Crop information processing system, crop information processing method, crop information processing device, and crop information processing program

Abstract

[Problem] To more easily and accurately estimate the form quantity of crops. [Solution] Provided is a crop information processing system 1000 characterized by comprising: a data acquisition unit 210 that acquires geocoded data of an agricultural field; a data processing unit 220 that calculates a feature amount of the data; and a crop form quantity estimation unit 230 that estimates a crop form quantity with the feature amount serving as an explanatory variable. The data can be a height map obtained by associating the information of the heights of crops which grow in the agricultural field with respective coordinates and performing mapping.

Description

Crop information processing system, crop information processing method, crop information processing device, and crop information processing program

[0001] The present invention relates to a crop information processing system, a crop information processing method, a crop information processing device, and a crop information processing program.

[0002] Growth diagnosis of crops such as rice has been performed by calculating indices such as the number of stems, leaf area index, leaf age, number of leaves, number of branches, or number of nodes. Therefore, there is a need for a technology that can easily and accurately obtain growth diagnostic indices for crops grown in fields.

[0003] In this regard, Patent Document 1 discloses a farm field breeding data collection device that generates image processing data (two-dimensional orthomosaic, digital surface model, and point cloud) from one of a plurality of image groups and the shooting conditions, and analyzes plant characteristics in the farm field from the image processing data. Patent Document 2 discloses a plant sensor device that uses measurement light and reflected light to determine the distance D from a light emitting unit to a growth condition measurement target, and uses the distance D to determine the plant height value H of the growth condition measurement target.

[0004] However, the invention of Patent Document 1 focuses on identifying the crowns of individual fruit trees and does not acquire detailed indicators for crop growth diagnosis.Furthermore, the invention of Patent Document 2 is specialized in determining plant height, and is therefore insufficient for detailed crop growth diagnosis.

[0005] JP 2023-052423 A Patent No. 6526474 A

[0006] One of the objectives is to estimate the morphological quantity of crops simply and accurately.

[0007] In order to achieve the above-mentioned object, a crop information processing system according to one aspect of the present invention is characterized by having a data acquisition unit that acquires geocoded data in a farm field, a data processing unit that calculates feature quantities of the data, and a crop morphological quantity estimation unit that estimates crop morphological quantities using the feature quantities as explanatory variables.

[0008] The data may be a height map in which information on heights formed by crops growing in the field is mapped in association with each of the coordinates.

[0009] The data may be a reflection intensity map in which the reflection intensity at each coordinate of a specific wavelength of the light beam irradiated onto the farm field is mapped in association with each coordinate of the farm field.

[0010] The data processing unit may calculate a spatial differential value of the data as the feature amount.

[0011] The data acquisition unit may acquire both a height map in which information on heights formed by the field and the crops growing in the field is mapped and linked to each coordinate, and a reflection intensity map in which the reflection intensity of a specific wavelength irradiated onto the field is mapped and linked to each coordinate of the field, the data processing unit may calculate a spatial differential value of the height map and a spatial differential value of the reflection intensity as feature quantities, and the crop morphological quantity estimation unit may estimate the crop morphological quantity using both the spatial differential value of the height map and the spatial differential value of the reflection intensity as explanatory variables.

[0012] The crop morphological quantity may include any one of the number of stems, the leaf area index, the number of leaves, the number of branches, and the number of nodes of the crop.

[0013] The crop physical quantity estimating unit may calculate the crop physical quantity for each point or partial region of the field.

[0014] The system may further include a memory unit that stores the type of crop and the feature amount in correspondence with each other, and the crop morphological quantity estimation unit may refer to the memory unit and estimate the crop morphological quantity using the feature amount, which differs for each crop, as an explanatory variable.

[0015] The crop morphological quantity estimation unit may be configured to estimate the crop morphological quantities from the feature quantities by referring to the storage unit and using the estimation formula.

[0016] The memory unit may store a different estimation formula for each type of crop, and the crop morphological quantity estimation unit may refer to the memory unit to determine the estimation formula according to the type of crop, and estimate the crop morphological quantity using the determined estimation formula.

[0017] In order to achieve the above-mentioned object, a crop information processing method according to another aspect of the present invention includes a data acquisition step of acquiring geocoded data for a farm field, a data processing step of calculating feature quantities of the data, and a crop morphological quantity estimation step of estimating crop morphological quantities using the feature quantities as explanatory variables, all of which are executed by a computer.

