Information processing apparatus, information processing method, and program

Abstract

To provide an information processing apparatus, an information processing method, and a program, which enable more accurate estimation of a state of a field while reducing facility cost.SOLUTION: An information processing apparatus of an embodiment includes an acquisition unit that acquires an image capturing a predetermined region of a field, an analysis unit that analyzes, from the image acquired by the acquisition unit, a crack state of soil contained in the predetermined region, and an estimation unit that estimates a state of the field including at least a midseason drainage period of the field, based on the crack state of the soil analyzed by the analysis unit.SELECTED DRAWING: Figure 3

Description

[Technical Field]

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

[0002] Traditionally, the impact of methane emissions from rice paddies on global warming has been attracting attention. It is known that methane emissions can be reduced by extending the period of mid-drying, in which water is drained from rice paddies to dry the surface (see, for example, Non-Patent Document 1). Conventionally, methods that use soil moisture sensors or water level sensors installed in rice paddies have been known as methods for monitoring the mid-drying period (see, for example, Patent Documents 1 to 3). In recent years, the J-Credit Scheme has become known, in which the government certifies greenhouse gas emission reductions as "credits" and makes them tradeable (see, for example, Non-Patent Document 2). Under this system, extending the mid-drying period of rice paddies beyond the conventional period is certified as a credit, and revenue can be earned by selling the credits. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-49102 [Patent Document 2] Japanese Patent Publication No. 2022-82636 [Patent Document 3] Patent No. 7152097 [Non-patent literature]

[0004] [Non-Patent Document 1] Shigeto Sudo, Masayuki Ito, Kazuyuki Yagi, "New Water Management Technology Manual for Suppressing Paddy Field Methane Emissions", National Institute for Agro-Environmental Sciences, August 2012 [Non-patent document 2] "About the J-Credit Scheme for 'Extending the Mid-Drying Period in Paddy Rice Cultivation'," Ministry of Agriculture, Forestry and Fisheries, Agricultural Products Bureau, Agricultural Environment Division, September 2023 Summary of the Invention [Problem to be solved by the invention]

[0005] However, conventional technology requires the installation of soil moisture sensors and water level sensors at multiple points in the field, which increases the cost of installing the sensors and places a heavy operational burden on the field, such as aggregating sensor results to determine the mid-drying period, making it difficult to accurately estimate the condition of the field.

[0006] The aspects of the present invention have been made in consideration of these circumstances, and aim to provide an information processing device, an information processing method, and a program that can estimate the state of a farm field more accurately while reducing equipment costs. [Means for solving the problem]

[0007] The information processing device, the information processing method, and the program according to the present invention employ the following configuration. A first aspect of the information processing device of the present invention is an information processing device comprising: an acquisition unit that acquires an image of a specified area within a field; an analysis unit that analyzes the crack state of the soil included in the specified area from the image acquired by the acquisition unit; and an estimation unit that estimates the field state including at least the mid-drying period of the field based on the crack state of the soil analyzed by the analysis unit.

[0008] In a second aspect of the information processing device of the present invention, the acquisition unit further acquires information regarding the soil moisture content or water level of the field, and the estimation unit estimates at least the mid-drying period of the field based on the crack state of the soil and at least one of the soil moisture content and the water level.

[0009] In the information processing device according to a third aspect of the present invention, the estimation unit further estimates at least one of a soil moisture content and a water level of the field based on a state of cracks in the soil.

[0010] The information processing device of the fourth aspect of the present invention further includes a learning unit that generates a trained model that inputs information about the crack state and outputs the inter-drying period based on information that corresponds the crack state and the inter-drying period for each past field, and the estimation unit obtains the inter-drying period by inputting at least information about the crack state of the soil into the trained model trained by the learning unit.

[0011] The information processing device of the fifth aspect of the present invention further includes a learning unit that generates a trained model that inputs the image acquired by the acquisition unit and outputs the interim drainage period based on information that associates past images containing soil in the field with the interim drainage period, and the estimation unit inputs the image acquired by the acquisition unit into the trained model trained by the learning unit, thereby acquiring the interim drainage period of the image contained in the image.

[0012] In the information processing device according to a sixth aspect of the present invention, the crack state further includes information on the amount of cracks contained in the image.

[0013] In the information processing device according to a seventh aspect of the present invention, the estimation unit further estimates the end time of mid-season drainage for the field based on a change in the state of the cracks over time.

[0014] In the information processing device according to an eighth aspect of the present invention, the analysis unit further analyzes the growth of crops in a field included in the image.

[0015] The information processing device according to a ninth aspect of the present invention further comprises an information providing unit that provides information including the mid-drying period estimated by the estimation unit.

[0016] In a tenth aspect of the information processing device of the present invention, the analysis unit further analyzes the degree of dryness of the specified area, and the information providing unit provides information regarding water supply to the field based on the degree of dryness determined by the analysis unit.

[0017] An eleventh aspect of the present invention is an information processing method in which a computer acquires an image of a specified area within a field, analyzes the crack condition of the soil contained in the specified area from the acquired image, and estimates the field condition including at least the mid-season drainage period of the field based on the analyzed crack condition of the soil.

