection stage of palm trees from sensing data (described below).
[0045] [Operation of the tree infection monitoring system] Next, the operation of the tree infection monitoring system configured as described above will be described. The operation is performed by the cooperation of hardware such as the CPU 11 and communication unit 19 of the monitoring server 100 and software stored in the storage unit 18. For convenience, in the following description, the CPU 11 of the monitoring server 100 is the subject of operation.
[0046] First, a description will be given of the learning model generation process performed by the monitoring server 100. Fig. 4 is a flowchart showing the flow of the learning model generation process.
[0047] As shown in the figure, the CPU 11 of the monitoring server 100 acquires sensing data obtained by remote sensing palm trees in multiple different palm tree plantations from microsatellites, flying vehicles, and ground terminals, and then performs preprocessing on this data to generate a learning model (step 41).
[0048] Specifically, distortions in the acquired image data (spectral images or RGB images) are corrected, and wavelength shifts between images are corrected so that multiple spectral images overlap for the same wavelength.
[0049] Next, the CPU 11 sets explanatory variables and objective variables for machine learning based on the preprocessed data (step 42). Specifically, the preprocessed sensing data and data such as palm tree location, age, species, generation, and soil type received from the user terminal 200 are set as explanatory variables for machine learning, and the infection stage of the disease that the palm tree is suffering from (e.g., infection stage 0 to 4) is set as the objective variable for machine learning (classification problem).
[0050] Next, the CPU 11 generates training data and validation data from the preprocessed dataset (step 43). That is, various teacher data with different weights and biases according to the infection stage of the disease are generated, and validation data is generated to evaluate the performance of hyperparameters that set the behavior of the machine learning algorithm.
[0051] For example, as the infection stage of a diseased tree progresses, the area that can be determined from the spread of the leaves when viewed from above, the maximum diameter equivalent to a circle that can be drawn by connecting the tips of the leaves, or the leaf area becomes smaller, the temperature tends to change, the color of the leaves tends to change, etc., and the spectrum also differs for each infection stage.Therefore, the CPU 11 extracts data with different values for these items from the above sensing data and generates training data having parameters corresponding to each infection stage.
[0052] The validation data is also used to determine hyperparameters related to the structure and configuration of the model (for example, in the case of a neural network, numerical values such as the number of layers and the learning rate) and to select a model that is expected to have high performance.
[0053] After preprocessing, the rest of the dataset, other than the training data and validation data, is left as test data to test the versatility of the trained model.
[0054] Next, the CPU 11 constructs a learning model based on the explanatory variables and the objective variables, and trains the learning model based on the training data and the verification data (step 44).
[0055] The CPU 11 then evaluates the versatility of the trained model using the test data (step 45). By comparing the evaluation results of different trained models, the CPU 11 selects and generates an optimal machine learning model that describes the relationship between the explanatory variables and the target variable.
[0056] Here, examples of machine learning algorithms that may be used include logistic regression, random forest, boosting, and SVM (support vector machine), which are suitable for image processing. While the above examples show supervised learning, unsupervised learning may also be used. Furthermore, CNN (convolutional neural network), VAE (variational autoencoder), GAN (generative adversarial network), and deep learning algorithms may also be used.
[0057] Next, a tree treatment method determination process using the learning model generated above will be described. Fig. 5 is a flowchart showing the flow of the tree treatment method determination process performed by the monitoring server 100.
[0058] As shown in the figure, first, the CPU 11 of the monitoring server 100 determines whether a request to confirm the infection status of Ganoderma palm trees has been received from the user terminal 200 (step 51). The infection status confirmation request includes location information for the palm tree growing area in the user's farm for which the user wishes to confirm the infection status. The infection status confirmation request may also include at least one of growth information and environmental information for the palm tree for which the user wishes to confirm the infection status. Growth information includes the seedling variety, tree age, generation, tree height, etc., while environmental information includes the soil type and terrain type in which the plant is growing.
[0059] Next, the CPU 11 transmits instruction information to the remote sensing device based on the received location information to instruct the remote sensing device to perform sensing (step 52). The instruction information includes the location information as well as information such as the sensing date and time, the sensing method (type of sensor to be used), etc. If multiple remote sensing devices are used, the instruction information is transmitted to each of them.
[0060] Next, the CPU 11 determines whether or not sensing data has been received from the remote sensing device (step 53).
[0061] If it is determined that sensing data has been received (Yes in step 53), the CPU 11 uses the above-mentioned learning model to determine the stage of infection with Ganoderma in each palm based on the sensing data and information about the palm trees (or farms) whose infection status is being confirmed (step 54).
[0062] Next, the CPU 11 refers to the treatment information database 33 and determines the treatment (treatment method, prevention method) to be taken for each palm tree according to the determined infection stage (step 55).
[0063] Figure 6 is a table showing an example of a tree treatment method according to the progression stage in this embodiment. As shown in the figure, in this embodiment, the infection stages of Ganoderma are set and determined by the learning model as five stages: infection stage 0 (healthy), infection stage 1 (early infection), infection stage 2 (mid-stage infection), infection stage 3 (severe infection), and infection stage 4 (very severe infection). Note that the infection stages of plant diseases vary depending on the plant, and may be standardized by government agencies or academic research institutions of the country where the plant grows.
[0064] A typical example of treatment in this embodiment is the addition of a pesticide and an adjuvant (a pesticide efficacy enhancer) to enhance the effect of the pesticide. In this embodiment, the higher the infection stage, the more frequent the treatment, and the higher the concentration of the adjuvant used together with the pesticide.
[0065] Palm trees at infection stage 0 are healthy trees that are located around infected trees at infection stage 1 or higher. As shown in the figure, palm trees at infection stage 0 are sprayed with an aqueous solution of pesticide and adjuvant (pesticide alone is sufficient as the tree is healthy) onto the trunk surface as a preventative measure. Note that the aqueous solution refers to a dispersion liquid with water as the solvent, and may also be referred to as a treatment liquid. Hereinafter, adding a pesticide or adjuvant can be said to mean adding these treatment liquids.