[0018] In order to achieve the above-mentioned object, a crop information processing device according to yet another aspect of the present invention includes a data acquisition unit that acquires geocoded data in a farm field, a data processing unit that calculates feature quantities of the data, and a crop morphological quantity estimation unit that estimates crop morphological quantities using the feature quantities as explanatory variables.

[0019] To achieve the above object, a crop information processing program according to another aspect of the present invention causes a computer to execute the following: data acquisition instructions for acquiring geocoded data of a farm field, data processing instructions for calculating feature quantities of the data, and crop morphological quantity estimation instructions for estimating crop morphological quantities using the feature quantities as explanatory variables. Note that the computer program can be provided by downloading via a network such as the Internet, or by recording it on various computer-readable recording media such as a CD-ROM.

[0020] This allows for easy and accurate estimation of crop morphological quantities.

[0021] 1 is a schematic perspective view of a drone included in the crop information processing system according to the present invention. It is a functional block diagram of the drone, user interface device, and crop information processing device included in the crop information processing system. It is a diagram showing (a) an example of a height map, which is one example of data acquired by a data acquisition unit included in the crop information processing system, and (b) an example of a reflectance map, which is another example of data acquired by the data acquisition unit. It is a conceptual diagram explaining how spatial differential values ​​as feature quantities are calculated by a data processing unit included in the crop information processing system, where (a) is a conceptual diagram explaining how spatial differential values ​​are calculated, and (b) is a conceptual diagram explaining how spatial differential values ​​are calculated for an image of a leaf. It is (a) an example of a spatial differential value map calculated from the height map by the data processing unit, and (b) an example of a spatial differential value map calculated from the reflectance map by the data processing unit. It is a conceptual diagram explaining another example of how spatial differential values ​​are calculated by the data processing unit. It is a flowchart for estimating crop morphological quantities by the crop information processing system. It is a scatter plot showing an example of the relationship between the actually measured number of stalks of a crop and the number of stalks estimated by the crop information processing system. Scatter diagrams showing an example of the relationship between the measured values ​​of the number of stalks of a crop and the DSM derivatives obtained by the crop information processing system, plotting the correspondence between the measured values ​​and the DSM derivatives for a plurality of varieties. (a) A scatter diagram showing an example of the relationship between the measured values ​​of the number of stalks of a crop and the number of stalks estimated by the crop information processing system, and a diagram showing a linear regression line. (b) A scatter diagram showing an example of the relationship between the measured values ​​of the number of stalks of a crop and the number of stalks estimated by the crop information processing system, and a diagram showing a non-linear regression curve.

[0022] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following detailed description will be described with reference to the drawings. All figures are illustrative. In the following detailed description, for purposes of explanation, certain specific details are set forth in order to provide a thorough understanding of the disclosed embodiments. However, the embodiments are not limited to these specific details. Additionally, well-known structures and devices are schematically depicted to simplify the drawings.

[0023] First, the configuration of the drone according to the present invention will be described. In this specification, a drone refers to any flying object with multiple rotors, regardless of the power source (electricity, prime mover, etc.) or control method (wireless or wired, autonomous flight type or manual flight type, etc.).

[0024] As shown in FIG. 1 , rotors 101-1a, 101-1b, 101-2a, 101-2b, 101-3a, 101-3b, 101-4a, and 101-4b (also referred to as rotors) are used to fly the drone 100. Eight rotors (four sets of two-stage rotors) are provided, taking into consideration a balance between flight stability, aircraft size, and power consumption. Each rotor 101 is positioned on all four sides of the housing 110 of the drone 100 by an arm extending from the housing 110. Specifically, rotors 101-1a and 101-1b are positioned at the rear left of the drone 100's traveling direction, rotors 101-2a and 101-2b at the front left, rotors 101-3a and 101-3b at the rear right, and rotors 101-4a and 101-4b at the front right. Note that the drone 100's traveling direction in FIG. 1 is the +x direction.

[0025] A lattice-shaped propeller guard is provided on the outer periphery of each set of rotors 101 to prevent interference with foreign objects. Rod-shaped legs 107-1, 107-2, 107-3, and 107-4 extend downward from the rotation axis of rotors 101.