[0018] A twelfth aspect of the present invention is a program that causes a computer to acquire an image of a specified area within a field, analyze the crack condition of the soil contained in the specified area from the acquired image, and estimate the field condition including at least the mid-drying period of the field based on the analyzed crack condition of the soil. [Effects of the Invention]

[0019] According to the aspects of the present invention, it is possible to estimate the state of a farm field more accurately while reducing equipment costs. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a configuration diagram of a farm land management system 1 to which an information processing device according to an embodiment is applied. [Figure 2] FIG. 2 is a diagram illustrating an example of a functional configuration of a terminal device 100. [Figure 3] FIG. 2 is a diagram illustrating an example of a functional configuration of an information processing device 200. [Figure 4] FIG. 10 is a diagram showing an example of the contents of farm field management information 292. [Figure 5] FIG. 10 is a diagram for explaining differences in crack states based on soil moisture content. [Figure 6] 10 is a process for explaining an image analysis process according to an embodiment. [Figure 7] 10A and 10B are diagrams for explaining analysis of the amount of cracks in the target area AR1. [Figure 8] FIG. 10 is a diagram illustrating a first generation method of a trained model 294 trained by a training unit 270. [Figure 9]FIG. 10 is a diagram illustrating a second generation method for a trained model 294. [Figure 10] 10 is a diagram showing an example of an image IM30 provided by an information providing unit 280. FIG. [Figure 11] 10 is a flowchart illustrating an example of processing executed by the information processing device 200 according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0021] Hereinafter, an information processing device, an information processing method, and a program according to an embodiment of the present invention will be described with reference to the drawings. Note that the following describes an example in which the information processing device is applied to a farm field management system that acquires information about a farm field and manages the state of the farm field, such as the mid-season drainage period.

[0022] [Overall configuration] 1 is a configuration diagram of a farm land management system 1 to which an information processing device according to an embodiment is applied. The farm land management system 1 includes, for example, one or more terminal devices 100 and an information processing device 200. The terminal devices 100 and the information processing device 200 are communicably connected, for example, via a network NW. The network NW includes, for example, a Wi-Fi network, a cellular network, the Internet, a WAN (Wide Area Network), a LAN (Local Area Network), a provider device, a wireless base station, etc.

[0023] The terminal device 100 is, for example, a smartphone, a tablet terminal, a camera device, or the like. The terminal device 100 includes at least an imaging unit (camera) that captures images of the surroundings, and a communication unit that transmits the captured images to the information processing device 200 via a network NW. In the example of FIG. 1, terminal devices 100-1 to 100-4 are shown, but the number and type are not limited to this. The terminal device 100 may be mounted on an air vehicle such as a drone or a small unmanned aerial vehicle (UAV) that is equipped with an imaging unit and a communication unit. The terminal device 100 captures, for example, an image of a farm field F that includes at least soil. The images captured by the terminal device 100 may be captured by a user who uses the terminal device 100, or may be captured while the terminal device 100 is installed in a fixed position.

[0024] The information processing device 200 is, for example, a general-purpose PC (Personal Computer) or a server device. The information processing device 200 may also be a cloud computing system realized by a server device or a storage device. The information processing device 200 may also be a communication terminal (terminal device) such as a smartphone or a tablet terminal. The information processing device 200 acquires image information, location information, etc. from the terminal device 100 via the network NW, manages the acquired images, analyzes the images to analyze the state of the field F, and estimates the mid-season drainage period of the field F based on the analysis results. The state of the field F is, for example, the state of cracks in the soil when the field F is a paddy field. The state of the field F may also include the growth (vegetation) state of crops and the like in the field F. The information processing device 200 also manages the analysis and estimation results, etc.

[0025] In addition to the above-described configuration, the farm land management system 1 may also include a measuring device 300 equipped with a sensor for measuring the soil moisture content (or soil moisture rate; the same applies below) in the farm field F and a sensor for measuring the water level in the farm field F. In this case, the measuring device 300 transmits the measurement results from the sensor to the information processing device 200 via the network NW from a communication unit provided within the device. The information processing device 200 then analyzes the condition of the farm field F and estimates the mid-season drainage period for the farm field F based on an image of the farm field F and the measurement results of the measuring device 300 (e.g., at least one of the soil moisture content and the water level). The measuring device 300 may also include a function for observing (measuring) surrounding meteorological data. The meteorological data includes, for example, weather, temperature, humidity, atmospheric pressure, wind direction, wind speed, precipitation, and sunshine duration. By acquiring meteorological data from the measuring device 300, the information processing device 200 can analyze the condition of the farm field F in more detail according to the surrounding environment.

[0026] The functional configurations of the terminal device 100 and the information processing device 200 will be specifically described below. [Terminal Device] FIG. 2 is a diagram illustrating an example of the functional configuration of the terminal device 100. Note that the configuration illustrated in FIG. 2 illustrates, as an example, the configuration when the terminal device 100 is a smartphone or a tablet terminal. The terminal device 100 includes, for example, a communication unit 110, an imaging unit 120, an output unit 130, a location information acquisition unit 140, a control unit 150, an application execution unit 160, and a storage unit 170. Some or all of the location information acquisition unit 140, the control unit 150, and the application execution unit 160 are realized by, for example, a hardware processor such as a CPU (Central Processing Unit) executing a program (software). Furthermore, some or all of these components may be realized by hardware (including circuitry) such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), a GPU (Graphics Processing Unit), or a SOC (System On Chip), or may be realized by a combination of software and hardware. The program may be stored in advance in a storage device such as an HDD or flash memory (a storage device with a non-transitory storage medium), or may be stored in a removable storage medium (a non-transitory storage medium) such as a DVD or CD-ROM, and installed in the storage device by inserting the storage medium into a drive device or a card slot of the terminal device 100.

[0027] The storage unit 170 may be realized by the various storage devices described above, or a solid-state drive (SSD), an electrically erasable programmable read-only memory (EEPROM), a read-only memory (ROM), a random-access memory (RAM), etc. The storage unit 170 stores, for example, a farm field condition management application 172, programs, and various other information.

