JPWO2020040063A1 - Field crop photography method and photography drone - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a drone (unmanned aerial vehicle) capable of appropriately photographing the root of a crop in a field, the side surface of a leaf, and the tip of an ear. SOLUTION: A camera pointed backward in the direction of travel of the drone is installed at the lower part of the drone, and a crop temporarily knocked down by the downdraft of the rotary blade of the drone is photographed. In addition, it is equipped with an altitude and speed shooting mode that enables appropriate shooting of the tip of the ear, and each mode can be switched. The direction of the camera and the flight speed / altitude of the aircraft may be adjustable so that the desired part can be photographed. In addition, the shooting mode may be automatically switched based on the curvature of the leaves of the fallen crop. [Selection diagram] Fig. 6

Description

The present invention relates to a crop photographing method for grasping a crop growth situation in a field by image analysis, and an unmanned air vehicle (drone) suitable for the photographing.

A method is known in which a field is photographed from the sky by a drone (unmanned air vehicle, multicopter) or the like and the photographed image is analyzed in order to grasp the growth situation of a crop (for example, Patent Document 1). By using a drone that flies over the field, it is possible to obtain more accurate image information than when using satellite images, etc., but there is still the problem that sufficient information cannot be obtained for accurate analysis. there were.

For example, insect pests of crops such as planthoppers often occur in the root portion of the plant, but it was difficult to properly photograph the root portion of the plant from above. Similarly, it was difficult to photograph lesions occurring on the root and weeds occurring on the water surface. In addition, especially for rice, if the image of the leaf shape when it receives wind can be obtained, the amount of silicon accumulated can be grasped, and the growth rate of rice can be estimated based on this to optimize the fertilizer plan. It is possible, but it was difficult to grasp it properly when shooting from above.

Japanese Patent Laid-Open Publication No. 2003-9664

(EN) A drone (unmanned aerial vehicle) that can appropriately capture the origin, tip, and side of leaves of a crop in a field.

The present invention is an unmanned aerial vehicle for field photography, which includes a camera, a rotary wing, and flight control means, and shows the crop stock exposed by the downdraft generated by the rotary wing causing the crop to fall. A first flight mode in which the flight control unit flies the unmanned aerial vehicle at an altitude and a speed that can be captured by the camera, and the crop exposed by the descending airflow generated by the rotor blades. The above problem is solved by providing an unmanned aerial vehicle having a second flight mode in which the flight control means flies the unmanned aerial vehicle at an altitude and speed at which the camera can capture the tip of the ear.

Further, the present invention solves the above problem by providing an unmanned aerial vehicle according to paragraph 0006, in which the altitude in the second flight mode is higher than the altitude in the first flight mode.

Also, in the present invention, the altitude in the second flight mode is the same as the altitude in the first flight mode, and the speed in the second flight mode is higher than the speed in the first flight mode. The above problem is solved by providing an unmanned aerial vehicle according to item 1.

Further, the present invention further comprises an exposure range measuring means for the crop due to the descending air flow, and the flight control means uses the first flight mode according to the exposure range of the crop measured by the exposure range measuring means. The above problem is solved by providing an unmanned aerial vehicle according to paragraph 0006, paragraph 0007, or paragraph 0008 for switching between the second flight mode and the second flight mode.

Further, the present invention solves the above problems by providing the unmanned air vehicle according to paragraph 0009, in which the exposure range measuring unit measures the exposure range of the crop by measuring the curvature of the leaves of the crop that has been laid down. To do.

Further, the present invention provides the unmanned air vehicle according to paragraph 0009, wherein the exposure range measuring means measures the exposure range of the crop by hue, lightness, or saturation of the image of the crop that has been fallen down. Solve the problem.

Further, the present invention further comprises a third flight mode, wherein the downdraft in the third flight mode is weaker than the downdraft in the second flight mode, and when the camera is not taking a picture, The above problem is solved by providing an unmanned air vehicle according to paragraph 0006, paragraph 0007, paragraph 0008, paragraph 0009, paragraph 0010, or paragraph 0011, in which flight control means automatically switches to the third flight mode. To solve.