[0026] The motors 102-1a, 102-1b, 102-2a, 102-2b, 102-3a, 102-3b, 102-4a, and 102-4b are means (typically electric motors, but may also be engines, etc.) for rotating the rotors 101-1a, 101-1b, 101-2a, 101-2b, 101-3a, 101-3b, 101-4a, and 101-4b, and one motor is provided for each rotor. The motor 102 is an example of a propulsion device. The upper and lower rotors (e.g., 101-1a and 101-1b) in a set and their corresponding motors (e.g., 102-1a and 102-1b) have axes that are on the same straight line and rotate in opposite directions to ensure the drone's flight stability, etc.

[0027] The housing 110 contains, for example, a flight controller 501, which is a component responsible for overall control of the drone 100. Specifically, the flight controller may be an embedded computer including a CPU, memory, related software, and the like. The flight controller controls the flight of the drone 100 by controlling the rotation speeds of the motors 102-1a, 102-1b, 102-2a, 102-2b, 102-3a, 102-3b, 104-a, and 104-b via control means such as an ESC (Electronic Speed ​​Control) based on input information received from a controller and input information obtained from various sensors (described below). The actual rotation speeds of the motors 102-1a, 102-1b, 102-2a, 102-2b, 102-3a, 102-3b, 104-a, and 104-b are fed back to the flight controller 501, allowing monitoring of whether the motors are rotating normally. Alternatively, an optical sensor or the like may be provided on the rotor 101 so that the rotation of the rotor 101 is fed back to the flight controller 501.

[0028] The fields are rice paddies, farmlands, etc. that are the subject of photography by the drone 100.

[0029] Typically, drone 100 takes off from a takeoff and landing point outside the field and returns to takeoff and landing point 406 after photographing the field or when charging is required. The flight path (entry path) from the takeoff and landing point to the target field may be stored in advance in a group of computers such as an agricultural cloud or related software, or may be input by the user before takeoff. The takeoff and landing point may be a virtual point defined by coordinates stored in drone 100, or there may be a physical takeoff and landing platform.

[0030] The drone 100 includes a measurement unit that measures the field and acquires data for growth diagnosis. The measurement unit may be, for example, a camera that receives visible light. The measurement unit may also be a multispectral camera that receives multiple light beams with different wavelengths. The multiple light beams may be light beams in any frequency range, such as red light (wavelength of approximately 650 nm) and near-infrared light (wavelength of approximately 774 nm). Because red light corresponds to the absorption spectrum of chlorophyll contained in plants, a configuration that receives and photographs red light allows for special photography of plants.

[0031] Overview of the Control System As shown in Fig. 2, the crop information processing system 1000 is a system including, for example, a drone 100, a crop information processing device 200, and a user interface device 300, which are communicatively connected to each other via a network NW. The crop information processing device 200 may be configured as hardware or may be configured on the cloud. The drone 100, the crop information processing device 200, and the user interface device 300 may be connected to each other wirelessly, or some or all of them may be connected by wire.

[0032] 2 is an example, and a component may include another component, or a functional unit of each component may be provided by another component. For example, some or all of the functions of the crop information processing device 200 may be installed in the drone 100.

[0033] The user interface device 300 only needs to include an input unit and a display unit for an operator, and may be realized by the functions of the operation device 401. The user interface device 300 may also be a personal computer, and information may be input to a UI on the web via a web browser installed on the personal computer and displayed.

[0034] ●Functional parts of the drone The drone 100 is equipped with an arithmetic unit such as a CPU (Central Processing Unit) for performing information processing, and storage devices such as RAM (Random Access Memory) and ROM (Read Only Memory), and therefore has at least a flight control unit 1001 and a sensing unit 1002 as software resources.

[0035] The flight control unit 1001 is a functional unit that operates the motor 102 and controls the flight and takeoff and landing of the drone 100. The flight control unit 1001 is realized by, for example, a flight controller, and controls the flight altitude, flight speed, and flight path to fly the drone 100 above the field.

[0036] The sensing unit 1002 is a functional unit that acquires images of crops growing in the field via the measurement unit while the drone 100 is flying above the field. The growth information includes images of the crops for diagnosing the growth state of the crops.

[0037] The sensing unit 1002 acquires images of the crops using the measurement unit. The sensing unit 1002 may have a beam splitter and acquire only light rays within a predetermined frequency range from the light source. The light rays received by the sensing unit 1002 include light rays transmitted from the sensing unit 1002 that are mainly reflected from the crops. The drone 100 flies over the field using the flight control unit 1001, and acquires growth information about the crops growing in the field by receiving reflected light from the field using the sensing unit 1002.