[0066] The pesticide used is, for example, hexaconazole, at a concentration of 0.1%. The adjuvant used is Kao Adjuvant A-134, at a concentration of 0.1%. The total volume of both is 2 L, and treatment is to be carried out at least once a year.
[0067] Palm trees at infection stage 1 have white buttons at the base and symptoms on the leaves that are difficult to see with the naked eye. As shown in the figure, for palm trees at infection stage 1, treatment stage 1 is proposed, which involves spraying an aqueous solution of pesticide and adjuvant onto the surface of the trunk, injecting the solution into the trunk and irrigating the soil, and also adding microorganisms.
[0068] Hexaconazole is used as the pesticide and Kao Adjuvant A-134 as the adjuvant for all treatments: trunk surface spray, trunk injection, and soil drench. The pesticide concentration is 0.1-0.3% for all treatments, and the adjuvant concentration is 1.0% for trunk surface spray and trunk injection, and 0.1-1.0% for soil drench. The combined volume of both is 2 L for trunk surface spray and trunk injection, and 20 L for soil drench at the base of the tree. All treatments are performed at least twice a year.
[0069] Microorganisms are added using fertilizers containing microorganisms (microorganism species: Trichoderma, Hendersonia, Actinomycetes, etc.). For example, when using a fertilizer with a trade name of Bio-fertilizer (Gano EF), the concentration and frequency of application are 50 g / tree three times / year for young trees, 500 g / hole once / hole when replanting trees, 2 kg / tree once / year for trees four years or younger, and 4 kg / tree once / year for trees four years or older.
[0070] Palm trees at infection stage 2 are judged comprehensively based on the presence of white buttons and mushrooms at the base of the tree, rotting of less than 30%, and yellowing of the leaves. As shown in the figure, for palm trees at infection stage 2, treatment stage 2 is proposed, which involves spraying a solution of pesticide and adjuvant onto the surface of the trunk, injecting the solution into the trunk and irrigating the soil, and also adding microorganisms.
[0071] In the treatment phase 2, whether spraying onto the trunk surface, injecting into the trunk, or drenching the soil, the pesticides and adjuvants used will be the same as those for the palm trees in the infection phase 1, but the dosage will be increased and the frequency of treatments will be increased to a minimum of four times per year for all treatments. Furthermore, the same agents, concentrations, and frequency of microorganism additions will be used as for the palm trees in the treatment phase 1.
[0072] Palm trees at infection stage 3 have white buttons and mushrooms at the base of the tree, decay of less than 50%, and can be identified comprehensively by the phenomenon of leaf breaking. As shown in the figure, for palm trees at infection stage 2, treatment stage 3 is proposed, which involves spraying a solution of pesticide and adjuvant onto the surface of the trunk, injecting the solution into the trunk and irrigating the soil, and also adding microorganisms.
[0073] In treatment stage 3, whether sprayed onto the trunk surface, injected into the trunk, or drenched in the soil, the pesticides and adjuvants used will be the same as those used for palm trees in infection stages 1 and 2, but increased dosages are proposed and the frequency of treatments will be increased to a minimum of six times per year for all treatments. Microorganisms will also be added at the same agent, concentration, and frequency as those used for palm trees in treatment stages 1 and 2.
[0074] Palm trees in infection stage 4 are fallen or dead palm trees, and as shown in the figure, treatment stage 4 involves felling, replanting the palm trees, replacing the soil, and adding microorganisms.
[0075] The microorganisms used are the same as those in the above-mentioned therapeutic treatment steps 1 to 3, and the amount and frequency of treatment are 500 g / whole, once / whole.
[0076] Examples of pesticide raw materials used in pesticides that can be used in this embodiment include, but are not limited to, those listed in the 2011 edition of the Pesticide Handbook (Japan Plant Protection Association). Furthermore, the pesticide efficacy enhancer composition of the present invention can be used safely without causing phytotoxicity to various crops. As described above, this embodiment uses the fungicide hexaconazole ((RS)-2-(2,4-dichlorophenyl)-1-(1H-1,2,4-triazol-1-yl)hexan-2-ol), but is not limited thereto. The following fungicides can also be used.
[0077] Other fungicides include sulfur-based fungicides such as zineb (zinc ethylenebisdithiocarbamate), maneb (manganese ethylenebisdithiocarbamate), manzeb (zinc ion-coordinated manganese ethylenebisdithiocarbamate), and polycarbamate (bisdimethyldithiocarbamoylzinc ethylenebisdithiocarbamate), and benzimidazole-based fungicides such as benomyl (methyl-1-(butylcarbamoyl)-2-benzimidazole carbamate). , thiophanate methyl (1,2-bis(3-methoxycarbonyl-2-thioureido)benzene), dicarboximide derivatives vinclozolin (3-(3,5-dichlorophenyl)-5-methyl-5-vinyl-1,3-oxazolidine-2,4-dione), iprodione (3-(3,5-dichlorophenyl)-N-isopropyl-2,4-dioxoimidazolidine-1-carboxamide), procymidone (N-(3,5-di chlorophenyl)-1,2-dimethylcyclopropane-1,2-dicarboximide), as well as triazine (2,4-dichloro-6-(2-chloroanilino)-1,3,5-triazine), triflumizole ((E)-4-chloro-α,α,α-trifluoro-N-(1-imidazol-1-yl-2-propoxyethylidane)-o-toluidine), iminoctadine acetate (1,1-iminodi(octamethylene)diguanidinium) triacetate), organic copper (Oxine-copper), cupric hydroxide (Coside Bordeaux, etc.), antibiotic disinfectants (streptomycin, tetracycline, polyoxy, blasticidin S, kasugamycin, validamycin), triadimefon (1-(4-chlorophenoxy)-3,3-dimethyl-1-(1,2,4-triazol-1-yl)-2-butanone), isoprothiolane (diisopropyl 1,3-dithiolan-2-ylidenemalonate), TPN (tetrachloroisophthalonitrile), etc., and preferred examples include organocopper (Oxine-copper), cupric hydroxide, triflumizole ((E)-4-chloro-α,α,α-trifluoro-N-(1-imidazol-1-yl-2-propoxyethylidane)-o-toluidine), iprodione (3-(3,5-dichlorophenyl)-N-isopropyl-2,4-dioxoimidazolidine-1-carboxamide), hexaconazole ((RS)-2-(2,4-dichlorophenyl)-1-(1H-1,2,4-triazol-1-yl)hexan-2-ol), and two or more of these may be used in combination.
