Information processing apparatus, information processing method, and program

Abstract

To improve work efficiency while suppressing the cost associated with pesticide spraying in the field. 【Solution means】An acquisition unit that acquires a plurality of aerial images obtained by imaging the field from different positions above the field by a camera mounted on an unmanned aircraft, a generation unit that generates a height difference map representing the height difference of the surface of the field from the plurality of aerial images, and a calculation unit that calculates a spraying position for spraying pesticides and a spraying amount corresponding to the spraying position in the field based on the height difference map. An information processing apparatus comprising:

Description

Technical Field

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

Background Art

[0002] Conventionally, a technique is known in which an aerial image of a field is acquired by an unmanned aircraft, and based on the acquired aerial image, a plan for spraying agricultural chemicals on the field is made. For example, Patent Document 1 describes that based on an aerial image of a field, the progress of a disease occurring in a crop is determined, and when the progress is in the late stage, spraying of agricultural chemicals is determined as a countermeasure.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The above conventional technique determines the spraying concentration of agricultural chemicals according to the progress of the disease. On the other hand, in recent years, damage to paddy rice by the rice water weevil has become apparent, and the damage by the rice water weevil tends to concentrate in places where the water depth in the paddy field is deeper than 4 centimeters. In this regard, the conventional technique uniformly sprays agricultural chemicals throughout the field regardless of the distribution density of the rice water weevil, and while the cost associated with spraying agricultural chemicals increases, the working efficiency decreases.

[0005] One object of the present invention is to provide an information processing apparatus, an information processing method, and a program that can improve the working efficiency while suppressing the cost associated with spraying agricultural chemicals on a field in consideration of such circumstances.

Means for Solving the Problems

[0006] An information processing apparatus according to an aspect of the present invention includes an acquisition unit that acquires a plurality of aerial images obtained by imaging a field from different positions above the field by a camera mounted on an unmanned aircraft, a generation unit that generates a height difference map representing the height difference of the surface of the field from the plurality of aerial images, and a calculation unit that calculates a spraying position for spraying agricultural chemicals and a spraying amount corresponding to the spraying position in the field based on the height difference map.

[0007] The generation unit may generate a DSM (digital surface model) model representing the surface of the field from the plurality of aerial images, calculate an average value of elevation values of each position on the surface indicated by the DSM model, and generate the height difference map by subtracting the average value from the elevation value of each position on the surface to obtain the height difference.

[0008] The calculation unit may determine, as the spraying position, a position corresponding to a height difference that is lower than a predetermined value or more, based on a height difference corresponding to a threshold value among the height differences of each position indicated by the height difference map.

[0009] The threshold value may be a height difference representing the water surface height of the field when water enters the field.

[0010] The calculation unit may classify the height differences of each position indicated by the height difference map into a plurality of levels and calculate the spraying amounts corresponding to the plurality of levels.

[0011] The information processing apparatus may further include an output unit that outputs a spraying instruction map indicating the spraying position and the spraying amount in the field to the unmanned aircraft.

[0012] The output unit may divide the field into meshes of a predetermined size, determine representative values of the spraying position and the spraying amount included in each divided mesh, and output the determined representative values of the spraying position and the spraying amount as the spraying instruction map.

[0013] The field is a paddy field for rice cultivation, and the agricultural chemical may be one for controlling the rice water weevil.

[0014] An information processing method according to an aspect of the present invention is a method in which a computer acquires a plurality of aerial images obtained by imaging a field from different positions above the field by a camera mounted on an unmanned aircraft, generates a height difference map representing the height difference of the surface of the field from the plurality of aerial images, and calculates a spraying position for spraying an agricultural chemical and a spraying amount corresponding to the spraying position in the field based on the height difference map.

[0015] A program according to an aspect of the present invention causes a computer to acquire a plurality of aerial images obtained by imaging a field from different positions above the field by a camera mounted on an unmanned aircraft, generate a height difference map representing the height difference of the surface of the field from the plurality of aerial images, and calculate a spraying position for spraying an agricultural chemical and a spraying amount corresponding to the spraying position in the field based on the height difference map.