Further, the present invention solves the above problem by providing an unmanned aerial vehicle according to paragraph 0012, in which the altitude in the third flight mode is higher than the altitude in the second flight mode.

Further, in the present invention, the altitude in the third flight mode is the same as the altitude in the second flight mode, and the speed in the third flight mode is higher than the speed in the second flight mode. The above problem is solved by providing an unmanned aerial vehicle according to item 1.

Further, the present invention is a method of photographing a plant origin, a side surface of a leaf, or a tip of a crop in a field, using an unmanned air vehicle for field photography, which includes a camera and a rotary wing, and uses the camera for the rotation. The problem is solved by providing a method that includes a step of photographing a plant base, a side surface of a leaf, or a tip of the crop exposed when the crop is laid down by a downdraft generated by a wing.

Further, the present invention provides the method according to paragraph 0015, further comprising the steps of measuring the exposure range of the crop and adjusting the altitude or speed of the airframe according to the measured exposure range. Solves the above problem.

Further, the present invention solves the above-mentioned problems by providing the method according to paragraph 0016, wherein the step of measuring the exposure range of the crop includes the step of measuring the curvature of the leaves of the crop that has fallen.

Further, the present invention provides the method according to paragraph 0016, wherein the step of measuring the exposure range of the crop includes the step of measuring the exposure range of the crop by the hue, lightness, or saturation of the image of the crop that has been collapsed. The above problems can be solved by providing them.

Further, the present invention is a program for causing an unmanned aerial vehicle for field photography equipped with a camera and a rotor to photograph a crop plant in the field, side surfaces of leaves, or ears, and the camera includes the rotor. The above problem is solved by providing a program that causes a computer to execute a group of instructions for photographing a plant origin, a side surface of a leaf, or a tip of a crop exposed when the crop is laid down by the generated downdraft.

Further, the present invention is a program according to paragraph 0019, which further causes a computer to execute an instruction group for measuring an exposure range of a crop and an instruction group for adjusting an altitude or a speed of an airframe according to the measured exposure range. The above problem is solved by providing

Further, the present invention solves the above problem by providing the program according to paragraph 0020, in which the command group for measuring the exposure range of the crop includes a command group for measuring the curvature of the leaves of the crop that has fallen.

In the present invention, the instruction group for measuring the exposure range of the crop is described in paragraph 0020, which includes a group of instructions for measuring the crop exposure range according to hue, lightness, or saturation of an image of a crop that has been knocked down. The above problems are solved by providing a program.

Provided is a drone (unmanned aerial vehicle) capable of appropriately photographing the plant origin, the tip of a crop, and the side surface of leaves in a field.

It is a top view of the Example of the drone for field photography which concerns on this invention. It is a front view of the Example of the drone for field photography which concerns on this invention. It is a right side view, a rear view, and a perspective view of an example of a field photography drone concerning the present invention. 1 is an example of an overall conceptual view of a field photography system using an embodiment of a field photography drone according to the present invention. It is a schematic diagram showing the control function of the Example of the drone for field photography which concerns on this invention. It is a figure which shows the basic idea of the photography mode by the Example of the drone for farmland photography which concerns on this invention. It is a figure which shows the basic idea of the altitude control in the plant root partial photography by the Example of the drone for farmland photography which concerns on this invention. It is a figure which shows the basic way of thinking of control at the time of a direction change of the Example of the drone for field photography which concerns on this invention.

Hereinafter, modes for carrying out the present invention will be described with reference to the drawings. The figures are all examples.

FIG. 1 is a plan view of an embodiment of a drone (100) according to the present invention, FIG. 2 is a bottom view thereof, FIG. 3-a is a front view thereof (as seen from the traveling direction side), and FIG. A right side view and a perspective view are shown in FIG. In the specification of the present application, the drone means any power means (electric power, prime mover, etc.) and control method (whether wireless or wired, autonomous flight type or manual control type, etc.). Instead, it refers to an entire flying body having a plurality of rotors or flight means.