[0078] In the case of insecticides, pyrethroid insecticides include permethrin ((3-phenoxybenzyl = (1RS,3RS)-(1RS,3RS)-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropanecarboxylate), cyvermethrin ((RS)-α-cyano-3-phenoxybenzyl = (1RS,3RS)-(1RS,3RS)-3-(2,2-dichlorovinyl)-2,2-dimethylcyclopropanecarboxylate), and fenvalerate (α-cyano-3-phenoxybenzyl-2-(4-chlorophenyl)-3-methylbutanoate), and organophosphate insecticides include DDVP (dimethyl 2,2-dichlorovinyl phosphate) and Sumithion (MEP). (O,O-dimethyl-O-(3-methyl-4-nitrophenyl)thiophosphate), Malathion (S-[1,2-bis(ethoxycarbonyl)ethyl]dimethyl phosphorothiolthionate), Dimethoate (dimethyl S-(N-methylcarbamoylmethyl)dithiophosphate), Elsan (S-[α-(ethoxycarbonyl)benzyl]dimethyl phosphorothiolthionate), Bayjit (O,O-dimethyl-O-(3-methyl-4-methylthiophenylthiophosphate)), Carbamate insecticides include Bassa (O-sec-butylphenyl methylcarbamate), MTMC (m-tolyl methylcarbamate), Meopal (3,4-dimethylphenyl-N-methylcarbamate), Methomyl (S-methyl-N Suitable examples include permethrin, DDVP (dimethyl 2,2-dichlorovinyl phosphate), and methomyl (S-methyl-N-[(methylcarbamoyl)oxy]thioacetimide).
[0079] Furthermore, examples of natural insecticides include pyrethrins derived from pyrethrum, piperonyl butoxide, rotenone derived from the legume shrub derris, nicotine (3-(1-methyl-2-pyrrolidinyl)pyridine sulfate), etc. Examples of insect growth regulators (IGRs) include diflubenzuron (1-(4-chlorophenyl)-3-(2,6-difluorobenzoyl)urea), teflubenzuron (1-(3,5-dichloro-2,4-difluorophenyl)-3-(2,6-difluorobenzoyl)urea), chlorfluazuron (1-[3,5-dichloro-4-(3-chloro-5-trifluoromethyl-2-pyridyloxy)phenyl]-3-(2,6-difluorobenzoyl)urea), etc.
[0080] Acaricides include CPCBS (parachlorophenyl parachlorobenzenesulfonate), phenisobromorate (isopropyl 4,4'-dibromobenzilate), tetradifon (2,4,5,4'-tetrachlorodiphenyl sulfone), fenothiocarb (S-4-phenoxybutyl dimethylthiocarbamate), fenpyroximate (tert-butyl (E)-α-(1,3-dimethyl-5-phenoxypyrazol-4-ylmethyleneaminooxy)-p-toluate), fluacinam (3-chloro-N-(3-chloro-5-trifluoromethyl-2-pyridyl)-α,α,α-trifluoro-2,6-dinitro-p-toluidine), fenbutatin oxide (hexakis(β,β-dimethylphenylethyl)distannoxane), and hexythiazox (trans-5-(4-chlorophenyl)-N -cyclohexyl-4-methyl-2-oxothiazolidine-3-carboxamide), amithrastin (3-methyl-1,5-bis(2,4-xylyl)-1,3,5-triazapenta-1,4-diene), etc., and preferred examples include phenisobromorate (isopropyl 4,4'-dibromobenzilate), amithrastin (3-methyl-1,5-bis(2,4-xylyl)-1,3,5-triazapenta-1,4-diene), and fenpyroximate (tert-butyl (E)-α-(1,3-dimethyl-5-phenoxypyrazol-4-ylmethyleneaminooxy)-p-toluate).
[0081] Examples of herbicides include acid amide herbicides such as Stam (3,4-dichloropropionanilide, DCPA). Examples of urea herbicides include DCMU (3-(3,4-dichlorophenyl)-1,1-dimethylurea). Examples of bipyridylium herbicides include paraquat (1,1'-dimethyl-4,4'-bipyridium dichloride) and diquat (6,7-dihydrodipyrido[1,2-a:2',1'c]pyrazinedium dibromide). Examples of diazine herbicides include bromacil (5-bromo-3-sec-butyl-6-methyluracil). Examples of triazine herbicides include simazine (2-chloro-4,6-bis(ethylamino)-1,3,5-triazine). Examples of nitrile herbicides include DBN (2,6-dichlorobenzonitrile). Examples of dinitroaniline herbicides include trifluralin (α,α,α-trifluoro-2,6-dinitro-N,N-dipropyl-p-toluidine). Examples of carbamate herbicides include benthiocarb (Saturn) (S-p-chlorobenzyl-N,N-diethylthiocarbamate). Examples of diphenyl ether herbicides include NIP (2,4-dichlorophenyl-p-nitrophenyl ether). Examples of phenolic herbicides include PCP (sodium pentachlorophenoxide). Examples of benzoic acid herbicides include MDBA (dimethylamine-3,6-dichloro-o-anisate). Examples of phenoxy herbicides include 2,4-D sodium salt (sodium 2,4-dichlorophenoxyacetate). Examples of amino acid herbicides include glyphosate (N-(phosphonomethyl)glycine or a salt thereof) and glyphosinate (ammonium-DL-homoalanin-4-yl(methyl)phosphinate).Examples of aliphatic herbicides include TCA sodium salt (sodium trichloroacetate), and preferred herbicides include DBN (2,6-dichlorobenzonitrile), DCMU (3-(3,4-dichlorophenyl)-1,1-dimethylurea), paraquat (1,1'-dimethyl-4,4'-bipyridium dichloride), and diquat (6,7-dihydrodipyrido[1,2-a:2',1'c]pyrazinedium dibromide).