Advantages of the Invention

[0016] According to the present invention, it is possible to provide an information processing apparatus, an information processing method, and a program that can improve work efficiency while suppressing the cost associated with spraying agricultural chemicals on a field.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

[0018] [Overview] Hereinafter, with reference to the drawings, the information processing apparatus 100 according to an embodiment of the present invention will be described. FIG. 1 is a diagram showing an example of the usage environment and configuration of the information processing apparatus 100. The information processing apparatus 100 operates in cooperation with, for example, the aircraft 10, the relay device 20, and the terminal device 30.

[0019] The flying object 10 is, for example, a drone (unmanned aerial vehicle) with a camera 12 attached thereto as an imaging device. The flying object 10 is assumed to be equipped with an RTK (Real-Time Kinematics) function for performing relative positioning by the RTK method in combination with the camera 12. The flying object 10 flies over a field F where rice is growing, images the field F from different positions in the sky with the camera 12, and acquires longitude, latitude, and altitude information (hereinafter referred to as position information) of the flying object 10 at the time of imaging using the RTK function. In this case, the flying object 10 may fly along a specified route programmed by the software installed therein to image the field F, or may be manually operated by, for example, a relay device 20. The flying object 10 stores a plurality of aerial images captured by the camera 12 together with the position information of the aerial images in a memory card mounted on the flying object 10 or transmits them to the relay device 20. The flying object 10 is an example of an "unmanned aerial vehicle". In the present embodiment, the field F is a paddy field (paddy field for rice cultivation), and the flying object 10 is configured to image the field F before flooding.

[0020] The relay device 20 is a terminal device such as a tablet terminal equipped with an application for operating the flying object 10 and receiving and displaying aerial images transmitted from the flying object 10. The relay device 20 displays the aerial image as an RGB image, or transmits the received aerial image and position information to the information processing device 100 via the network NW. The network NW is an arbitrary network such as, for example, a LAN, a WAN, or an Internet line, and may be wired or wireless.

[0021] The terminal device 30 is a computer device such as, for example, a personal computer, a smartphone, or a tablet terminal. The terminal device 30 communicates with the information processing device 100 via the network NW and displays the information received from the output unit 140 of the information processing device 100 described later.

[0022] The information processing device 100 is a server device such as a web server. The information processing device 100 receives a plurality of aerial images and position information from the relay device 20, and uses the method described later to generate a height difference map representing the height difference of the surface of the farm field F from the plurality of aerial images and the position information. Based on the generated height difference map, the spraying position of the agricultural chemical in the farm field F and the spraying amount corresponding to the spraying position are calculated. Here, the pest to be controlled by the agricultural chemical is, for example, the rice water weevil that inhabits paddy fields. Hereinafter, the details of the processing executed by the information processing device 100 will be described.

[0023] [Functional Configuration] The information processing apparatus 100 includes, for example, an acquisition unit 110, a generation unit 120, a calculation unit 130, an output unit 140, and a storage device 150. Each of the acquisition unit 110, the generation unit 120, the calculation unit 130, and the output unit 140 is realized, for example, by a hardware processor such as a CPU (Central Processing Unit) executing a program (software). 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), or a GPU (Graphics Processing Unit), or may be realized by the cooperation of software and hardware. The program may be stored in advance in a storage device (a storage device having a non-transitory storage medium) such as an HDD (Hard Disk Drive) or a flash memory, or may be stored in a removable storage medium (a non-transitory storage medium) such as a DVD or a CD-ROM, and may be installed by mounting the storage medium on a drive device. The storage device 150 stores, for example, a height difference map 150A generated by the generation unit 120, a histogram distribution 150B calculated by the calculation unit 130, and a scatter instruction map 150C output by the output unit 140. The storage device 150 is, for example, an HDD, a flash memory, a RAM (Random Access Memory), or the like.

[0024] The acquisition unit 110 reads an aerial image and position information from a memory card mounted on the airframe 10 main body, or receives a plurality of aerial images and position information obtained by imaging the field F from different positions above the field F by the camera 12 of the airframe 10 from the relay device 20. When the acquisition unit 110 acquires a plurality of aerial images and position information, the acquisition unit 110 delivers the aerial images and position information to the generation unit 120.

[0025] First, based on the aerial image and the position information acquired by the acquisition unit 110, the generation unit 120 generates an orthomosaic image and a DSM (digital surface model) model representing the farmland F using a known method. The DSM model is three-dimensional data including the ground surface of the farmland F and the elevations of the surfaces of features on the ground surface, and includes the heights of structures such as buildings and trees. As described above, since the position information is highly accurate positioning data by the RTK function, GCP (ground control point) calibration is not required. However, when the aircraft 10 performs, for example, GNSS single-point positioning, GCP calibration is required.