Rotors (101-1a, 101-1b, 101-2a, 101-2b, 101-3a, 101-3b, 101-4a, 101-4b) (also called rotors) are used to fly the drone (100). Considering the balance of flight stability, airframe size, and battery consumption, it is desirable to have 8 aircraft (4 sets of two-stage rotary blades).

The motors (102-1a, 102-1b, 102-2a, 102-2b, 102-3a, 102-3b, 102-4a, 102-4b) are rotary blades (101-1a, 101-1b, 101-2a). , 101-2b, 101-3a, 101-3b, 101-4a, 101-4b) is a means for rotating (typically an electric motor, but may be an engine, etc.) It is desirable to have one. The upper and lower rotor blades (eg 101-1a and 101-1b) and their corresponding motors (eg 102-1a and 102-1b) in one set are for drone flight stability etc. It is desirable that the axes be collinear and rotate in opposite directions.

The drug nozzles (103-1, 103-2, 103-3, 103-4) are means for spraying the drug downward, and are preferably provided with four machines. The chemical|medical agent tank (104) is a tank for storing the chemical|medical agent sprayed, and it is desirable to be provided in the position close|similar to the gravity center of a drone (100), and a position lower than a gravity center from a viewpoint of a weight balance. The drug hose (105-1, 105-2, 105-3, 105-4) connects the drug tank (104) with each drug nozzle (103-1, 103-2, 103-3, 103-4). It may be made of a hard material and may also serve to support the medicine nozzle. The pump (106) is a means for discharging the medicine from the nozzle.

The farm field photographing camera (107) is a camera for photographing the farm field (farmland) from the drone (100), and is preferably installed in a direction for photographing the farm field behind the traveling direction of the machine body. Further, it is desirable that the angle between the field photographing camera (107) and the machine body be variable by means of a stepping motor or the like.

FIG. 4 shows an overall conceptual diagram of an embodiment of a field photography system using the drone (100) according to the present invention. This figure is a schematic diagram and the scale is not accurate. The pilot (401) sends a command to the drone (100) by the operation of the user (402), and also receives information (for example, position, drug amount, remaining battery level, field photography camera) received from the drone (100). It is a means for displaying images and the like, and may be realized by a portable information device such as a general tablet terminal that runs a computer program. It is desirable that the drone (100) according to the present invention be controlled to perform autonomous flight, but it is desirable that it can be manually operated during basic operations such as takeoff and return, and in an emergency. It is desirable that the pilot (401) and the drone (100) perform wireless communication by Wi-Fi or the like.

The farm field (403) is a rice field, a field or the like to be imaged by the drone (100). In reality, the topography of the farm field (403) is complicated, and there are cases where the topographic map cannot be obtained in advance, or the topographic map and the situation at the site are inconsistent. Normally, the field (403) is adjacent to a house, a hospital, a school, a field for other crops, a road, a railroad, or the like. In addition, there may be obstacles such as buildings and electric wires in the field (403).

The base station (404) is a device that provides a master device function of Wi-Fi communication, etc. It is desirable that it also functions as an RTK-GPS base station and can provide an accurate position of the drone (100). (Wi-Fi communication master function and RTK-GPS base station may be independent devices). The farm cloud (405) is typically a group of computers operated on a cloud service and related software, and it is desirable that the farm cloud (405) be wirelessly connected to the control unit (401) by a mobile phone line or the like. The farming cloud (405) may analyze the image of the field (403) captured by the drone (100), grasp the growing condition of the crop, and perform a process for determining a flight route. Further, the drone (100) may be provided with the stored topographical information of the field (403). In addition, the history of the drone (100) flight and captured images may be accumulated and various analysis processes may be performed.

Usually, the drone (100) takes off from the landing point (406) outside the field (403) and sprays the drug on the field (403), or when it becomes necessary to replenish or recharge the drug. Return to (406). The flight route (entry route) from the landing point (406) to the target field (403) may be stored in advance in the farm cloud (405), etc., or before the user (402) starts taking off. You may enter in.