[0038] In this embodiment, the drone 100 is configured to be equipped with a sensing unit 1002, but the technical idea of ​​the present invention is not limited to this, and the sensing unit 1002 may be provided by a land-based machine that is provided by the crop information processing system 1000 or connected to the crop information processing system 1000, or by a fixed device that can take bird's-eye views of the field.

[0039] Functional Units of the Crop Information Processing System As shown in Figure 2, the crop information processing device 200 is a device that diagnoses the plants, i.e., the crops, growing in the field over which the drone 100 flies, based on information acquired by the drone 100. The crop information processing device 200 includes a calculation device such as a CPU (Central Processing Unit) for executing information processing, and storage devices such as RAM (Random Access Memory) and ROM (Read Only Memory), and thus includes at least a data acquisition unit 210, a data processing unit 220, a crop morphological quantity estimation unit 230, a result output unit 240, and a memory unit 250 as software resources.

[0040] The data acquisition unit 210 is a functional unit that acquires geocoded data of the farm field. The data acquisition unit 210 acquires data measured by the sensing unit 1002 of the drone 100.

[0041] The data acquisition unit 210 orthorectifies the acquired data and creates an orthogonally projected aerial photograph image. In the example shown in the figure, the farm field is approximately rectangular. The orthorectified data is divided into multiple small areas, and data is associated with each small area. In other words, the data is geocoded.

[0042] The data may be, for example, a height map in which height information formed by crops growing in a field is mapped to each coordinate. More specifically, the data may be a height map in which heights obtained using a digital surface model (DSM) are mapped. The data acquisition unit 210, for example, extracts feature points from the data and generates three-dimensional point cloud data. Furthermore, the height of each area is calculated by comparing feature points from multiple images. Figure 3(a) shows an example of a height map obtained using a digital surface model. In this example, height information is mapped in color to areas corresponding to the field.

[0043] The data acquired by the data acquisition unit 210 may also be a reflection intensity map in which the reflection intensity at each coordinate of a specific wavelength of the light beam irradiated onto the field is mapped to each coordinate of the field. For example, a reflection intensity map in which the reflection intensity of wavelengths absorbed by plants, such as red light or near-infrared light, is mapped can clearly distinguish between areas where plants are present and areas where plants are not present. The data acquired by the data acquisition unit 210 may also be a reflectance map in which the reflectance calculated from the reflection intensity is mapped. The data acquisition unit 210 may acquire the reflection intensity, calculate the reflectance, and generate a reflectance map. Figure 3(b) shows a reflectance map in which the reflectance calculated based on the reflection intensity of near-infrared light is mapped.

[0044] Furthermore, the data acquired by the data acquisition unit 210 may be any suitable three-dimensional measurement data that can ensure a predetermined or higher resolution, and may be data measured by a laser, for example.

[0045] The data processing unit 220 is a functional unit that calculates the feature amounts of the data acquired by the data acquisition unit 210. The data processing unit 220 calculates, for example, the spatial differential value of the geocoded data as the feature amount.

[0046] 4 is a conceptual diagram illustrating how the data processing unit 220 calculates spatial differential values ​​as feature quantities. As shown in FIG. 4(a), the measurement data acquired by measuring a farm field is subdivided into multiple microregions M11 to M33, and data is linked to each microregion. For example, for a central target microregion M22, the data processing unit 220 calculates the feature quantities of the microregion M22 based on the values ​​of the adjacent microregions M11 to M21 and M23 to M33.

[0047] For example, the data processing unit 220 calculates the difference between the target microregion M22 and each of the adjacent microregions M11-M21 and M23-M33, and sets the maximum of the calculated differences as the feature amount of the target microregion M22. Alternatively, the data processing unit 220 may calculate the difference between the microregions M12, M21, M23, and M32 adjacent to the target microregion M22 on the top, bottom, left, and right, or the diagonally adjacent microregions M11, M13, M31, and M33, and set the maximum value as the feature amount of the target microregion M22. Furthermore, the data processing unit 220 may perform a second differentiation process on either or both of the microregions M12, M21, M23, and M32 adjacent to the central microregion M22 on the top, bottom, left, and right, and the diagonally adjacent microregions M11, M13, M31, and M33, and set the maximum value as the feature amount of the microregion M22. In the following description, the maximum value of the difference between the target microregion M22 and the adjacent microregions M12, M21, M23, and M32 on the height map using the digital surface model is also referred to as the DSM differential coefficient of the target microregion M22. In other words, the DSM differential coefficient may be used as the feature value of the target microregion M22.