[0026] Next, the generation unit 120 calculates the average value of the elevation values at each position on the surface represented by the generated DSM model, and subtracts the average value from the elevation value at each position on the surface to obtain the height difference, thereby generating a height difference map 150A. FIG. 2 is a diagram showing an example of the height difference map 150A generated by the generation unit 120. As an example, FIG. 2 shows a height difference map 150A in which the average value is set to 0 cm and the obtained height differences are classified into ranges of 5 cm each. However, the present invention is not limited to such a configuration, and the height difference map 150A may be generated by classifying the obtained height differences in any unit (which may be in meters).

[0027] Also, as an example, FIG. 2 shows contour lines of the height difference map 150A at intervals of 5 cm, but the contour lines may not be displayed. When the farmland F is a paddy field for rice cultivation, the height difference map 150A may target only the limited range (for example, from -20 cm to +20 cm) of the surface represented by the DSM model. In that case, portions deviating from the limited range will be excluded from the generation target of the height difference map 150A as ridges or access roads.

[0028] First, the calculation unit 130 calculates a histogram distribution 150B representing the frequency of occurrence of height differences at each position in the farm field F based on the height difference map 150A. FIG. 3 is a diagram showing an example of the histogram distribution 150B calculated by the calculation unit 130. In the graph of FIG. 3, the horizontal axis represents the height difference, and the vertical axis represents the occurrence density. In FIG. 3, the point P1 represents a height difference (an example of the "threshold value" in the claims) that coincides with the water surface height when the water inflow into the farm field F is completed. The height difference that coincides with the water surface height is, for example, preset as the upper 95% point or 97.5% point of the histogram distribution 150B.

[0029] When the calculation unit 130 identifies the threshold value, next, based on the height difference corresponding to the threshold value, it determines the positions corresponding to height differences of 4 cm or more in depth as the pesticide spraying positions. For example, in the case of the graph shown in FIG. 2, the calculation unit 130 identifies the height difference P2 of 4 cm based on the height difference of the point P1 identified as the threshold value, and determines the positions corresponding to the height differences below the point P2 as the spraying positions. That is, in the case of the graph shown in FIG. 2, the positions corresponding to the height differences located on the left side of the point P2 are determined as the pesticide spraying range. This is because the apple snail generally tends to cause damage at positions where the water depth is 4 cm or less. 4 cm is an example of the "predetermined value" in the claims.

[0030] The output unit 140 outputs a spraying instruction map 150C indicating the spraying positions and spraying amounts in the farm field F to the aircraft 10 or the terminal device 30 via the network NW. Here, the format of the output spraying instruction map 150C may be, for example, a vector format such as a shape file (SHP), GeoJSON, etc., or a raster file format such as GeoTiff, or an ISOXML format used for general agricultural machinery.

[0031] FIG. 4 is a diagram showing an example of the spraying instruction map 150C output by the output unit 140. The spraying instruction map 150C shown in FIG. 4 represents a case where, as an example, the output unit 140 divides the area representing the farm field F into 1 m-sized meshes and determines the spraying amount according to the height difference at the position corresponding to the representative point among the plurality of coordinates included in each mesh. The representative point in this case may be any point such as a randomly selected point. When the height difference of the representative point belongs to the pesticide spraying range of the graph shown in FIG. 3 (when the water depth is 4 cm or less), the output unit 140 determines the mesh including the representative point as a spraying position for spraying pesticides at a predetermined spraying amount (in FIG. 4, for example, 40 kg / ha). On the other hand, when the height difference of the representative point does not belong to the pesticide spraying range of the graph shown in FIG. 3, the output unit 140 determines the mesh including the representative point as a non-spraying position where pesticides at a predetermined spraying amount are not sprayed. In this way, by spraying pesticides only at positions where there is a high possibility of damage caused by the rice water weevil, it is possible to improve the working efficiency while suppressing the cost associated with pesticide spraying in the farm field.