[0082] Further examples of adjuvants include, but are not limited to, the wetting agents described in the 2011 Agricultural Chemicals Handbook (published by the Japan Plant Protection Association on February 25, 2011). In this embodiment, as described above, the adjuvant used is the product name "Kao Adjuvant A-134", but is not limited thereto, and the following may also be used.
[0083] Other commercially available spreading agents include Makupika (93.0% by mass of polyoxyethylene methylpolysiloxane, manufactured by Ishihara Biosciences Co., Ltd.), Squash (70% by mass of sorbitan fatty acid ester, 5.5% by mass of polyoxyethylene resin acid ester, manufactured by Kao Corporation), Avion-E (24% by mass of paraffin, manufactured by Avion Co., Ltd.), Petan V (42.0% by mass of paraffin, manufactured by Agrokanesho Co., Ltd.), Submerge (50% by mass of sodium alkylbenzenesulfonate, manufactured by Syngenta), and Braid. Examples of such surfactants include Xulu (80% by mass of polyoxyalkyleneoxypropylheptamethyltrisiloxane, 20% by mass of polyoxyalkylenepropenyl ether, manufactured by Sankei Chemical Co., Ltd.), Agridex (71.5% by mass of paraffin oil, manufactured by BASF), Agristic (40% by mass of alkylaryl polyglycol ether, manufactured by BASF), D-Flava (82% by mass of oleic acid, manufactured by Citaagungjaya), and Miracle 240 (75% by mass of trisiloxane, manufactured by Evonik).
[0084] The adjuvant contains a surfactant, and examples of such surfactants include nonionic surfactants, anionic surfactants, cationic surfactants, and amphoteric surfactants, or mixtures thereof.
[0085] Examples of nonionic surfactants include polyoxyethylene alkyl ethers, polyoxyethylene alkylaryl ethers, polyoxyethylene alkylaryl ether formaldehyde condensates, polyoxyalkylene aryl ethers, polyoxyalkylene alkyl sorbitol esters, polyoxyalkylene alkyl glycerol esters, polyoxyalkylene block copolymers, polyoxyalkylene block copolymer alkylglycerol esters, polyoxyalkylene alkylsulfonamides, polyoxypropylene block copolymers, polyoxyethylene oleyl ethers, polyoxyalkylene alkylphenols, polyoxyalkylene alkyl polyglycosides, and mixtures of two or more of these.
[0086] Cationic surfactants include alkylamine ethylene oxide adducts, alkylamine propylene oxide adducts, such as tallowamine ethylene oxide adducts, oleylamine ethylene oxide adducts, soyamine ethylene oxide adducts, cocoamine ethylene oxide adducts, synthetic alkylamine ethylene oxide adducts, octylamine ethylene oxide adducts, dialkylamine derivatives, and mixtures thereof. Examples of the dialkylamine derivatives include dialkylmonomethylhydroxyethylammonium propionate, dialkylmonomethylbenzalkonium chloride, and dialkylmonomethylethylammonium ethyl sulfate.
[0087] Typical anionic surfactants can be obtained in aqueous solution or solid state, and examples thereof include sodium mono- and di-alkylnaphthalenesulfonates, sodium alpha-olefinsulfonate, sodium alkanesulfonate, alkyl sulfosuccinates, alkyl sulfates, polyoxyalkylene alkyl ether sulfates, polyoxyalkylene alkylaryl ether sulfates, polyoxyalkylene styrylphenyl ether sulfates, mono- and di-alkylbenzenesulfonates, alkylnaphthalenesulfonates, formaldehyde condensates of alkylnaphthalenesulfonates, alkyldiphenylethersulfonates, and olefinic sulfonic acids. salts, mono- and di-alkyl phosphates, polyoxyalkylene mono- and di-alkyl phosphates, polyoxyalkylene mono- and diphenyl ether phosphates, polyoxyalkylene mono- and di-alkyl phenyl ether phosphates, polycarboxylates, fatty acid salts, linear and branched alkyl polyoxyalkylene ether acetates or salts thereof, alkenyl polyoxyalkylene ether acetates or salts thereof, linear and branched alkylamido polyoxyalkylene ether acetates or salts thereof, stearic acid and salts thereof, oleic acid and salts thereof, N-methyl fatty acid taurides, mixtures of two or more thereof, and the like (including sodium, potassium, ammonium, and amine salts).
[0088] Examples of amphoteric surfactants include alkyldimethylaminoacetic acid betaine, alkylcarboxymethylhydroxyethylimidazolinium betaine, alkylamidopropyl betaine, and alkyldimethylamine oxide, and one or more of these can be used as a mixture.
[0089] Returning to Figure 5, the CPU 11 then generates an infection stage map that visualizes the progression stage of Ganoderma infection in each palm tree on a map image as treatment suggestion information based on the infection stage determined above and the treatment decided upon (step 56), and transmits the generated progression stage map to the user terminal 200 (step 57).
[0090] After the CPU 11 generates treatment proposal information (infection stage map) and sends it to the user terminal 200, after a predetermined period of time (e.g., several months, six months, etc.) has passed, it may have the remote sensing device sense the palm again to obtain sensing data, and then generate treatment proposal information (infection stage map) again based on the sensing data and send it to the user terminal 200.
[0091] FIG. 7 is a diagram showing an example of an infection stage map generated in this embodiment and displayed on the application of the user terminal 200.
[0092] As shown in the figure, the infection stage map shows the area R of each palm tree on an RGB image (aerial photograph) of the target area of the palm plantation taken by the remote sensing device (mainly a drone) in a different color according to the infection stage of Ganoderma determined above.
[0093] The CPU 11 colors each palm region R on the RGB image in a color corresponding to the determined infection stage (for example, infection stage 0 is colorless, infection stage 1 is green, infection stage 2 is yellow, infection stage 3 is orange, and infection stage 4 is red), and refers to the treatment information database 33 to set a link for each palm region R that allows access to information indicating the determined treatment.