[0028] The communication unit 110 communicates with the information processing device 200 and other external devices via, for example, a network NW. The communication unit 110 may also communicate with other terminal devices 100 via a short-range communication network or the network NW.

[0029] The imaging unit 120 is, for example, a digital camera using a solid-state imaging element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor). The imaging unit 120 may be a stereo camera. The imaging unit 120 captures an image of a predetermined area of ​​the field F. The predetermined area is an area that includes at least the soil in the field F. The captured image may be, for example, a visible image, a multispectral image, a hyperspectral image, or another digital image. The imaging unit 120 is fixedly installed in or near the field F and captures an image of a predetermined area based on an angle of view (e.g., magnification, etc.) from the installation position. Furthermore, the imaging unit 120 captures an image of a predetermined area based on an angle of view in the direction in which the user is pointing while held by the user. The imaging unit 120 periodically captures images repeatedly or captures images based on user operation. The captured image may be transmitted to the information processing device 200 via the communication unit 110 each time it is captured, or may be temporarily stored in the storage unit 170 and transmitted to the information processing device 200 at a predetermined timing.

[0030] The output unit 130 includes, for example, a display unit 132 and a speaker 134. The output unit 130 outputs predetermined information to the display unit 132 and the speaker 134. The display unit 132 is, for example, an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display. The display unit 132 displays various information in the embodiment. The display unit 132 may also be a touch panel having the function of an input unit that accepts user input operations. The speaker 134 outputs predetermined sound. For example, the output unit 130 outputs images and sound corresponding to information provided by the information processing device 200, and outputs images captured by the imaging unit 120, information to be input by the user, and information input by the user.

[0031] The location information acquisition unit 140 acquires location information of the terminal device 100. The location information acquisition unit 140 acquires location information (e.g., latitude and longitude) of the terminal device 100, for example, by a GPS (Global Positioning System) device built into the terminal device 100. Furthermore, the location information acquisition unit 140 may acquire information on the imaging direction and angle of view (e.g., magnification information) of the imaging unit 120 of the terminal device 100 in addition to the location information.

[0032] The control unit 150 controls all functions of the terminal device 100. For example, the control unit 150 controls communication by the communication unit 110, input / output by the output unit 130, and execution of the farm field condition management application 172 by the application execution unit 160.

[0033] The application execution unit 160 is realized by executing a field condition management application 172 stored in the storage unit 170. The field condition management application 172 is, for example, downloaded from an external device via the network NW and installed in the terminal device 100. The field condition management application 172 transmits information on images captured by the imaging unit 120, location information acquired by the location information acquisition unit 140, and the like to the information processing device 200 via the communication unit 110. The field condition management application 172 may also transmit identification information (terminal ID) of the terminal device 100, identification information (user ID) of the user who captured the image, and date and time information such as the date and time of shooting and the start date of mid-season drying, in association with the image information, to the information processing device 200. In this case, the field condition management application 172 may perform tampering prevention control using a timestamp or the like to prevent date and time information from being tampered with by an unauthorized input by a user or the like. For example, when associating date and time information with image information, the farm field condition management application 172 acquires the date and time information from an external source (for example, a certified time authentication business operator) via the network NW and assigns the date and time information to the image information. The farm field condition management application 172 also receives information transmitted from the information processing device 200, and performs imaging using the imaging unit 120 based on the received information, outputs the received information to the output unit 130, or stores the information in the storage unit 170.

[0034] If the terminal device 100 is a camera device, it may not have some of the components of the output unit 130 among the above-mentioned functions, and an app other than the farm field condition management app 172 may be installed.

[0035] [Information processing device] 3 is a diagram illustrating an example of the functional configuration of the information processing device 200. The information processing device 200 includes, for example, a communication unit 210, an acquisition unit 220, an output unit 230, a management unit 240, an analysis unit 250, an estimation unit 260, a learning unit 270, an information provision unit 280, and a storage unit 290. Some or all of the acquisition unit 220, the management unit 240, the analysis unit 250, the estimation unit 260, the learning unit 270, and the information provision unit 280 are realized by, for example, a hardware processor such as a CPU executing a program (software). Furthermore, some or all of these components may be realized by hardware (including circuitry) such as an LSI, ASIC, FPGA, GPU, or SOC, or may be realized by a combination of software and hardware. The program may be stored in advance in a storage device (non-transitory storage medium) such as a HDD or flash memory, or may be stored in a removable storage medium (non-transitory storage medium) such as a DVD or CD-ROM, and installed by inserting the storage medium into a drive device of the information processing device 200. The information processing device 200 may be realized by an operator of a farm field condition management service installing the program in a cloud server, in which case the owner of the hardware of the information processing device 200 may be different from the operator of the service.

[0036] The memory unit 290 may be realized by the various storage devices described above, or an SSD, an EEPROM, a ROM, a RAM, or the like. The memory unit 290 stores, for example, farm field management information 292, a trained model 294, programs, and various other information. At least a portion of the information included in the memory unit 290 may be stored in an external device (e.g., a database server) capable of communicating with the information processing device 200. The trained model 294 is a model (estimation) used by the estimation unit 260 to estimate at least the mid-season drainage period of the farm field F. The trained model 294 may also be, for example, a model (analysis model) used by the analysis unit 250 to analyze the state of cracks from an image. The trained model 294 may be trained by the learning unit 270 or acquired from an external device connected via the network NW.

[0037] 4 is a diagram showing an example of the contents of the field management information 292. For example, the field management information 292 associates a field ID, which is identification information for identifying the field F, and imaging location information with image information, soil moisture information, optional parameter information, mid-season drainage period information, and field environment prediction information. Note that the types of information stored in the field management information 292 are not limited to those listed above, and some information may not be stored.