[0032] The output unit 140 transmits the spraying instruction map 150C to the aircraft 10 loaded with pesticides, and the aircraft 10 sprays pesticides on the farm field F according to the spraying instruction map 150C. At this time, the farm field F where pesticides are sprayed is a paddy field after the flooding for rice cultivation is completed. Further, as another aspect, the output unit 140 may transmit the spraying instruction map 150C to the terminal device 30, and the user of the terminal device 30 may check the content of the displayed spraying instruction map 150C. In this case, the terminal device 30 causes the user to confirm whether or not to consent to the generated spraying instruction map 150C, and when the user consents to the generated spraying instruction map 150C, information indicating consent is transmitted to the information processing device 100, and then the spraying instruction map 150C may be transmitted to the aircraft 10. As yet another aspect, the terminal device 30 may enable the user to edit the generated spraying instruction map 150C, and after the user edits and finalizes the spraying instruction map 150C, the finalized spraying instruction map 150C may be transmitted to the information processing device 100.

[0033] [Another method for determining the spraying amount and spraying position] In the above description, based on the height difference that matches the water surface height, the position corresponding to a height difference of 4 cm or more is determined as the pesticide spraying position, and a predetermined spraying amount (40 kg / ha) is sprayed on the determined spraying position (that is, there are two options: spraying or not spraying). Hereinafter, another method for determining the pesticide spraying amount and spraying position will be described.

[0034] FIG. 5 is a diagram showing another example of the height difference map 150A generated by the generation unit 120. FIG. 5 is a height difference map 150A obtained by imaging an area of a field F different from that in FIG. 2. Similar to FIG. 2, the average value is set to 0 cm, and the obtained height differences are classified into ranges of every 5 cm.

[0035] The calculation unit 130 calculates a histogram distribution 150B representing the appearance frequency of the height differences at each position in the field F based on the height difference map 150A by the same method as described in FIG. 3. FIG. 6 is a diagram showing another example of the histogram distribution 150B calculated by the calculation unit 130. In the graph of FIG. 6, the horizontal axis represents the height difference, and the vertical axis represents the appearance density. As shown in FIG. 6, the calculation unit 130 classifies the height differences at each position indicated by the height difference map into a plurality of levels (for example, less than -5 cm, -5 cm or more and less than 5 cm, 5 cm or more), and calculates the spraying amounts corresponding to the plurality of levels. For example, in the case of FIG. 6, the calculation unit 130 determines to spray pesticides with a spraying amount of 40 kg / ha at positions with a height difference of less than -5 cm, determines to spray pesticides with a spraying amount of 10 kg / ha at positions with a height difference of -5 cm or more and less than 5 cm, and determines not to spray pesticides at positions with a height difference of 5 cm or more. In this way, the calculation unit 130 may spray a plurality of levels of spraying amounts at the spraying positions according to a plurality of levels of height differences, rather than just two options of spraying or not spraying a predetermined spraying amount at the spraying position.

[0036] FIG. 7 is a diagram showing another example of the spraying instruction map 150C output by the output unit 140. Similar to the one shown in FIG. 4, the spraying instruction map 150C shown in FIG. 7 also represents the case where the output unit 140 divides the area representing the field F into 1 m-sized meshes and determines the spraying amount according to the height difference at the position corresponding to the representative point selected from a plurality of points included in each mesh. Different from the spraying instruction map 150C in FIG. 4, the spraying instruction map 150C in FIG. 7 indicates whether to spray pesticides with a spraying amount of 40 kg / ha, spray pesticides with a spraying amount of 10 kg / ha, or not spray pesticides according to the level division of the height difference at each position in the area of the field F. In this way, by dividing and determining the spraying amount of pesticides into multiple levels according to the likelihood of damage caused by the apple snail, it is possible to improve the working efficiency while suppressing the cost associated with pesticide spraying in the field.

[0037] Note that the method for determining the spraying amount does not have to be the two methods described above, and may be determined by any calculation formula that takes at least the damage tendency of the apple snail into account, uses the height difference of the position in the field F as an input variable, and uses the spraying amount as an output variable. In that case, parameters such as other growth indices calculated from a plurality of aerial images captured by the camera 12 may be considered.

[0038] Also, the spraying position of the pesticide is not limited to the position where the water depth is 4 cm or less, and may be changed and set in advance by the administrator of the field F in consideration of the characteristics of the field F. For example, when the field F has an overall flat shape, the administrator of the field F may set 3 cm as a predetermined value, and the calculation unit 130 may determine the position corresponding to the height difference 3 cm or deeper as the spraying position of the pesticide based on the height difference that coincides with the water surface height.