[0094] When the user of the user terminal 200 selects each region R, for example by tapping, information indicating the treatment or therapeutic procedure determined for the infection stage corresponding to the region is displayed, for example by a pop-up.
[0095] FIG. 8 is a diagram showing an example in which treatment information (or therapeutic treatment information) is displayed when one region R of the infection stage map is selected by the user.
[0096] As shown in the figure, a pop-up 80 showing treatment information is displayed starting from the region R selected by the user. The pop-up 80 displays the sequence number and coordinate information (latitude and longitude information) of the palm trees in that region R, as well as treatment information 81 showing the infection stage of Ganoderma on the palm trees and the corresponding treatment. In the example shown in the figure, for example, information suggesting spraying pesticides and adjuvants on the surface of the tree trunk is displayed. The user can also display the other treatments mentioned above (injection into the interior of the tree trunk, soil irrigation, and microbial addition) by switching screens on the pop-up.
[0097] Additionally, for example, at the bottom of the pop-up 80, a navigation button 82 is displayed for launching a map application and navigating the user terminal 200 to the location of each palm tree. The user can check the treatment for each palm tree on the infection stage map, prepare the agent required for treatment (treatment or prevention) such as pesticide or adjuvant, and begin treatment by starting navigation using the navigation button 82.
[0098] Furthermore, not only is a pop-up 80 displayed when a user selects each region R on the infection stage map, but when a user selects each infection stage on the infection stage map, the range of palm trees that should be treated according to the infection stage is highlighted on the map, and information indicating the area, etc. may be output. Each infection stage may be selected, for example, by selecting the box for each infection stage in the legend G.
[0099] Furthermore, when determining treatment for each palm tree, the CPU 11 may take into consideration not only the infection stage of Ganoderma on that palm tree, but also the infection stage of Ganoderma on palm trees adjacent to that palm tree, thereby enabling more accurate treatment proposals.
[0100] That is, the CPU 11 may set a link to information indicating a predetermined treatment similar to the treatment for palms determined to be in a diseased stage (e.g., infection stages 1 to 4) in the region R of palms determined to be in the non-disease stage (infection stage 0) on the infection stage map and adjacent to palms determined to be in the diseased stage (e.g., infection stages 1 to 4), but may not set such a link in the region of palms not adjacent to palms determined to be in the diseased stage. The predetermined treatment may be the same as the treatment for palms determined to be in infection stage 1.
[0101] Furthermore, the CPU 11 may change the treatment for a palm tree at infection stage 0 depending on whether the palm tree is adjacent to palm trees at infection stage 1 or higher in multiple directions or adjacent to the palm tree in only one direction. That is, even if the palm tree is healthy, if it is adjacent to an infected palm tree in multiple directions, it is considered that there is a high possibility of future infection, so the importance of prevention (concentration of pesticides or adjuvants) may be higher for the healthy palm tree than for the healthy palm tree adjacent to an infected palm tree in only one direction.
[0102] Furthermore, the CPU 11 may set links to information indicating different treatments for the region R of palm trees adjacent to palm trees determined to be at infection stage 2 (or infection stage 3) on the infection stage map, among palm trees determined to be at infection stage 1 (or infection stage 2), and for the region R of palm trees adjacent to palm trees determined to be at infection stage 3 (or infection stage 4). That is, even for palm trees determined to be at infection stage 1, if the region R is adjacent to a palm tree at infection stage 2, it is expected that the infection stage will worsen more in the latter case, and therefore the importance of treatment (concentration of pesticides or adjuvants) may be increased for the latter case.
[0103] The CPU 11 may also display a treatment map, such as that shown in Figure 9, upon request, for palm trees determined to be at infection stage 1 or 2, taking into account the areas of neighboring palm trees as described above. In the example shown in Figure 9, different types of lines surround different treatment stages, superimposed on the infection stage map. However, different colors may be used for different treatment stages, separate from the infection stage map. The infection stage map allows the disease state of the palm trees to be understood, and the treatment map can be used to support the plantation's treatment of infection. The treatment map can also display the order quantities of additives, such as pesticides and adjuvants, to be prepared by the plantation. Furthermore, maps for each treatment method, such as pesticide application areas and pesticide and adjuvant application areas, can also be displayed.
[0104] In determining the treatment stage according to the infection stage of surrounding plants, the adjacent range of the plant may be set appropriately according to the type of plant. The adjacent range may be set in units of the number of plants or in units of plots (areas). Setting information regarding the adjacent range may be input each time via the user terminal 200 or the operation reception unit 17 of the monitoring server 100, or information on the adjacent range predetermined according to the type of plant may be stored in the storage unit 18.
[0105] As shown in Figure 9, in this embodiment, the adjacent range is set in units of adjacent palm trees. However, the adjacent range is not limited to one tree. For example, in the case of a disease caused by spore dispersion, the disease may affect a section of multiple plants, so the adjacent range can be set in units of section (area), as in the example of soybeans described below.
[0106] [Summary] As described above, according to this embodiment, the tree infection monitoring system can determine a treatment according to the infection stage of the disease (Ganoderma) in the plant (palm tree) and suggest it to the user.
[0107] [Modifications] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made within the scope of the gist of the present invention.
[0108] In the above-described embodiment, the tree infection monitoring system monitors the infection status of the trees in the farm with the pathogenic fungus, determines the infection stage, and proposes a treatment (cure or prevention) method according to the infection stage. In addition, the tree infection monitoring system may perform a process after the treatment to confirm whether the treatment was effective.
[0109] Specifically, the monitoring server 100 performs the processing shown in Figure 5 above, and after the user takes the necessary measures (such as spraying pesticides on palm trees) according to the infection stage map, it performs the monitoring processing of steps 52 to 54 of Figure 5 again after a certain period of time (e.g., several months) has passed.
[0110] Once the infection stage of the palm tree is determined by the second monitoring process (step 54), the determination result is compared with the infection stage determination result from the first monitoring process stored in the monitoring information database 32 to confirm the control effect of the treatment carried out after the first monitoring process.