FIG. 5 is a schematic view showing the control function of the embodiment of the drug spraying drone according to the present invention. The flight controller (501) is a component that controls the entire drone, and specifically may be an embedded computer that includes a CPU, memory, related software, and the like. The flight controller (501) uses a control means such as an ESC (Electronic Speed Control) to control the motor (102-) based on the input information received from the flight controller (401) and the input information obtained from various sensors described later. 1a, 102-1b, 102-2a, 102-2b, 102-3a, 102-3b, 104-a, 104-b) to control the flight of the drone (100). The actual rotation speed of the motor (102-1a, 102-1b, 102-2a, 102-2b, 102-3a, 102-3b, 104-a, 104-b) is fed back to the flight controller (501) and is normal. It is configured so that it can be monitored whether it is rotating normally. Alternatively, the rotary blade (101) may be provided with an optical sensor or the like so that the rotation of the rotary blade (101) is fed back to the flight controller (501).

The software used by the flight controller (501) is preferably rewritable through a storage medium or the like for function expansion/change, problem correction, or the like, or through communication means such as Wi-Fi communication or USB. In addition, a part of the calculation process used by the flight controller (501) for control may be executed by another computer existing on the flight controller (401), the farming support cloud (405), or elsewhere. Good. Since the flight controller (501) is highly important, some or all of its constituent elements may be duplicated.

The battery (502) is a means for supplying power to the flight controller (501) and other components of the drone, and is, for example, rechargeable. The battery (502) is connected to the flight controller (501) via a fuse or a power supply unit including a circuit breaker and the like. The battery (502) may be a smart battery having a function of transmitting its internal state (amount of stored electricity, accumulated usage time, etc.) to the flight controller (501) in addition to the power supply function.

The flight controller (501) communicates with the control device (401) via the Wi-Fi slave device function (503) and further via the base station (404), and sends necessary commands from the control device (401). It is possible to receive and send necessary information to the pilot (401). The base station (404) may have the function of the RTK-GPS base station in addition to the communication function by Wi-Fi. By combining the signal from the RTK base station and the signal from the GPS positioning satellite, the GPS module (504) can measure the absolute position of the drone (100) with an accuracy of a few centimeters. Since the GPS module (504) is of high importance, it is duplicated and multiplexed, and each redundant GPS module (504) uses a different satellite to cope with the failure of a specific GPS satellite. Controlled.

The 6-axis gyro sensor (505) is means for measuring accelerations of the drone body in three directions orthogonal to each other (further, means for calculating velocity by integration of accelerations). The 6-axis gyro sensor (505) is a means for measuring the change in the attitude angle of the drone aircraft in the three directions described above, that is, the angular velocity. The geomagnetic sensor (506) is a means for measuring the direction of the drone body by measuring the geomagnetic field. The barometric pressure sensor (507) is a means for measuring the barometric pressure, and can indirectly measure the altitude of the drone. The laser sensor (508) is a means for measuring the distance between the drone body and the ground surface by utilizing the reflection of laser light, and for example, an IR (infrared) laser may be used. The sonar (509) is a means for measuring the distance between the drone body and the ground surface by utilizing the reflection of sound waves such as ultrasonic waves. These sensors may be selected depending on the drone's cost goals and performance requirements. Further, a gyro sensor (angular velocity sensor) for measuring the tilt of the machine body, a wind force sensor for measuring wind force, and the like may be added. Further, it is desirable that these sensors are duplicated or multiplexed.