[0048] 4(b) is a conceptual diagram illustrating how a spatial differential value is calculated for an image of a leaf L. In the figure, a leaf L is present in a microregion M11, while a leaf L is absent in a microregion M22. The data processing unit 220 calculates differences D11, D13, D31, and D33 between the geocoding data of the target microregion M22 and the geocoding data of microregions M11, M13, M31, and M33 diagonally adjacent to the target microregion M22, respectively, and sets the maximum value as the spatial differential value of the target microregion M22.

[0049] Fig. 5(a) is a diagram showing an example of mapping spatial differential values ​​calculated from the height map of Fig. 3(a) by the calculation method described with reference to Fig. 4(a). Fig. 5(b) is a diagram showing an example of mapping spatial differential values ​​calculated from the reflectance map of Fig. 3(b) by the calculation method described with reference to Fig. 4(a).

[0050] FIG. 6 is a conceptual diagram illustrating another example of a calculation method for calculating spatial differential values ​​as feature quantities. In this example, the data processing unit 220 calculates the feature quantity of the central target microregion M33 based on the values ​​of the microregions M23 and M43 adjacent above and below the central target microregion M33, and the values ​​of the microregions M13 and M53 adjacent above and below the microregions M23 and M43. More specifically, the difference values ​​between the microregions M13, M23, M43, and M53 and the target microregion M33 are calculated, and the difference between the maximum and minimum values ​​is used as the feature quantity. This configuration allows the calculation of feature quantities that take into account the values ​​of microregions two regions away from the target microregion M33. Note that, although the difference between the microregions two regions away is calculated here, feature quantities may also be calculated using values ​​of microregions multiple regions away.

[0051] The crop morphological quantity estimation unit 230 shown in FIG. 2 is a functional unit that estimates the crop morphological quantities using the feature quantities as explanatory variables.

[0052] In the aspect of generating a height map, the difference between a micro-region where leaf contours are measured and a micro-region where no leaves are present is larger than the difference in an area where leaves are continuously present. Therefore, the more leaf contours there are, i.e., the greater the number of leaves, the larger the spatial differential value of the target micro-region. Therefore, the crop morphological quantity estimating unit 230 estimates that the larger the spatial differential value of the micro-region, the larger the number of leaves. Furthermore, the crop morphological quantity estimating unit 230 estimates that the larger the spatial differential value of the target micro-region, the larger the leaf area index (LAI).

[0053] Furthermore, the difference between a micro-region where stem contours are measured and a micro-region where no stems are present is larger than the difference in areas where stems are continuously present. Therefore, the more stem contours there are, i.e., the more stems there are, the larger the spatial differential value of the target micro-region. Therefore, the crop morphological quantity estimating unit 230 estimates that the larger the spatial differential value of the micro-region, the larger the number of stems. Furthermore, the crop morphological quantity estimating unit 230 estimates that the larger the spatial differential value of the target micro-region, the larger the number of stems extending from one crop, i.e., the number of branches or tillers.

[0054] The storage unit 250 may store an estimation formula for estimating the crop morphological quantities from the feature quantities. The crop morphological quantity estimation unit 230 refers to the storage unit 250 and estimates the crop morphological quantities from the feature quantities by using the estimation formula. The estimation formula may be, for example, a linear line or a curve. The predetermined estimation formula can be obtained from the feature quantities and crop morphological quantity data by, for example, the least squares method. With this configuration, the morphological quantities of crops in a field can be easily and accurately determined based on the geocoding data acquired by the drone 100.

[0055] The crop morphological quantity estimation unit 230 may estimate the morphological quantities for each plant stalk. Alternatively, the crop morphological quantity estimation unit 230 may subdivide the field into a plurality of regions and estimate the morphological quantities for each region. The crop morphological quantity estimation unit 230 may subdivide the field into a mesh pattern, for example. Each region is different from the microregion for which feature quantities are calculated and is, for example, a 1 m square rectangle formed by a plurality of microregions. The crop morphological quantity estimation unit 230 may calculate the crop morphological quantities for each predetermined point in the field. Furthermore, the crop morphological quantity estimation unit 230 may estimate the crop morphological quantities for each wavelength at which reflection intensity is measured.