[0111] Specifically, if the result of the above comparison shows that the infection stage at the time of the second monitoring process for a particular palm tree is lower than the infection stage at the time of the first monitoring process, the CPU 11 determines that there was a control effect, and if the two infection stages are the same or the infection stage at the time of the second monitoring process is more advanced than the infection stage at the time of the first monitoring process, the CPU 11 determines that there was no control effect.
[0112] Information regarding the confirmed control effect is stored in the treatment information database 33 as information regarding the degree of improvement from the treatment at the initial or previous monitoring time, and is transmitted from the monitoring server 100 to the user terminal 200 in response to a request from the user or automatically.
[0113] In this way, the tree infection monitoring system can also diagnose the effectiveness of the treatments proposed and implemented by the user by performing the monitoring process multiple times to identify differences in the infection stage.
[0114] In the above-described embodiment, the monitoring server 100 stores treatment information corresponding to the infection stage in advance in the treatment information database 33, and when the infection stage is determined, the monitoring server 100 determines the treatment by referring to the stored information. Alternatively, the monitoring server 100 may not store treatment information in advance, but may, when the infection stage is determined, send request information requesting treatment information corresponding to the infection stage to the proposer terminal 300, and may determine the treatment based on the treatment information sent from the proposer terminal 300 in response, thereby generating the above-described progression stage map.
[0115] Although the above embodiment shows only one monitoring server 100, the processing executed by the monitoring server 100 may be distributed among a plurality of servers. For example, a server that stores remote sensing data and treatment information and a server that generates the learning model and determines the infection stage of Ganoderma based on the model may exist separately.
[0116] In the above-described embodiment, the monitoring server 100 may determine the next monitoring time based on the determination of the disease progression stage (infection stage) or based on time series information on the determination of the disease progression stage, and provide the user with suggested monitoring time information. For example, the monitoring frequency may be set to increase as the proportion of plants with serious disease progression stages at a certain time period increases, or as the disease progression rate increases over time.
[0117] The information on treatments according to the infection stage (the pesticides, adjuvants, and microorganisms to be used, and the amounts and frequency of use thereof) shown in Figure 6 above is merely an example and can be changed as appropriate depending on the plants to be treated, etc. Furthermore, the display modes of the infection stage maps shown in Figures 7 and 8 are also merely an example and can be changed as appropriate.
[0118] In the above-described embodiment, palm oil palm is used as an example of a plant and Ganoderma as a pathogen. However, the combination of plant and pathogen is not limited to this, and the method may be used to determine the presence of a disease in herbaceous plants. For example, the method may be used to determine the presence of Panama disease or new Panama disease in bananas, rust in soybeans, root rot in rubber, blast in rice, or rust in grapes. Furthermore, the method may be used to determine the presence of a disease in herbaceous plants, such as corn, potato, or rice. The method is suitable for diseases whose early stages are difficult to detect with the naked eye. Furthermore, in the present invention, plant disease is not limited to infections caused by fungi such as mushrooms or mold, bacteria, or viruses, but also includes diseases other than infectious diseases, such as those caused by pests or parasites. The "progression stage" refers not only to the infection stage of an infectious disease in a plant, but also to the progression stage of a disease other than an infectious disease in a plant. However, the present invention is suitable for treating plants infected with an infectious disease.
[0119] Hereinafter, an embodiment will be described in which the plant is soybean and the pathogen is soybean rust (hereinafter simply referred to as rust). There are two known species of rust: Asian species (Phakopsora pachyrhizi) and Latin American species (P. meibomiae), and the Asian species is particularly problematic.
[0120] FIG. 10 is a table showing examples of tree treatment methods according to the infection progression stage determined in this embodiment.
[0121] As shown in the figure, in this embodiment, the five infection stages of rust disease are set and determined by the learning model: infection stage 0 (healthy), infection stage 1 (early infection), infection stage 2 (mid-stage infection), infection stage 3 (severe infection), and infection stage 4 (very severe infection).
[0122] Preventive treatment of soybeans at infection stage 1 is also suggested by spraying with an aqueous solution of pesticide and adjuvant (pesticide alone is fine for healthy soybeans), or by adding microorganisms.
[0123] The pesticide used is, for example, hexaconazole, and its concentration is 0.01% to 0.001%. The adjuvant used is Kao Adjuvant A-134, and its concentration is 0.1% to 1%. The total amount of both added is 100L to 200L / 1000m. 2 The treatment frequency is set at least once a year.
[0124] To add microorganisms, fertilizer containing microorganisms (microorganism type: Bacillus, etc.) is used. For example, when using fertilizer under the trade name CEASE (BioWorks), the concentration and frequency of application are 0.05% to 0.3%, the amount of application is 100L to 200L per 1000m2, and the frequency of application is at least once per year.
[0125] To add silicon, a silicon-containing fertilizer is used. For example, when the fertilizer is AgroSilicio (manufactured by Harsco Minerais Ltda), the concentration and number of treatments are 100 to 800 kg / 1000 m. 2 The treatment frequency is set at least once a year.
[0126] Soybeans at infection stage 1 have small light brown to yellowish brown spots or raised patches on the lower leaves and on relatively mature leaves on the main stem from the lateral branches. As shown in the figure, for soybeans at infection stage 1, treatment stage 1 is proposed, which involves spraying an aqueous solution of pesticide and adjuvant onto the plant surface, as well as adding a microorganism.
[0127] In both the surface spray and soil drench treatments, hexaconazole is used as the pesticide and Kao Adjuvant A-134 is used as the adjuvant. The concentration is 0.1% to 1%. The total amount of both is 100L to 200L / 1000m 2 The treatment frequency is set at least once a year.
[0128] To add microorganisms, fertilizer containing microorganisms (microorganism type: Bacillus, etc.) is used. For example, when the fertilizer is a product name CEASE (BioWorks), the concentration and number of treatments are 0.05% to 0.3%, and the amount of addition is 100 L to 200 L / 1000 m. 2 The treatment frequency is set at least once a year.