The flow rate sensor (510) is a means for measuring the flow rate of the medicine, and is provided at a plurality of places in the path from the medicine tank (104) to the medicine nozzle (103). The liquid shortage sensor (511) is a sensor that detects that the amount of the medicine has become equal to or less than a predetermined amount. The field shooting camera (107) is a means for shooting the field (403) and acquiring data for image analysis, and is preferably a multi-spectral camera. The obstacle detection camera (513) is a camera for detecting an obstacle of the drone. Since the image characteristics and the direction of the lens are different from those of the field photography camera (107), it is a device different from the field photography camera (106). Is desirable. The switch (514) is a means for the user (402) of the drone (100) to make various settings. The obstacle contact sensor (515) is used to detect that the drone (100), in particular its rotor and propeller guards, has come into contact with obstacles such as electric wires, buildings, human bodies, trees, birds or other drones. Sensor of. The cover sensor (516) is a sensor that detects that the operation panel of the drone (100) and the cover for internal maintenance are open. The drug injection port sensor (517) is a sensor that detects that the injection port of the drug tank (104) is open. These sensors may be selected according to the drone's cost targets and performance requirements, and may be duplicated or multiplexed.

The flight controller (501) sends a control signal to the pump (106) to adjust the medicine ejection amount and stop the medicine ejection. It is desirable that the current status (for example, the number of revolutions) of the pump (106) be fed back to the flight controller (501).

The LED (517) is a display means for informing the drone operator of the status of the drone. Display means such as a liquid crystal display may be used instead of or in addition to the LEDs. The buzzer (518) is an output means for notifying a drone state (especially an error state) by an audio signal. The Wi-Fi slave device function (519) is an optional component for communicating with an external computer or the like for the transfer of software, for example, separately from the control device (401). In addition to or in addition to the Wi-Fi cordless handset function, other wireless communication means such as infrared communication, Bluetooth (registered trademark), ZigBee (registered trademark), NFC, or wired communication means such as USB connection May be used. The speaker (520) is an output means for notifying the drone state (particularly an error state) by the recorded human voice or synthetic voice. Depending on the weather conditions, it may be difficult to see the visual display of the drone (100) in flight. In such a case, it is effective to communicate the situation by voice. The warning light (521) is a display means such as a strobe light for indicating the state of the drone (in particular, an error state). These input/output means may be selected according to the cost target and performance requirements of the drone, or may be duplicated/multiplexed.

The drone (100) according to the present invention may be provided with a shoot-source shooting mode suitable for shooting a shoot-base part of a field crop, a tip-shooting mode suitable for shooting a shoot-head part of a crop in the field, or both. desirable. Shooting the root part is effective for measuring the number of sprouts (heading number) per stock, estimating the total yield per stock, and for pathological diagnosis and growth diagnosis near the root. Is effective for yield per unit number of ears and pathological diagnosis near leaf tips, etc., and can significantly improve the value of image analysis compared to the case of simply photographing the entire crop from the sky. it can.

Especially in the case of rice, it is desirable to perform both shoot-source and tip-shooting from the heading to the end of tillering. After the height of the spikes is sufficiently long, the shoots should be photographed frequently and the shoots should be photographed regularly to diagnose pathological growth. It is necessary to mainly shoot the tips when the harvest is approaching. Therefore, it is desirable for the drone (100) according to the present invention to be able to switch between the stock shoot mode and the tip shoot mode by a user's command or automatically.

FIG. 6 shows the basic idea of the shooting mode according to the embodiment of the field shooting drone according to the present invention (this drawing is a conceptual diagram, and the scale is not accurate). Generally, in a drone (100), the air flow (601) by the rotor blades flows backward in the traveling direction of the aircraft. The airflow from the rotor blades has the effect of temporarily pulling down the crops in the field. The crop that is most affected by the air flow created by the drone's rotor blades is behind the drone's direction of travel, and its depression angle is about 60 degrees. It is desirable to install the camera (107) at a position where the image can be taken in this direction. On the other hand, if the camera is designed to be able to shoot only in the direction directly below the drone, the drone moves at a very low speed to shoot the roots and sides of the leaves, or procedures such as shooting in the hovering state. Becomes necessary, and it becomes impossible to perform efficient shooting.

FIG. 6-a shows the stock shooting mode. Behind the direction of the drone (100), select the plant base of the crop (602) caused by the downdraft and the area (604) where the side of the leaf is exposed to the sky with the field photography camera (107). It is possible to acquire the image of the plant base of the crop and the side surface of the leaf by capturing the image, or by extracting a region (604) obtained by image processing by capturing a region wider than this region. The image of the region (604) in which the side surface of the leaf is exposed to the sky has a great difference in lightness and saturation as compared with the other regions, and therefore extraction by image processing is easy. In addition, the thickness and hardness of the leaves can be estimated from the curved shape of the leaves of the crop when exposed to the wind.