[0039] [Processing flow] Next, with reference to FIG. 8, the processing flow executed by the information processing apparatus 100 will be described. FIG. 8 is a flowchart showing an example of the processing flow executed by the information processing apparatus 100. The processing of the flowchart shown in FIG. 8 is executed before the water is introduced into the field F, which is a paddy field for rice cultivation.

[0040] First, the acquisition unit 110 acquires a plurality of aerial images and position information obtained by imaging the farmland F (step S100). Next, the generation unit 120 generates an orthomosaic image and a DSM model of the farmland F based on the acquired plurality of aerial images and position information (step S102). Next, the calculation unit 130 obtains the average value of the height differences of the DSM model, and generates a height difference map 150A by subtracting the average value from the elevation values at each position of the farmland F (step S104).

[0041] Next, the calculation unit 130 generates a histogram distribution 150B from the height difference map 150A, and determines the spraying amount and spraying position of the agricultural chemical based on the generated histogram distribution 150B (step S106). Next, the output unit 140 divides the farmland F into a predetermined mesh size, and outputs a spraying instruction map 150C that indicates the spraying amount for each mesh (step S108). Thereby, the processing of this flowchart ends.

[0042] In the above embodiment, as an example, the case where the farmland F is a paddy field for rice cultivation and the pest to be controlled is *Scirpophaga incertulas* has been described. However, the present invention is not limited to such a premise, and the farmland F may be a cultivated land other than paddy fields where waterlogging and pests and diseases are likely to occur, such as upland crops and soybeans. Further, in the above embodiment, as an example, it is assumed that the farmland F to be imaged is a flat ground, but the farmland F may be a sloping ground. When the farmland F is a sloping ground, the principle of the present invention can be applied as it is by calculating the relative height difference with the sloping ground as the reference plane.

[0043] According to the present embodiment described as above, a plurality of aerial images obtained by imaging the farmland from different positions above the farmland are acquired by a camera mounted on an unmanned aerial vehicle, and a height difference map representing the height difference of the surface of the farmland is generated from the plurality of aerial images. Based on the height difference map, the spraying position of the agricultural chemical in the farmland and the spraying amount corresponding to the spraying position are calculated. Thereby, it is possible to improve the working efficiency while suppressing the cost associated with spraying the agricultural chemical on the farmland.

[0044] [Second Embodiment] In the embodiment described above, the average value of the height difference of the DSM model is obtained, and the height difference map 150A is generated by subtracting the average value from the elevation value at each position in the field F. The position corresponding to a height difference that is lower than a predetermined value by a threshold value or more is determined as the pesticide spraying position. However, depending on the shape of the field F (for example, when the field F is on a slope), even in an area where the elevation value is higher than the average value, water may easily accumulate due to a step or the like with the surrounding area, and it may be necessary to spray pesticides.

[0045] That is, when the field F is on a slope, in the method of the embodiment described above, there may be a case where a depression area or the like where pesticides need to be sprayed cannot be detected. For example, in the cultivation fields of sweet potatoes and vegetables, drainage measures are considered, and there are many sloping fields F. If there is a depression area where water accumulates even on a slope, diseases such as basal rot and stem and root rot bacteria are likely to occur. Therefore, it is necessary to detect the depression area on the slope and spray pesticides. The second embodiment addresses such problems. Note that the method described below is particularly effective for the field F on a slope, but even when the field F is not on a slope, a depression area can be detected.

[0046] FIG. 9 is a diagram showing an example of an aerial image of the field F on a slope that is the target for detecting a depression area. In FIG. 9, the area R surrounded by the dotted line represents the depression area to be detected. FIG. 9 shows a field F on a slope that slopes downward from the upper left to the lower right.

[0047] Similar to the first embodiment, the aircraft 10 flies over the field F, images the field F from different positions in the air by the camera 12, and transmits it to the relay device 20 together with the position information of the aerial image. The relay device 20 displays the aerial image as an RGB image, or transmits the received aerial image and position information to the information processing device 100 via the network NW.