[0129] In soybeans at infection stage 2, the leaves turn further yellow and turn brown to dark brown, or the plants emit uredospores (light brown powder). As shown in the figure, for soybeans at infection stage 2, spraying an aqueous solution of pesticide and adjuvant onto the plant surface is suggested as treatment stage 2.
[0130] In therapeutic treatment stage 2, the pesticides and adjuvants used are the same as those in infection stage 1 above, but the frequency of treatments is increased to a minimum of two times per year for each treatment.
[0131] Soybeans at infection stage 3 have polygonal, slightly swollen, dark brown spots (teliospore layers) visible under the epidermis around the urediospore layer. As shown in the figure, for soybeans at infection stage 3, spraying with an aqueous solution of pesticide and adjuvant is recommended as treatment stage 3.
[0132] In the treatment stage 3, the pesticides and adjuvants used are the same as those in the infection stage 1, but with the addition of 0.04-0.15% azoxystrobin as the pesticide. The frequency of treatments is increased to at least twice a year for each treatment.
[0133] Soybeans at infection stage 4 are soybeans whose damaged leaves have turned yellow and fallen off, and as shown in the figure, treatment stage 4 involves removing the infected soybean plants.
[0134] FIG. 11 is a diagram showing an example of a progress stage map generated in this embodiment.
[0135] The palm oil plant infection stage map shown in Figure 7 shows the regions R of each palm oil plant in different colors depending on the determined Ganoderma infection stage. As shown in the figure, the infection stage map of this embodiment also shows the regions Rs of each soybean plant in different colors depending on the determined rust infection stage. In addition, in this embodiment, the CPU 11 divides the entire area of the progression stage map into multiple sections Rf and displays the infection stage for each section Rf, for example, by a number. Instead of numbers, the infection stage may be indicated by a change in color or other display mode. The sections Rf correspond to a farm divided into predetermined units of area (e.g., 1 hectare).
[0136] That is, the infection stage of the soybean with the most advanced infection stage in each section Rf among the soybeans contained in that section Rf is shown as the infection stage of that section Rf with a corresponding number, for example, 0.5, 1, 2, or 3. Note that infection stage 0.5 means that the soybean is infected but has not yet reached infection stage 1.
[0137] This is because in the case of soybeans, the distance between soybeans is shorter than in the case of palm oil, and in the case of rust disease, germination, the lesions mature, and spores are dispersed, which can cause the infection to spread more widely from diseased soybeans to surrounding healthy soybeans. Therefore, it is considered appropriate to capture the disease stage not only in the soybeans adjacent to the diseased soybeans, but also in a wider area Rf.
[0138] In the progression stage map of this embodiment, a pop-up showing the treatment information shown in Figure 10 is also displayed, similar to the pop-up 80 in Figure 8, starting from the region Rs or section Rf selected by the user.
[0139] Moreover, instead of the infection stage map, the CPU 11 may display a treatment map similar to that shown in Fig. 9. Fig. 12 is a diagram showing an example of a treatment map generated in this embodiment.
[0140] The palm oil treatment map shown in Figure 9 is superimposed on the infection stage map, with different types of lines surrounding the regions R for each different treatment stage. On the other hand, as shown in the figure, in this embodiment, the infection stage of the soybean with the most advanced infection stage in each section Rf is set as the infection stage of that section Rf, and each section Rf is displayed in a different display mode depending on the treatment (treatment or prevention) corresponding to the infection stage of that Rf.
[0141] In the example shown in the same figure, the treatment map is superimposed on the infection stage map, with sections Rf surrounded by different types of lines for each different treatment stage, but separate from the infection stage map, sections Rf may also be shown in different colors for each different treatment stage.
[0142] Thus, in the case of soybeans, when determining the treatment stage according to the infection stage of surrounding soybeans, the adjacent range is set by the above-mentioned section Rf (area), the infection stage is determined based on the soybean with the highest infection stage in that section Rf, and the treatment stage can be determined based on the infection stage of that section Rf and the adjacent sections Rf.
[0143] As described above, in the infection stage map and treatment stage map, depending on the type of plant (characteristics of how the infection spreads), such as palm oil or soybean, the monitoring server 100 may determine and execute, according to pre-settings or input from the user terminal 200, whether to display the infection stage of the plant and its adjacent area on a single plant basis, as in the palm oil example, or on a plot basis, as in the soybean example (i.e., whether to treat on a single plant basis or on a plot basis).
[0144] Among the inventions described in the claims of this application, the invention described as an "information processing method" is one in which each step is automatically performed by at least one device such as a computer through information processing by software, and is not performed by a human using a device such as a computer. In other words, the "information processing method" is an information processing method using computer software, and is not a method in which a human operates a calculation tool called a computer.
[0145] The functions performed by the components described herein may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), a CPU (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to perform the described functions. A processor includes transistors and other circuits and is considered to be circuitry or processing circuitry. A processor may also be a programmed processor that executes programs stored in memory.
[0146] In this specification, a circuitry, unit, or means is hardware that is programmed to realize or performs the described functions, which may be any hardware disclosed herein or any hardware known to be programmed to realize or perform the described functions.
[0147] If the hardware is a processor that is considered to be a type of circuitry, the circuitry, means, or unit is a combination of hardware and software used to configure the hardware and / or processor. Also, configurations described as units or means in this specification can be implemented as circuitry.
[0148] 11...CPU 18...storage unit 19...communication unit 31...farm information database 32...monitoring information database 33...treatment information database 80...pop-up 81...treatment information 82...navigation button 100...monitoring server 200...user terminal 300...proposer terminal R...palm palm area Rs...soybean area Rf...map division section
Claims
1. An information processing system for monitoring plants, comprising: a memory unit that stores treatment information regarding treatments to be taken according to multiple stages of disease progression in the plant and a learning model that has been trained on the characteristics of each sensing data obtained by remotely sensing the plant at multiple stages of disease progression; and a control unit that acquires image data based on first sensing information obtained by imaging the plant with a spectral camera or RGB camera, which is a first sensing means possessed by a remote sensing device, determines the stage of disease progression in the plant in the image data based on the acquired image data and the learning model, and determines the treatment to be taken according to the determined stage of disease progression based on the determined stage of disease progression and the stored treatment information, and generates treatment proposal information that proposes the treatment.