FIG. 6-b shows the tip shooting mode. The basic idea is the same as in the stock shooting mode, but the altitude of the drone (100) is raised more than in the stock shooting mode to reduce the influence of the downdraft (601) toward the crop (602). With this, it is possible to create an area (605) in which the tip portion of the crop (602) is exposed to the sky, and it is possible to effectively acquire an image of only the tip portion as in the case of the stock shooting mode. ..

FIG. 6-c shows the normal movement mode. Since the normal movement mode is higher than the tip shooting mode, the effect of defeating the crop (602) is small. If you do not shoot the roots or tips, it is advisable to fly the drone (100) in normal transfer mode to minimize lodging and damage to the rice ears due to downdraft. Also, it is desirable that the movement when suspending/restarting the flight due to replacement of the dead battery during shooting is similarly set to the normal movement mode. In this case, it is desirable to automatically switch to the normal movement mode and return to the stock-source shooting mode or the tip shooting mode by a program installed in the flight controller (501).

It is desirable to switch the modes shown in FIG. 6 by controlling the altitude of the drone (100). The altitude in the stock shooting mode is approximately 0.9 m from the ground or 0.1 m from the tip, while the altitude in the tip shooting mode is 1.1 m to 2.0 m from the ground, or 0.3 m to 1.2 m from the tip, normal movement mode Experiments by the inventor have shown that it is desirable to control the drone (100) so that the altitude at time is 2 meters to 2.5 meters, or 1.2 meters to 1.7 meters from the tips. It is desirable that this altitude control be automatically performed by a program installed in the flight controller (501).

Instead of adjusting the altitude of the drone (100), or in accordance with the adjustment of the altitude of the drone (100), the shooting mode may be switched by changing the flight speed. Since the drone (100) flies by tilting the aircraft toward the direction of travel, lowering the flight speed will cause the downdraft to go directly below the aircraft, increasing the effect of defeating the crop. Therefore, the flight speed in the tip shooting mode may be faster than that in the stock shooting mode, and the flight speed in the normal flight mode may be faster than that in the tip shooting mode. For example, if the altitude of the drone (100) is fixed at 2 meters above ground, the speed will be about 7 kilometers per hour in the stock shooting mode, about 20 kilometers per hour in the tip shooting mode, and about 27 kilometers per hour in normal flight mode. Experiments by the inventor have revealed that such control is desirable.

The altitude and speed values in each mode above are examples, and it is desirable that they can be adjusted according to changes in the crop type and the technology adopted by the drone (100). The target altitude and speed of each shooting mode are set in advance, and the user can simply control the shooting mode to be automatically controlled by the program installed in the flight controller (501). Is desirable.

The direction of the camera that can effectively shoot the stock changes depending on the moving speed, so the speed sensor of the drone (100) can be used to change the shooting direction of the field shooting camera (107) with a stepping motor etc. to match the moving speed. It may be adjustable. In addition, the relative position of the crop that is temporarily cut down to the drone aircraft is affected by the wind force and wind direction. Therefore, a wind force/wind direction sensor is installed on the drone (100) to automatically set the shooting direction of the field shooting camera (107). It may be adjustable. Also, the image captured by the field shooting camera (107) is displayed in real time on the remote control device of the drone (100), and the operator can finely adjust the direction of the field shooting camera (103) while checking the image. Good. The field of the field where the plant base and the tip of the ear are exposed to the sky can be distinguished by image recognition due to the difference in brightness and color tone from other areas, so that the area should be near the center of the shooting range. In addition, the field shooting camera (107) may be automatically adjusted. When the drone (100) is hovering or is flying at a low speed (for example, about 3 meters or less per second), control may be performed such that the field shooting camera (107) does not shoot.