[0038] The imaging position information is, for example, the position information of the terminal device 100 when the terminal device 100 captured an image including a predetermined area of ​​the field F. The image information is, for example, information related to the image captured by the terminal device 100 or the image analysis results by the analysis unit 250 (described later). The image information may also include the image capture date and time, and may include multiple images captured at different dates and times. The soil moisture information is, for example, information related to the soil moisture content or water level near the imaging position (within a predetermined distance from the imaging position) including a predetermined area of ​​the field F, or the soil moisture content or water level of the imaged field F. The soil moisture content and water level may be obtained, for example, from the measuring device 300 via the network NW, or from an external server that manages the soil moisture content and water level in the field F. The optional parameter information is, for example, meteorological data near the field F, soil information (for example, soil temperature, soil components, soil map, etc.), growth conditions (for example, tiller number, wilting, vegetation cover rate, etc.), a heat balance model of the crop community, etc. The optional parameter information may be obtained, for example, from an external server via the network NW, or may be obtained from a device provided on the farm field management system 1 for obtaining the optional parameters. The mid-season drainage period information is, for example, information about the mid-season drainage period of the farm field F estimated by the estimation unit 260. The mid-season drainage period information may include information about the start date of mid-season drainage obtained from the terminal device 100. The input farm field environment prediction information is, for example, future prediction information about the farm field environment estimated by the estimation unit 260. The prediction information includes, for example, predictions of the time to supply water to the farm field F and the growth status of crops in the farm field F.

[0039] Returning to FIG. 3, the communication unit 210 communicates with the terminal device 100 and other external devices via a network NW or the like.

[0040] The acquisition unit 220 acquires various information from the terminal device 100 or an external device connected via the network NW, based on the information received by the communication unit 210. For example, the acquisition unit 220 acquires image information, position information, and the like from the terminal device 100, and acquires measurement results from the measuring device 300.

[0041] The output unit 230 includes, for example, a display unit 232 and a speaker 234. The output unit 230 outputs predetermined information to the display unit 232 and the speaker 234. The display unit 232 is, for example, an LCD or an organic EL display. The display unit 232 displays various information in the embodiment. The display unit 232 may also be a touch panel having a function as an input unit that accepts input operations from a user (e.g., a system administrator) of the information processing device 200. The speaker 234 outputs predetermined sound. For example, the output unit 230 outputs images and sound corresponding to information provided by the information processing device 200, or outputs information to be input by the user of the information processing device 200 or information input by the user.

[0042] The management unit 240 manages the state of the field F. For example, the management unit 240 stores information acquired by the acquisition unit 220 in the field management information 292. The management unit 240 also causes the learning unit 270 to perform updates to the trained model 294, and performs processing to acquire the trained model 294 from an external device. The management unit 240 may also manage, for example, the status of a J-Credit application (procedure) based on the mid-drying period of the field F. The management unit 240 may also store and manage information required for a J-Credit application (for example, the number of days for mid-drying) together with images in the memory unit 290. This makes it possible to prevent tampering with dates (date and time information), etc., and to manage the state of the field F more accurately. Prevention of tampering with the photographing date and time, the start date of mid-season drying, etc. may be controlled by the field condition management app 172 of the terminal device 100, or the management unit 240 may analyze the captured image to determine whether or not it has been tampered with, and if it has been tampered with, output a warning or prevent the application of J-Credits, etc. For example, the management unit 240 may output a warning to the terminal device 100 if a timestamp is not added to the image information acquired from the terminal device 100. The management unit 240 may also analyze the captured image with photographing date and time information added to acquire the surrounding conditions of the field F, such as the weather, and determine that tampering has occurred if the acquired weather, etc. differs from the weather in the weather data corresponding to the photographing date and time, and issue a warning, etc. The method of tampering determination by the management unit 240 is not limited to the above example.

[0043] The analysis unit 250 analyzes the state of cracks in the soil of a predetermined area in the field F based on the images etc. acquired by the acquisition unit 220. The analysis unit 250 may also analyze the growth status of crops and the dryness degree of the soil included in the images. The functions of the analysis unit 250 will be described in detail later.

[0044] The estimation unit 260 estimates at least the mid-season drainage period of the field F based on the state of cracks in the soil analyzed by the analysis unit 250. The estimation unit 260 may also estimate at least the mid-season drainage period of the field F based on the state of cracks in the soil and the soil moisture content or water level. The function of the estimation unit 260 will be described in detail later. The analysis results by the analysis unit 250 and the estimation results by the estimation unit 260 may be stored in the field management information 292.

[0045] The learning unit 270 receives information or image information relating to the state of cracks in the soil as input, and generates a trained model 294 that outputs at least the mid-season drainage period of the field F. The learning unit 270 may also receive image information as input, and generate a trained model 294 that outputs at least the state of cracks in the soil contained in the image. The generated trained model 294 is stored in the storage unit 290. Details of the functions of the learning unit 270 will be described later.

[0046] The information providing unit 280 provides information on the processing results by the analysis unit 250 and the estimation unit 260 to the user of the terminal device 100. The information to be provided includes, for example, information on the state of cracks in the soil analyzed by the analysis unit 250 and information on the mid-drying period of the field F estimated by the estimation unit 260. The information providing unit 280 generates at least one of image information and audio information corresponding to the information to be provided, and transmits the generated information to the terminal device 100 via the communication unit 210. Furthermore, the information providing unit 280 may output the generated information to the information processing device 200 from the output unit 230, thereby providing the information to the user of the information processing device 200.