[0048] FIG. 10 is a diagram showing an example of a height difference map 150A of a farm field F on a sloped ground that is a target for detecting a depression area. FIG. 10 sets the average value to 0 m and classifies the obtained height differences into ranges of every 0.2 m. In the case of FIG. 10, the height differences in the height difference map 150A exceed 0 m at most positions, indicating an area where the elevation value is higher than the average value (that is, in this case, with the method of the above-described embodiment, there may be a case where a depression area cannot be detected).

[0049] Similar to the first embodiment, when the generation unit 120 receives an aerial image, based on the aerial image and the position information, using a known method, an orthomosaic image and a DSM model representing the farm field F are generated. Next, the generation unit 120 calculates the average value of the elevation values of each position on the surface indicated by the generated DSM model, and subtracts the average value from the elevation value of each position on the surface to obtain a height difference, thereby generating a height difference map 150A.

[0050] Next, the calculation unit 130 sequentially applies a sliding window with a predetermined width (for example, 3×3 pixels) to each position indicated by the height difference map 150A, subtracts the minimum value from the pixel values (height differences) of each position included in the sliding window, and determines a position where the subtracted value is equal to or less than a predetermined value (for example, 5 cm) as a depression area.

[0051] FIG. 11 is a diagram for explaining a method of detecting a depression area from the height difference map 150A using a sliding window. FIG. 11 shows an example in which the calculation unit 130 applies a 3×3 pixel sliding window to the height difference map 150A. In the case of FIG. 11, the calculation unit 130 first specifies 2 cm, which is the pixel value in the lower right, as the minimum value in the sliding window. Next, the calculation unit 130 subtracts 2 cm from the value of each pixel, and determines a position where the subtracted value is equal to or less than a predetermined value (for example, 5 cm) as a pixel constituting the depression area. As a result, in FIG. 11, among the sliding window, the pixels in the rightmost column are determined as the depression area.

[0052] Note that, although FIG. 11 shows an example in which the calculation unit 130 determines a depression area from the farm field F using the height difference map 150A, the present invention is not limited to such a configuration, and the calculation unit 130 may determine a depression area from the farm field F using the DSM model. In that case, the calculation unit 130 subtracts the minimum value included in the sliding window from the pixel value (elevation value) of each position of the farm field F indicated by the DSM model, and determines a position where the subtraction value is equal to or less than a predetermined value as the depression area.

[0053] FIG. 12 is a diagram showing an example of a depression area detected from the farm field F on a sloping ground. FIG. 12 shows an example in which a depression area is detected from the aerial image shown in FIG. 9 and the depression area is drawn superimposed on the image. As shown in FIG. 12, while the area R is detected as the depression area, it can be seen that positions on the farm field F that are not depressions are also partially detected as the depression area. Therefore, the calculation unit 130 may apply a method such as clustering to the pixel group determined as the depression area, and determine only a pixel group having an area equal to or larger than a predetermined value (that is, continuous) as the depression area. Thereby, noise misdetected as the depression area can be removed.

[0054] The calculation unit 130 may determine the same spraying amount for each position determined as the depression area, or may classify the height differences of each position determined as the depression area into a plurality of levels and determine the spraying amounts corresponding to the plurality of levels.

[0055] [Flow of processing] FIG. 13 is a flowchart showing an example of the flow of processing in which the calculation unit 130 according to the second embodiment detects a depression area. The processing of the flowchart shown in FIG. 13 is executed, for example, at the timing when the generation unit 120 generates the DSM model.

[0056] First, the calculation unit 130 applies a Gaussian filter or the like to the generated DSM model to remove noise from the DSM model (step S200). Next, the calculation unit 130 sequentially applies a sliding window to the DSM model to detect local lowest points (step S202). Next, the calculation unit 130 extracts pixels whose subtraction value obtained by subtracting the lowest point from the pixel value is equal to or less than a predetermined value as a pixel group representing a depression area (step S204). Next, the calculation unit 130 determines a continuous pixel group as a depression area from the extracted pixel group (step S206). Next, the calculation unit 130 determines the amount of pesticide to be sprayed on the determined depression area (step S208).

[0057] FIG. 14 is a flowchart showing an example of the processing flow executed by the information processing apparatus 100 according to the second embodiment. First, the acquisition unit 110 acquires a plurality of aerial images and position information obtained by imaging the farm field F (step S300). Next, the generation unit 120 generates an orthomosaic image and a DSM model of the farm field F based on the acquired plurality of aerial images and position information (step S302). Next, the calculation unit 130 determines continuous depression areas according to the flow of the flowchart shown in FIG. 13 (step S304).