[0056] The target crop for which the crop information processing device 200 estimates morphological quantities is assumed to be wheat, but the technical scope of the present invention is not limited to this and may be any plant for which morphological quantities can be estimated mainly by photographing from the air.

[0057] The crop information processing device 200 may be capable of estimating crop morphological quantities for multiple types of crops. The storage unit 250 may store a different estimation formula for each type of crop, and estimate the morphological quantities from the feature quantities using a different estimation formula for each type or variety. Alternatively, the storage unit 250 may store a different feature quantity for each type of crop, and the crop morphological quantity estimation unit 230 may estimate the crop morphological quantities using the different feature quantity for each crop as an explanatory variable. This configuration allows for more accurate morphological quantity estimation that captures morphological characteristics that differ depending on the type of crop.

[0058] The crop information processing device 200 may estimate crop morphological quantities by acquiring and referencing multiple types of geocoded data. That is, the data acquisition unit 210 may acquire either a height map or a reflection intensity map, or both. The data processing unit 220 may calculate the spatial differential values ​​of the height map and the spatial differential values ​​of reflection intensity as feature quantities. Furthermore, the crop morphological quantity estimation unit 230 may estimate crop morphological quantities using both the spatial differential values ​​of the height map and the spatial differential values ​​of reflection intensity as explanatory variables. In this case, for example, the storage unit 250 may pre-store an equation indicating the correlation between the spatial differential values ​​of the height map, the spatial differential values ​​of reflection intensity, and the crop morphological quantities. Furthermore, the crop morphological quantity estimation unit 230 estimates the crop morphological quantities from the spatial differential values ​​of the height map and the spatial differential values ​​of reflection intensity by referencing the equation.

[0059] The result output unit 240 is a functional unit that outputs the estimation results of the crop morphological quantities. The result output unit 240 outputs, for example, any of the number of stalks, leaf area index, leaf age, number of leaves, number of branches, and number of nodes of the crop. The result output unit 240 may cause the user interface device 300 to display the determination result. Furthermore, the result output unit 240 may cause the determination result to be displayed on the screen of a personal computer, or may cause the determination result to be displayed on a web-based UI via a web browser installed on the personal computer. When the user interface device 300 accepts a selection of the crop morphological quantities, the result output unit 240 may output the selected crop morphological quantities to the user interface device 300.

[0060] Flowchart for Estimating Crop Morphological Quantities As shown in Figure 7, first, the drone 100 flies over the field, measures the field using the sensing unit 1002 of the drone 100, and the data acquisition unit 210 acquires the measurement data (S11). Next, the data acquisition unit 210 generates an orthoimage based on the acquired image (S12). Next, the data acquisition unit 210 acquires a height map or a reflection intensity map as geocoding data (S13). Next, the data processing unit 220 of the crop information processing device 200 calculates feature quantities at each coordinate of the geocoding data (S14). Next, the crop morphological quantity estimation unit 230 estimates the crop morphological quantities using the feature quantities as explanatory variables (S15).

[0061] Example (1) Figure 8 shows an example graph comparing the number of stalks estimated by the crop information processing device 200 according to the present invention with the actual measured number of stalks of a crop. The horizontal axis represents the measured number of stalks, and the vertical axis represents the estimated number of stalks. The circle plots represent the estimated number of stalks calculated without performing image selection processing. The triangle plots represent the estimated number of stalks calculated using images extracted through image selection processing. Image selection processing selects images suitable for morphological estimation, such as whether the crop is properly included in the image. Known image analysis techniques can be used. Line L1 indicates the position where the estimated number of stalks matches the actual measured number of stalks. It can be seen that the circle and triangle plots are concentrated near line L1. Furthermore, the triangle plots selected through image selection are even closer to line L1. In other words, the crop information processing device 200 can accurately estimate the number of stalks of a crop by calculating the number of stalks from feature values ​​using a predetermined estimation formula.

[0062] Figure 9 is an example of a scatter plot plotting the relationship between the measured values ​​of the number of stalks of a crop and the DSM derivative for each variety. The horizontal axis represents the measured values, and the vertical axis represents the DSM derivative. Although there is a strong positive correlation between the measured values ​​and the DSM derivative within each variety, the DSM derivative varies depending on the variety even for the same number of stalks. In other words, by using a different formula for calculating the estimated number of stalks from the DSM derivative for each variety, the number of stalks can be estimated more accurately. In this case, the memory unit 250 stores the type of crop and the feature values ​​in association with each other. The crop morphological quantity estimation unit 230 references the memory unit 250 and estimates the crop morphological quantities using the feature values, which differ for each crop, as explanatory variables.