[0047] [Analysis Department] Next, a detailed description will be given of the functions of the analysis unit 250. For example, in the soil of a paddy field or the like, when the soil moisture content decreases due to drying, the soil shrinks and cracks occur. Furthermore, when water is added to cracked soil, the soil expands and the cracks become smaller (or disappear).

[0048] Fig. 5 is a diagram for explaining differences in crack conditions based on soil moisture content. Fig. 5(A) to Fig. 5(D) show examples of images captured by the terminal device 100 at the same point from above to below (the ground surface side) of a farm field F (e.g., a rice paddy). Fig. 5(A) shows an image IM10A taken after a first predetermined time has elapsed in a dry state of the farm field F (a state in which water has not been supplied to the rice paddy), Fig. 5(B) shows an image IM10B taken after a second predetermined time, which is longer than the first predetermined time, has elapsed while the field F remains dry, Fig. 5(C) shows an image IM10C taken when water has been supplied after the second predetermined time has elapsed, and Fig. 5(D) shows an image IM10D taken after a third predetermined time has elapsed after water has been supplied in Fig. 5(C).

[0049] As shown in Figure 5(A), when a paddy field is dried, cracks (CRK) appear on the surface of the ground. Further drying causes the soil to shrink, resulting in larger cracks (CRK) as shown in Figure 5(b) (the area of ​​the grooves of the cracks (CRK) increases when viewed from above). Furthermore, when water is supplied to the paddy field as shown in Figure 5(C), the moisture content of the soil increases, causing the soil to expand and the cracks to become smaller. After that, the cracks (CRK) become unrecognizable in the image as shown in Figure 5(D) (this also includes cases where they become unrecognizable in the image due to the water level on the ground). Note that Figures 5(C) and 5(D) show images with different water levels on the ground. As such, it can be seen that the state of the cracks changes depending on the drying period and the water supply conditions after drying, and that the drying period corresponds to the mid-drying period. Therefore, the analysis unit 250 analyzes the characteristics of the image of the soil (e.g., the ground surface) contained in the image and analyzes the state of the cracks in the soil from the analysis results.

[0050] 6 illustrates an image analysis process according to an embodiment. The analysis unit 250 sets a target area AR1 to be analyzed from an image IM20 captured by the terminal device 100. The target area AR1 may be designated in advance by a user of the terminal device 100, or the target area AR1 for analyzing the crack state may be extracted from color information of each pixel included in the image IM20. The target area AR1 may also be the entire image IM20.

[0051] For example, the analysis unit 250 performs edge extraction, color extraction, brightness extraction, shape extraction using pattern matching processing, etc. on the target area AR1 using existing image analysis processing, and extracts feature information of the image based on the extraction results.The analysis unit 250 then analyzes the state of cracks in the soil (ground) included in the image based on the feature information.

[0052] Specifically, the analysis unit 250 extracts edge points within the target area AR1 through image analysis processing, and extracts the shapes (contours, etc.) of objects (including cracks CRK, crops CRP, etc.) contained in the image IM20 by connecting the extracted edge point sequences. The analysis unit 250 may also acquire color and brightness information for each pixel and extract the object shapes by connecting edge points with similar acquired colors and brightness. The analysis unit 250 also identifies the soil area from the extracted shape information, color information, and brightness information, and further analyzes the crack state of the soil within the target area AR1 based on the positional relationship between the color of the soil in the soil area and the color of the cracks CRK. The crack state includes, for example, the amount of cracks contained in the target area AR1.

[0053] FIG. 7 is a diagram for explaining the analysis of the crack amount in the target area AR1. In the example of FIG. 7, the analysis unit 250 defines the groove (crack) portion of the crack CRK included in the image of the target area AR1 as a crack area AR2, and extracts the area of ​​the crack area AR2 as the crack amount. The analysis unit 250 may also extract the ratio (crack rate) of the area of ​​the crack area AR2 to the area of ​​the target area AR1 as the crack amount. The analysis unit 250 may also extract the crack amount according to, for example, the width of the groove of the crack CRK, or according to the length of the crack CRK or the number of cracks (number) that are equal to or longer than a predetermined length. The analysis unit 250 may also generate a crack extraction image based on the crack extraction results. The crack extraction image may be, for example, an image in which the crack area AR2 is displayed in a distinguishable color or pattern on the image of the target area AR1, as shown in FIG. 7, or an image in which the crack amount is associated with the image of the target area AR1.

[0054] Furthermore, when there are multiple images taken at different times at the same location, the analysis unit 250 performs the above-described analysis on each image. Furthermore, in the extraction of cracks, the analysis unit 250 may perform analysis using deep learning, such as semantic segmentation, instance segmentation, or panoptic segmentation. Furthermore, the analysis unit 250 may predict the future state of a crack based on changes in the state of the crack over a predetermined time. Furthermore, the analysis unit 250 may analyze the dryness level of the target area AR1 based on the state of the crack.

[0055] Furthermore, since the color and brightness of the soil vary depending on the water level as shown in FIGS. 5(C) and 5(D), the analysis unit 250 may analyze the water level based on color information and brightness information of the target area AR1. The analysis unit 250 may also analyze the amount of change in the cracks CRK over time and analyze the soil moisture content based on the analysis results. The analysis unit 250 may also analyze the growth status of the crop CRP based on the color information and shape of the crop CRP included in the image IM20. The growth status may include, for example, the number of tillers, degree of wilting, growth rate, and vegetation coverage of the crop CRP.