[0058] Next, the calculation unit 130 determines the spraying amounts for the depression area and the other areas (step S306). More specifically, the calculation unit 130 may determine a fixed spraying amount for the depression area, and determine the spraying amount for the other areas by the method described with reference to FIGS. 5 and 6. Next, the output unit 140 divides the farm field F into a predetermined mesh size and outputs a spraying instruction map 150C indicating the spraying amount for each mesh (step S308). Thereby, the processing of this flowchart ends.

[0059] According to the second embodiment described above, even when the farm field F is located on a sloping ground and the altitude value is high, it is possible to detect the depression area where pesticides should be sprayed from the farm field F.

[0060] As described above, the embodiments for carrying out the present invention have been described using the embodiments. However, the present invention is not limited to such embodiments, and various modifications and substitutions can be made without departing from the gist of the present invention.

Explanation of Reference Numerals

[0061] 10 Aircraft 12 Camera 20 Relay Device 30 Terminal Device 100 Information Processing Device 110 Acquisition Unit 120 Generation Unit 130 Calculation Unit 140 Output Unit 150 Storage Device

Claims

1. An acquisition unit that acquires a plurality of aerial images obtained by imaging the farmland from different positions above the farmland using a camera mounted on an unmanned aircraft; A generation unit that generates a height difference map representing the height difference of the surface of the farmland from the plurality of aerial images; A calculation unit that calculates a spraying position for spraying agricultural chemicals and a spraying amount corresponding to the spraying position in the farmland based on the height difference map, and an information processing apparatus. Information processing apparatus.

2. The generation unit generates a DSM (digital surface model) model representing the surface of the farmland from the plurality of aerial images, calculates an average value of elevation values at each position on the surface indicated by the DSM model, and subtracts the average value from the elevation value at each position on the surface to obtain the height difference, thereby generating the height difference map. The information processing apparatus according to Claim 1.

3. The calculation unit determines, as the spraying position, a position corresponding to a height difference that is lower than a predetermined value or more based on a height difference corresponding to a threshold value among the height differences at each position indicated by the height difference map. The information processing apparatus according to Claim 1.

4. The threshold value is a height difference representing the water surface height of the farmland when water enters the farmland. The information processing apparatus according to Claim 3.

5. The calculation unit classifies the height differences at each position indicated by the height difference map into a plurality of levels, and calculates the spraying amounts corresponding to the plurality of levels. The information processing apparatus according to Claim 1.

6. The information processing apparatus further includes an output unit that outputs a spraying instruction map indicating the spraying position and the spraying amount in the farmland to the unmanned aircraft. The information processing apparatus according to Claim 1.

7. The output unit divides the farmland into meshes of a predetermined size, determines representative values of the spraying position and the spraying amount included in each divided mesh, and outputs the determined representative values of the spraying position and the spraying amount as the spraying instruction map. The information processing apparatus according to Claim 6.

8. The calculation unit extracts a depression area from the farmland based on the height difference of the surface of the farmland, and determines the extracted depression area as the spraying position. The information processing apparatus according to Claim 1.

9. The calculation unit sequentially applies a sliding window with a predetermined width to each position in the field, subtracts the minimum value from the pixel values of each position included in the sliding window, and determines a position where the subtracted value is equal to or less than a predetermined value as the depression area. The information processing apparatus according to claim 8.

10. The field is a paddy field for rice cultivation, The agricultural chemical is for controlling Semisulcospira libertina. The information processing apparatus according to any one of claims 1 to 9.

11. A computer, acquires a plurality of aerial images obtained by imaging the field from different positions above the field with a camera mounted on an unmanned aerial vehicle, generates a height difference map representing the height difference of the surface of the field from the plurality of aerial images, calculates a spraying position for spraying an agricultural chemical and a spraying amount corresponding to the spraying position in the field based on the height difference map. An information processing method.

12. A computer, is caused to acquire a plurality of aerial images obtained by imaging the field from different positions above the field with a camera mounted on an unmanned aerial vehicle, is caused to generate a height difference map representing the height difference of the surface of the field from the plurality of aerial images, is caused to calculate a spraying position for spraying an agricultural chemical and a spraying amount corresponding to the spraying position in the field based on the height difference map. A program.

Images