2. The information processing system of claim 1, wherein the remote sensing device acquires second sensing information of the plant other than the first sensing information using a second sensing means, and the control unit determines the stage of progression of the disease of the plant in the image data based on the first sensing information, the second sensing information, and the learning model.
3. The information processing system according to claim 1 or 2, wherein the control unit transmits the generated treatment proposal information to a user terminal of a user who owns or manages the plant.
4. An information processing system as described in any one of claims 1 to 3, wherein the control unit receives information from the user terminal regarding at least one of the plant seedling variety, tree age, generation, tree height, soil type in which the plant grows, and terrain type, and determines the stage of progression of the plant disease based on the received information, the image data, remote sensing data, and the learning model.
5. The information processing system of claim 4, wherein, when information received from the user terminal is missing information regarding the plant seedling variety, age, generation, or tree height, the control unit calculates and complements the missing information based on the sensing information.
6. An information processing system as described in any one of claims 1 to 5, wherein the control unit transmits information relating to the stage of progress to a proposer terminal proposing the treatment, receives treatment information corresponding to the stage of progress transmitted from the proposer terminal, and stores it in the memory unit.
7. An information processing system as described in any of claims 3 to 4, wherein the control unit receives location information of an area of the user's farm where multiple plants grow from the user terminal, transmits the location information to the remote sensing device, and receives the image data of the area captured based on the location information from the remote sensing device.
8. An information processing system according to any one of claims 3 to 7, wherein the control unit generates, as the generated treatment proposal information, a progression stage map image in which the areas of each plant on the image are shown in different colors according to the different progression stages.
9. The information processing system according to claim 8, wherein the control unit sets a link in each area of the progression stage map image that allows access to information indicating the treatment.
10. The information processing system described in claim 9, wherein the control unit sets a link to information indicating a predetermined treatment similar to the treatment for the plant determined to be at a diseased stage in areas of plants determined to be at a non-disease stage on the progression stage map image that are adjacent to plants determined to be at a diseased stage, and does not set the link in areas of plants that are not adjacent to the plant determined to be at a diseased stage.
11. An information processing system as described in any one of claims 8 to 10, wherein the disease progression stages include a first disease stage, a second disease stage that is more advanced than the first disease stage, and a third disease stage that is more advanced than the second disease stage, and the control unit sets links to information indicating different treatments in a first adjacent area of a plant adjacent to a plant determined to be at the second disease stage and a second adjacent area of a plant adjacent to a plant determined to be at the third disease stage on the disease progression stage map image.
12. The information processing system of claim 8, wherein the control unit divides the entire area represented by the progression stage map image into a plurality of sections, and displays each section of the progression stage map image in a different display mode depending on the treatment corresponding to the progression stage of the plant that is most advanced among the plants contained in each section.
13. An information processing system as described in any one of claims 3 to 11, wherein the control unit acquires new image data of the plant captured by the remote sensing device after a predetermined period has elapsed since the generation of the treatment proposal information, generates the treatment proposal information based on the image data, and transmits it to the user terminal.
14. An information processing device for monitoring plants, comprising: a memory unit that stores treatment information regarding treatments to be taken according to multiple stages of disease progression in the plant and a learning model that has been trained on the characteristics of each sensing data obtained by remotely sensing the plant at multiple stages of disease progression; and a control unit that receives image data based on first sensing information obtained by capturing images of the plant with a spectral camera or RGB camera, which is a first sensing means possessed by the remote sensing device, determines the stage of disease progression in the plant in the image data based on the received image data and the learning model, and determines the treatment to be taken according to the determined stage of disease progression based on the determined stage of disease progression and the stored treatment information, and generates treatment proposal information that proposes the treatment.
15. An information processing device as described in claim 14, wherein the remote sensing device acquires second sensing information of the plant other than the first sensing information by a second sensing means, and the control unit determines the stage of progression of the disease of the plant in the image data based on the first sensing information and second sensing information received from the remote sensing device and the learning model.
16. The information processing device according to claim 14 or 15, wherein the control unit transmits the generated treatment proposal information to a user terminal of a user who owns or manages the plant.
17. An information processing device described in any one of claims 14 to 16, wherein the control unit receives information from the user terminal regarding at least one of the plant seedling variety, tree age, generation, tree height, soil type in which the plant grows, and terrain type, and determines the stage of progression of the plant disease based on the received information, the image data, remote sensing data, and the learning model.
18. An information processing device as described in claim 16 or 17, wherein the control unit generates, as the generated treatment proposal information, a progression stage map image in which the areas of each plant on the image are shown in different colors according to the different progression stages.
19. The information processing device according to claim 18, wherein the control unit sets a link in each area of the progression stage map image that allows access to information indicating the treatment.
20. The information processing device described in claim 19, wherein the control unit sets a link to information indicating a predetermined treatment similar to the treatment for the plant determined to be in the diseased stage in areas of plants determined to be in the non-disease stage on the progression stage map image that are adjacent to plants determined to be in the diseased stage, and does not set the link in areas of plants that are not adjacent to the plant determined to be in the diseased stage.
21. The information processing device according to claim 18, wherein the control unit divides the entire area represented by the progression stage map image into a plurality of sections, and displays each section of the progression stage map image in a different display mode depending on the treatment corresponding to the progression stage of the plant that is most advanced among the plants contained in each section.
22. An information processing method for monitoring plants, comprising: storing treatment information regarding treatments to be taken according to multiple stages of disease progression in the plant; and a learning model that has been trained on the characteristics of each sensing data obtained by remotely sensing the plant at multiple stages of disease progression; acquiring image data based on first sensing information obtained by capturing an image of the plant with a spectral camera or RGB camera, which is a first sensing means possessed by a remote sensing device; determining the stage of disease progression in the plant in the image based on the acquired image data and the learning model; determining the treatment to be taken according to the determined stage of disease progression based on the determined stage of disease progression and the stored treatment information; and generating treatment proposal information that proposes the treatment.