The target altitude of the drone (100) in each mode may be controlled in real time based on an image of a crop of a field photographed by the field photographing camera (107) instead of a preset fixed value. By measuring the curvature (1/r) of the stalk (701) of the crop that has been cropped down as shown in Fig. 7-a by image analysis, it can be estimated whether the crop has been collapsed enough to shoot the plant. Good. If the curvature obtained by image analysis is not the desired curvature obtained by previous experiments, etc., the drone (100 ) Is desirable.

As shown in FIG. 9-b, in the image (902) of the entire field, the areas of the region (903) where the stem of the crop is exposed and the region (904) where the tip of the crop are exposed are obtained by a preliminary experiment or the like. If the desired range is not reached, the drone (100) altitude, speed, or both may be controlled so as to be within the desired range. Note that the area (903) and the area (904) are significantly different in hue, lightness, or saturation compared with the image (902) of the entire field, so that the range can be specified in real time by image analysis. It is possible.

By analyzing the image data of the stock origin photographed by the field photographing camera (106), various information which could not be obtained in the past can be obtained. For example, the presence of chlorophyll can be detected by analysis of near-infrared images (eg, around 780 nm wavelength), which allows extraction of only crop parts from the image. By applying edge detection to the image of the crop part and extracting the contour line of the leaf, it is possible to measure the degree of bending of the leaf when it receives wind. This makes it possible to estimate the leaf thickness and the growth status of the crop. In particular, when the crop is rice, it is possible to grasp the amount of silicon accumulated (because the amount of silicon increases the hardness of the leaves). Further, it can be presumed that the portion where the straight line portion (detected by the edge detection processing) is dense from the water surface area, which is the portion without the crop portion, which is found by the analysis of the near infrared image, is the stock origin. Edge detection is performed on the part where the amount of near-infrared rays is lower than that of the plant, and when a spotted area is seen, it can be estimated that the planter is attached. It can be presumed that the bacterial wilt disease is occurring when there is a deep red area in the stock. Moreover, since the plants are usually planted at equal intervals of 20 to 30 cm, it can be estimated that weeds are occurring when the water surface regions do not appear at equal intervals on the image. In addition to these conventional image analysis methods, a large number of image data examples can be used as input to a neural network (preferably deep neural network) for machine learning to enable efficient and highly accurate analysis. Has been clarified by the inventor's experiment.

Similarly, with regard to the image of the tip, by performing the tip shape recognition using deep learning and the NDVI analysis using the multispectral camera in real time in flight, it is possible to achieve much better results compared to conventional mere shooting from the sky. Useful information can be obtained.

FIG. 8 shows an example of the operation of the drone for field photography according to the present invention when changing the direction (this figure is a conceptual diagram and the scale is not accurate). The drone (800) in FIG. 8 is a simplified schematic representation of the drone shown in FIGS. 1, 2 and 3 so that the orientation of the aircraft can be easily understood. Normally, when the drone changes its flight direction, as shown in FIG. 8-A, control is performed such that only the moving direction is changed without changing the absolute direction of the airframe. However, in such a direction changing method, in order to allow the field shooting camera (107) to always shoot the rear side of the traveling direction of the drone (800), a plurality of fields (typically four) are used. It is necessary to provide (107) and switch appropriately for shooting, or to change the position of the field shooting camera (107) according to the traveling direction of the machine by means of a stepping motor or the like. It is not preferable because of increase in cost, cost, and control complexity.

As an alternative method, it is desirable to change the absolute orientation of the aircraft itself when changing directions, as shown in FIG. By adopting such a method, it is possible to always photograph the rear side in the traveling direction of the drone (800) simply by installing one field photographing camera (107) on the drone (800). Although the depression angle may need to be adjusted as described above, it is advantageous in terms of cost, weight, and accuracy as compared with the method of providing a plurality of cameras or the method of changing the camera directions themselves. Is.