[0056] [Estimation and learning parts] Next, the functions of the estimation unit 260 and the learning unit 270 will be described in detail. The estimation unit 260 estimates at least the interim drainage period of the field F based on the analysis results by the analysis unit 250. For example, the estimation unit 260 performs image estimation using a trained model 294 that models the relationship between an image containing soil cracks and the interim drainage period. The trained model 294 may be trained by the learning unit 270, for example, or may be acquired from an external device connected via the network NW. Here, the generation of the trained model 294 by the learning unit 270 will be described using diagrams.

[0057] FIG. 8 is a diagram illustrating a first generation method of the trained model 294 trained by the learning unit 270. The learning unit 270 generates a mid-drying period estimation model that takes crack conditions as input and outputs the mid-drying period, based on, for example, information associating previously captured images of a specific area of ​​the field F with the mid-drying period (ground truth data) for that area. Specifically, the learning unit 270 inputs, as required data, a crack extraction image including the crack conditions analyzed by the analysis unit 250 for an image of a specific area of ​​the field F and the mid-drying period for the image, and performs learning using a predetermined learning method to generate a trained model 294 for mid-drying period estimation (estimation model) that takes the crack extraction image as input and outputs at least the mid-drying period of the field. Here, the predetermined learning method includes, for example, machine learning and deep learning that perform regression (predicting one data (numerical value) from another data (numerical value)) and classification (predicting the class to which the target data belongs). Furthermore, the learning unit 270 can generate a highly robust model by using images captured at multiple locations and shooting angles within the field F. The multiple locations include, for example, locations where rice is planted, locations where the rice has been partially harvested to expose the soil, and the edge of the field F. The shooting angles include, for example, directly below and obliquely (a state tilted at a predetermined angle from directly below). Furthermore, the learning unit 270 may input images captured under different conditions, such as different weather conditions and times, to learn from.

[0058] Furthermore, the learning unit 270 may include information on the soil moisture content and the soil water level in addition to the mid-drying period as required data. This allows the generation of a trained model 294 that takes a crack extraction image as input and outputs the mid-drying period and at least one of the soil moisture content and the water level. Therefore, by using the trained model 294, the estimation unit 260 can more accurately estimate the field condition and manage the field. The learning unit 270 updates the trained model 294 based on update information at a predetermined timing.

[0059] Furthermore, the learning unit 270 may input at least some of the optional parameter information (weather data in the vicinity of the field F, soil information (e.g., soil temperature, soil components, soil map, etc.), growth conditions (e.g., tiller number, wilting, vegetation coverage, etc.), and heat balance model of the crop community) stored in the field management information 292 during learning as optional data for learning. This makes it possible to generate a trained model that can more accurately estimate the mid-season drainage period, including more detailed information about the field, such as the field environment and crop condition.

[0060] The learning unit 270 may also directly input a captured image instead of a crack extraction image to estimate at least the interim drainage period. FIG. 9 is a diagram for explaining a second generation method of the trained model 294. The example of FIG. 9 differs from the example of FIG. 8 in that the information input for model generation is a captured image rather than a crack extraction image. In the case of the second generation method, the learning unit 270 may perform, as a predetermined learning method, deep learning using the image analysis results displayed by the analysis unit 250, or deep learning using semantic segmentation, instance segmentation, panoptic segmentation, etc., in addition to machine learning and deep learning that perform regression and classification. By generating the trained model 294 using the second generation method, the estimation unit 260 can efficiently estimate the interim drainage period (and at least one of the soil moisture content and the water level) from the image acquired by the acquisition unit 220.

[0061] As described above, according to the embodiment, it is possible to quantitatively estimate the mid-season drainage period based on image information without variation among users, thereby enabling more accurate estimation of the state of the farm field.

[0062] Furthermore, instead of (or in addition to) estimating the interim drainage period using the trained model 294 described above, the estimation unit 260 may estimate the interim drainage period and at least one of the soil moisture content and water level based on the degree of change in the crack state at the same location (same area) imaged at different times. For example, if the amount of cracking has increased after a predetermined time has elapsed, the estimation unit 260 outputs an estimation result in which the soil moisture content is reduced according to the increase, and conversely, if the amount of cracking has decreased, the soil moisture content is increased according to the decrease. Furthermore, since the color and brightness of the image of the soil portion vary depending on the water level on the ground, as shown in Figures 5(C) and 5(D), for example, the estimation unit 260 may analyze the water level of the soil based on the color information and brightness information of the soil analyzed by the analysis unit 250.

[0063] Furthermore, the estimation unit 260 may estimate the number of days remaining until the mid-season drainage period of the field F reaches the number of days for which the J-Credit Scheme is applied (the end time of the mid-season drainage), based on the degree of change in the crack state over time. Furthermore, the estimation unit 260 may estimate information regarding the water supply to the field F (whether or not water supply is necessary or the timing of water supply) based on the degree of dryness of the target area AR1 analyzed by the analysis unit 250.

[0064] Furthermore, the estimation unit 260 may estimate the growth status of the crop CRP from the crack state (crack extraction image) or the captured image instead of (or in addition to) the mid-season drainage period. The estimation unit 260 estimates the growth status and the timing of watering based on the degree of change in the amount of cracks over time and the degree of wilting of the crop CRP analyzed by the analysis unit 250, or determines whether watering the soil is necessary. When estimating the growth status of the crop CRP, the learning unit 270 may generate a trained model (estimation model) that receives the crack state (crack extraction image) and the captured image and outputs information about the growth status, and estimate the growth status using the generated model. This makes it possible to estimate the status of the field F in more detail and manage the field F.

[0065] [About providing information] The information providing unit 280 generates information relating to the analysis results by the analysis unit 250 and the estimation results by the estimation unit 260, and transmits the generated information to the terminal device 100 that transmitted the captured image via the communication unit 210. For example, the information providing unit 280 generates an image showing a crack extraction image extracted by the analysis unit 250 for the captured image or an image showing the estimation result, and provides the generated image to the terminal device 100.