[0063] Example (2) Figure 10(a) is another example of a scatter plot plotting the relationship between the measured values ​​of crop stalk number and image features (DSM derivatives). The circle plots show image features calculated without image selection processing, with a correlation coefficient of 0.84. The triangle plots show image features calculated using images extracted through image selection processing, with a correlation coefficient of 0.90. This indicates that both image features have a strong positive correlation with the measured values. Line L11 is a regression line for the circle plots. The RMSE (root-mean-square error) between this regression line L11 and the plot is 33.8. Line L12 is a regression line for the triangle plots. The RMSE (root-mean-square error) between this regression line L12 and the plot is 15.7.

[0064] Figure 10(b) shows a quadratic regression curve L21 plotted against the circle plots in Figure 10(a). In this way, the number of stalks may be estimated from the image feature values ​​by performing nonlinear regression between the image feature values ​​and the measured number of stalks.

[0065] (Technically Significant Effects of the Present Invention) In the crop information processing system according to the present invention, the morphological quantity of a crop can be estimated more simply and accurately.

Claims

1. A crop information processing system comprising: a data acquisition unit that acquires geocoded data of a farm field; a data processing unit that calculates feature quantities of the data; and a crop morphological quantity estimation unit that estimates crop morphological quantities using the feature quantities as explanatory variables.

2. The crop information processing system according to claim 1, wherein the data is a height map in which height information formed by crops growing in the field is linked to each of the coordinates and mapped.

3. The crop information processing system according to claim 1, wherein the data is a reflection intensity map in which the reflection intensity at each coordinate of a specific wavelength of the light beam irradiated to the field is linked to each coordinate of the field.

4. The crop information processing system according to claim 1, wherein the data processing unit calculates a spatial differential value of the data as the feature amount.

5. The crop information processing system of claim 1, wherein the data acquisition unit acquires both a height map in which information on height formed by the field and the crops growing in the field is linked to each coordinate and mapped, and a reflection intensity map in which the reflection intensity of a specific wavelength irradiated to the field is linked to each coordinate of the field, the data processing unit calculates the spatial differential value of the height map and the spatial differential value of the reflection intensity as feature values, and the crop morphological quantity estimation unit estimates the crop morphological quantities using both the spatial differential value of the height map and the spatial differential value of the reflection intensity as explanatory variables.

6. The crop information processing system according to claim 1, wherein the crop morphological quantities include any one of the number of stalks, the leaf area index, the number of leaves, the number of branches and the number of nodes of the crop.

7. The crop information processing system according to claim 1, wherein the crop morphological quantity estimating unit calculates the crop morphological quantities for each point or partial area of ​​the field.

8. The crop information processing system of claim 1, further comprising a memory unit that stores the crop type and the feature amount in correspondence with each other, and the crop morphological quantity estimation unit refers to the memory unit and estimates the crop morphological quantities using the feature amount, which differs for each crop, as an explanatory variable.

9. A crop information processing system as described in claim 1, further comprising a memory unit that stores an estimation equation for estimating the crop morphological quantities from the feature quantities, wherein the crop morphological quantity estimation unit estimates the crop morphological quantities from the feature quantities by referring to the memory unit and using the estimation equation.

10. The crop information processing system of claim 9, wherein the memory unit stores a different estimation formula for each type of crop, and the crop morphological quantity estimation unit determines the estimation formula according to the type of crop by referring to the memory unit, and estimates the crop morphological quantity using the determined estimation formula.

11. A crop information processing method, comprising: a data acquisition step of acquiring geocoded data of a farm field; a data processing step of calculating feature quantities of the data; and a crop morphological quantity estimation step of estimating crop morphological quantities using the feature quantities as explanatory variables, the data acquisition step being carried out by a computer.

12. A crop information processing device comprising: a data acquisition unit that acquires geocoded data of a farm field; a data processing unit that calculates feature quantities of the data; and a crop morphological quantity estimation unit that estimates crop morphological quantities using the feature quantities as explanatory variables.

13. A crop information processing program that causes a computer to execute the following: data acquisition instructions for acquiring geocoded data of a farm field; data processing instructions for calculating features of the data; and crop morphological quantity estimation instructions for estimating crop morphological quantities using the features as explanatory variables.

Images