(Technically remarkable effect of the present invention)
With the drone according to the present invention, it is possible to efficiently acquire the image of the plant origin, the leaf side surface, and the tip of the entire crop in the field. Analyzing the obtained results can lead to effective and efficient pest control and fertilization planning. In addition, when the crop is rice, it is possible to optimize the fertilizer plan by grasping the amount of silicon accumulated from the image of the shape of the leaves when exposed to wind, estimating the growth rate of rice based on that It will be possible.

Claims (17)

An unmanned aerial vehicle for field photography, comprising a camera, a rotary wing, and flight control means,
A first flight in which the flight control means causes the unmanned air vehicle to fly at an altitude and a speed at which the camera can photograph the plant origin of the crop exposed by the downdraft of the crop caused by the descending airflow generated by the rotary wing. Mode,
A second flight mode in which the flight control means causes the unmanned aerial vehicle to fly at an altitude and speed at which the camera can shoot the tip of the crop exposed by the descending airflow generated by the rotor blades causing the crop to fall. An unmanned aerial vehicle equipped with. The unmanned air vehicle according to claim 1, wherein the altitude in the second flight mode is higher than the altitude in the first flight mode. The altitude in the second flight mode is the same as the altitude in the first flight mode,
The speed in the second flight mode is faster than the speed in the first flight mode,
The unmanned aerial vehicle according to claim 1. Furthermore, the exposure range measurement means by the downdraft of the crop is provided,
The flight control means performs switching between the first flight mode and the second flight mode according to the exposure range of the crop measured by the exposure range measuring means. The unmanned aerial vehicle according to claim 3. The unmanned air vehicle according to claim 4, wherein the exposure range measuring means measures the exposure range of the crop by measuring the curvature of the leaves of the crop that has fallen over. The unmanned air vehicle according to claim 4, wherein the exposure range measuring unit measures the exposure range of the crop based on the hue, lightness, or saturation of the image of the crop that has been fallen down. In addition, with a third flight mode,
The downdraft in the third flight mode is weaker than the downdraft in the second flight mode,
The flight control means automatically switches to the third flight mode when the camera is not shooting, claim 1, claim 2, claim 3, claim 4, claim 5, or Alternatively, the unmanned aerial vehicle according to claim 6. The unmanned air vehicle according to claim 7, wherein the altitude in the third flight mode is higher than the altitude in the second flight mode. The altitude in the third flight mode is the same as the altitude in the second flight mode,
The speed in the third flight mode is faster than the speed in the second flight mode,
The unmanned aerial vehicle according to claim 7. A method for photographing a plant origin, a side surface of a leaf, or a tip of a field crop,
Using an unmanned aerial vehicle for field photography equipped with a camera and a rotary wing,
A method comprising: causing the camera to photograph a plant origin, a leaf side surface, or a tip of the crop exposed by the descending airflow generated by the rotor blades causing the crop to fall. Further, measuring the exposure range of the crop,
Adjusting the altitude or speed of the airframe in response to the measured exposure range. 13. The method of claim 11, wherein measuring the exposed area of the crop comprises measuring the curvature of the leaves of a fallen crop. 12. The method of claim 11, wherein measuring the crop exposure range comprises measuring the crop exposure range by hue, lightness, or saturation of an image of the crop that has been collapsed. A program for making an unmanned aerial vehicle for field shooting equipped with a camera and a rotary wing image a plant origin of a field, a side surface of a leaf, or an ear,
A program that causes a computer to execute a group of instructions for causing the camera to photograph a plant origin, a leaf side surface, or a tip of the crop exposed by the descending airflow generated by the rotary wing causing the crop to fall. In addition, a group of commands to measure the exposure range of the crop,
The program according to claim 14, which causes a computer to execute a group of instructions for adjusting an altitude or a speed of an aircraft according to the measured exposure range. The program according to claim 15, wherein the instruction group for measuring the exposure range of the crop includes a group of instructions for measuring the curvature of the leaves of the crop that has fallen. 16. The program according to claim 15, wherein the command group for measuring the exposure range of the crop includes a group of commands for measuring the crop exposure range according to the hue, lightness, or saturation of the image of the crop that has been collapsed.

Images