[0066] Fig. 10 is a diagram showing an example of an image IM30 provided by the information providing unit 280. Note that the items and layout displayed in image IM30 are not limited to those shown here. The image IM30 shown in Fig. 10 includes, for example, a basic information display area AR31, an analysis image display area AR32, and an estimation result display area AR33. The basic information display area AR31 includes, for example, identification information that identifies the field F.

[0067] The analysis image display area AR32 includes information such as the crack state and crack extraction image analyzed by the analysis unit 250 for the captured image sent by the terminal device 100. Here, the information providing unit 280 can provide the results of the analysis of cracks by the analysis unit 250 to the user of the terminal device 100 more accurately by displaying the crack area of ​​the image to be displayed in a color that makes it distinguishable from other areas.

[0068] The estimation result display area AR33 includes, for example, the date of photography (or date and time), the start date of mid-drying, and information about the estimation result. The information about the estimation result includes, for example, the mid-drying period relative to the start date, the soil moisture content, the wilting status of the crops, etc. The start date of mid-drying may be input by the user or may be estimated from the captured image.

[0069] The image IM30 may also include an icon IC1. The icon IC1 is a GUI (Graphical User Interface) switch that accepts an instruction to end the display of the image IM30. When the icon IC1 is selected by the user of the terminal device 100, the display of the currently displayed image IM30 is ended. The information providing unit 280 may cause the generated terminal device 100 to output audio information corresponding to the image IM30.

[0070] Furthermore, the information providing unit 280 may provide information regarding the time and timing for adding water to the farm field F estimated by the estimation unit 260.

[0071] The information providing unit 280 may also provide information such as the number of days remaining until the mid-drying period reaches the number of days required for the J-Credit Scheme to be applied (the end time of the mid-drying period) and the determination result of whether the system can be applied. When it becomes possible to apply for J-Credit, the information providing unit may notify the user by providing information indicating this. The information providing unit 280 may also provide information regarding documents required for applying for J-Credit to the terminal device 100. In this case, the management unit 240 manages information required for applying for the J-Credit Scheme, stores and manages information for applying for the J-Credit Scheme entered by the terminal device 100 in the storage unit 290, and stores and manages credit information acquired through the application in association with the field ID in the storage unit 290. This allows the information processing device 200 to centrally manage everything from the management of the mid-drying period of field F to the management of J-Credit, thereby reducing the user's workload and enabling more appropriate field management.

[0072] [Processing flow] Next, an example of processing executed by the information processing device 200 in the embodiment will be described. Fig. 11 is a flowchart showing an example of processing executed by the information processing device 200 in the embodiment. In the example of Fig. 11, of the various processes executed by the information processing device 200, the description will mainly focus on processing for estimating the mid-drying period from crack information in an image and providing information on the estimation result.

[0073] In the example of FIG. 11, the acquisition unit 220 acquires an image captured by the terminal device 100 (step S100). Next, the management unit 240 stores and manages the acquired image information, etc. in the storage unit 290 (step S110). Next, the analysis unit 250 determines whether to analyze the crack state included in the captured image (step S120). In the process of step S120, for example, the determination of whether to analyze the crack state may be made based on an instruction from the user of the terminal device 100, or may be made based on information preset on the information processing device 200 side. If it is determined that the crack state is to be analyzed, the analysis unit 250 analyzes the crack state included in the captured image (step S130). Furthermore, after the process of step S130 or in the process of step S120, if it is determined that the crack state is not to be analyzed, the analysis unit 250 estimates the mid-drying period using a trained model corresponding to the input information (step S140). Next, the information providing unit 280 generates information on the analysis result by the analyzing unit 250 and the estimation result by the estimating unit 260, and provides it to the terminal device 100, etc. (Step S150). This ends the processing of this flowchart.

[0074] Next, an application example of the farm land management system 1 according to the embodiment will be described. [First application example] By applying the farmland management system 1 of the embodiment, for example, multi-point observation by the terminal device 100 becomes possible. For example, it becomes possible to take photographs at multiple locations using the terminal device 100 carried by the user, and by linking with the location information of the terminal device 100, it becomes possible to grasp the overall situation of the farmland F. Furthermore, according to the first application example, it is possible to estimate the state of crop growth (vegetation) as well as the soil, and to manage the wilting, number of stems, leaf color, etc. of the crops, and to link with other systems (for example, a cultivation management support system) via the network NW. Furthermore, by storing and managing the number of days for mid-season drying for J-Credit together with the images as farmland management information 292, it is possible to prevent tampering with dates, etc.

[0075] [Second application example] By applying the farmland management system 1 of the embodiment, remote observation becomes possible, for example, using a fixed camera device. In the second application example, the fixed camera device can continuously determine the implementation status of mid-season drainage, making it possible to observe the growth status throughout the growing period, including the mid-season drainage status. Furthermore, according to the second application example, similar to the first application example, it is possible to link with other systems, and by storing and managing the number of mid-season drainage days for J-Credit together with images, it is possible to prevent tampering with dates, etc. The application examples are not limited to the first and second application examples, and may be, for example, a configuration that combines the first and second application examples.

[0076] [Variations]

[0077] In the farmland management system 1 according to the embodiment, part of the configuration of the information processing device 200 may be provided in the terminal device 100, or part of the configuration of the terminal device 100 may be provided in the information processing device 200. For example, the terminal device 100 may be provided with the function of the analysis unit 250, and image analysis may be performed on the terminal device 100 side and the analysis results may be transmitted to the information processing device 200. Alternatively, the terminal device 100 may be provided with the functions of the analysis unit 250 and the estimation unit 260, and perform image analysis processing and estimation processing. In this case, the trained model 294 may be stored in the storage unit 170 of the terminal device 100, or the terminal device 100 may communicate with an external device such as the information processing device 200 to acquire information about the trained model 294. Furthermore, the function of the learning unit 270 of the information processing device 200 may be provided in an external device connectable via a network NW.

[0078] Furthermore, in an embodiment, when a trained model 294 (analysis model) that receives an image containing soil as input and outputs an analysis result of the crack state (e.g., crack volume) is stored in the memory unit 290, the analysis unit 250 may input an image acquired by the acquisition unit 220 to the trained model 294 to analyze the crack state. In this case, the trained model 294 may also be trained by the learning unit 270 or acquired from an external device.

[0079] According to the above-described embodiment, the information processing device 200 is equipped with an acquisition unit 220 that acquires an image of a predetermined position in the field, an analysis unit 250 that analyzes the crack state of the soil at the predetermined position from the image acquired by the acquisition unit 220, and an estimation unit 260 that estimates the field state including at least the mid-field drying period based on the crack state of the soil analyzed by the analysis unit 250, thereby making it possible to estimate the field state more accurately while reducing equipment costs.

[0080] Specifically, in this embodiment, by focusing on cracks (fissures) in the soil that occur when paddy fields or the like are drained, and quantifying the soil cracks from images, the drainage period can be obtained with higher accuracy. Furthermore, since the drainage period can be estimated from images alone without using soil moisture sensors or water level sensors, it is possible to reduce equipment costs and estimate the state of the field more accurately. Furthermore, according to this embodiment, the drainage state can be estimated from the state of cracks, which allows for optimization of water management for paddy fields.

[0081] Furthermore, according to the embodiment, the period of interim drainage can be quantitatively estimated based on the state of cracks in the soil contained in the image, thereby preventing excessive interim drainage from damaging crop roots and inhibiting growth. Furthermore, according to the embodiment, the water level, soil moisture content, and interim drainage period (such as the appropriate time to end interim drainage) can be estimated from images of the ground in a paddy field during the interim drainage period, making it possible to more accurately estimate the applicability of J-Credit applications and the end time of the interim drainage period, which previously relied on human experience and intuition. Therefore, more appropriate field management can be achieved.

[0082] While the present invention has been described above using the embodiments, the present invention is not limited to these embodiments, and various modifications and substitutions can be made without departing from the spirit of the present invention. For example, the above-described embodiments can be used by all producers who cultivate the target crops, from individual producers to large-scale production corporations. [Explanation of symbols]

[0083] 1...farm field management system, 100...terminal device, 110, 210...communication unit, 120...imaging unit, 130, 230...output unit, 132, 232...display unit, 134, 234...speaker, 140...position information acquisition unit, 150...control unit, 160...application execution unit, 170, 290...storage unit, 200...information processing device, 240...management unit, 250...analysis unit, 260...estimation unit, 270...learning unit, 280...information provision unit, 290...storage unit

Claims

1. an acquisition unit that acquires an image of a predetermined area in a farm field; an analysis unit that analyzes the state of cracks in the soil included in the predetermined area from the image acquired by the acquisition unit; an estimation unit that estimates a field state including at least a mid-season drainage period of the field based on the crack state of the soil analyzed by the analysis unit; An information processing device comprising:

2. the acquisition unit acquires information regarding the soil moisture content or water level of the field, The estimation unit estimates at least a mid-season drainage period of the field based on the crack state of the soil and at least one of the soil moisture content and the water level. The information processing device according to claim 1 .

3. the estimation unit estimates at least one of a soil moisture content and a water level in the field based on the state of cracks in the soil. The information processing device according to claim 1 .

4. Further provided is a learning unit that generates a trained model that inputs information about the crack state and outputs the interim drainage period based on information that associates the crack state and the interim drainage period for each past field, The estimation unit acquires the mid-drying period by inputting information on at least the crack state of the soil into the trained model trained by the learning unit. The information processing device according to claim 1 .

5. The system further includes a learning unit that generates a trained model that uses the image acquired by the acquisition unit as an input and outputs the interim drainage period based on information that associates past images including soil in the field with the interim drainage period, The estimation unit inputs the image acquired by the acquisition unit into a trained model trained by the learning unit, thereby acquiring the mid-drying period of the image included in the image. The information processing device according to claim 1 .

6. The crack state includes information on the amount of cracks included in the image. The information processing device according to claim 1 .

7. The estimation unit estimates the end time of mid-season drainage for the field based on changes in the crack state over time. The information processing device according to claim 1.

8. the analysis unit analyzes the growth of crops in the field included in the image; The information processing device according to claim 1 .

9. An information providing unit that provides information including the mid-drying period estimated by the estimation unit, The information processing device according to claim 1 .

10. The analysis unit analyzes the dryness degree of the predetermined region, the information providing unit provides information regarding water supply to the field based on the dryness degree determined by the analysis unit. The information processing device according to claim 9 .

11. The computer Acquire an image of a predetermined area in the field; Analyzing the state of cracks in the soil included in the predetermined area from the acquired image; and estimating a field condition including at least a mid-season drainage period of the field based on the analyzed crack condition of the soil. Information processing methods.

12. On the computer, Acquiring an image of a predetermined area in a farm field; analyzing the state of cracks in the soil included in the predetermined area from the acquired image; and estimating a field condition including at least a mid-season drainage period of the field based on the analyzed crack condition of the soil. program.

Images