KR20230067326A - Detecting system for forest disease using unmanned aerial vehicle - Google Patents

Abstract

The present invention relates to a system for detecting forest diseases using an unmanned aerial vehicle. According to one aspect of the present invention, a system for detecting forest diseases using an unmanned aerial vehicle. The unmanned aerial vehicle includes a collection part that collects image data about a forest; and a suspicious tree detection part that analyzes the collected image data to detect suspicious trees.

Description

Forest pest detection system using unmanned aerial vehicle {DETECTING SYSTEM FOR FOREST DISEASE USING UNMANNED AERIAL VEHICLE}

The present invention relates to a forest pest detection system using an unmanned aerial vehicle.

Recently, forest diseases and pests have rapidly increased due to the effects of climate change such as high temperatures and droughts caused by global warming. Pine wilt nematode disease caused by the pine wilt nematode, one of these forest pests, is a disease in which the pine nematode parasitic on the body of Monochamus alternatus invades the pine tree when it gnaws on the leaves of the pine tree, and the pine tree dies. The number and activity of Monochamus alternatus, an insect vector, is rapidly increasing, and the damage is spreading.

Accordingly, the Korea Forest Service is introducing a surveillance and control system using unmanned aerial vehicles to prevent the spread of pinewood nematode disease. However, the flight of the unmanned aerial vehicle for surveillance of pinewood nematode disease is mainly performed manually within the operator's visual range, and the captured image or video is transmitted to a server separately provided outside and analyzed. The disadvantage is that real-time detection is not possible.

Korean Registered Patent No. 10-2098259 (registered on 2020.04.01)

According to one embodiment of the present invention, it is intended to provide a forest disease and pest detection system using an unmanned aerial vehicle capable of detecting suspected infection in real time from an unmanned aerial vehicle.

In addition, it is intended to provide a forest pest detection system using an unmanned aerial vehicle that can automatically control the flight of the unmanned aerial vehicle according to the detection of a suspicious item in the unmanned aerial vehicle.

According to one aspect of the present invention, in a system for detecting forest diseases and pests using an unmanned aerial vehicle, the unmanned aerial vehicle includes: a collection unit for collecting image data of a forest; and a suspicious tree detection unit for detecting a suspicious tree by analyzing the collected image data. A system for detecting forest pests using an unmanned aerial vehicle may be provided.

The collection unit collects first image data of a forest at a first altitude, the suspicious tree detector analyzes the first image data to detect a suspicious tree, and the collection unit detects a suspicious tree by the suspicious tree detector. When detected, a forest pest detection system using an unmanned aerial vehicle that collects second image data for the suspicious tree detected at a second altitude may be provided.

In addition, a forest pest detection system using an unmanned aerial vehicle may be provided where the second altitude is an altitude lower than the first altitude.

In addition, the first altitude is an altitude of a predetermined height from the ground, and the second altitude is an altitude of a height spaced apart from the top of the suspect tree by a predetermined vertical distance A forest pest detection system using an unmanned aerial vehicle can be provided there is.

In addition, a forest pest detection system using an unmanned aerial vehicle may be provided in which the collection unit additionally collects third image data for the suspicious tree detected at a third altitude when a suspicious tree is detected by the suspicious tree detector.

In addition, a forest pest detection system using an unmanned aerial vehicle may be provided where the third altitude is an altitude lower than the second altitude.

In addition, the third altitude is an altitude lower than the top of the suspicious tree, and a forest pest detection system using an unmanned aerial vehicle may be provided.

In addition, a forest pest detection system using an unmanned aerial vehicle may be provided, wherein the unmanned aerial vehicle further includes an FC controller for controlling the position of the unmanned aerial vehicle.

In addition, a system for detecting forest diseases and pests using an unmanned aerial vehicle may further include a server for machine learning a suspicious item detection algorithm by the suspicious item detector.

The system for detecting forest pests using an unmanned aerial vehicle according to an embodiment of the present invention has the advantage of being able to detect suspected infection in real time from the unmanned aerial vehicle.

In addition, there is an advantage in that the flight of the aircraft can be automatically controlled according to the detection of a suspected infection.

1 is a diagram showing the configuration of a forest pest detection system using an unmanned aerial vehicle according to an embodiment of the present invention.
FIG. 2 is a diagram schematically showing sub-configurations of the unmanned aerial vehicle of FIG. 1 .
FIG. 3 is a diagram schematically illustrating sub-configurations of the server of FIG. 1 .

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the drawings.

In addition, in describing the present invention, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description will be omitted.

The forest pest detection system 1 using an unmanned aerial vehicle according to an embodiment of the present invention can be understood as a system for detecting trees infected with forest pests by collecting image data of a forest and analyzing the collected data. . In this embodiment, image data on pine trees is collected through the forest pest detection system 1 using an unmanned aerial vehicle, and the forest pest detected by the forest pest detection system 1 using an unmanned aerial vehicle is an example of a pine wilt nematode. However, the spirit of the present invention is not limited thereto.

Hereinafter, pine trees infected with forest pests, including pine wilt nematodes, are collectively referred to as "infected trees", and pine trees that can be detected as infected trees are referred to as "suspect trees".

1 is a diagram showing the configuration of a forest pest detection system using an unmanned aerial vehicle according to an embodiment of the present invention.

Referring to FIG. 1, the forest pest detection system 1 using an unmanned aerial vehicle according to an embodiment of the present invention is an unmanned aerial vehicle that detects suspicious and infected trees by photographing a suspected area of pine wilt nematode and analyzing the photographed image. (10), a server 20 for machine learning programs stored in the unmanned aerial vehicle 10 to detect suspicious items and infected items, and a network enabling mutual communication between the unmanned aerial vehicle 10 and the server 20 (30) may be included.

In the forest pest detection system 1 using an unmanned aerial vehicle according to an embodiment of the present invention, images are collected and analyzed by the unmanned aerial vehicle 10, and the operation of the unmanned aerial vehicle 10 is autonomously controlled according to the analyzed result. There is an advantage to being

Components constituting the forest disease and pest detection system 1 using an unmanned aerial vehicle may be connected through a network 30 to exchange data with each other or may be provided to transmit data from one side to the other. Here, the network 30 includes both a wired network and a wireless network, and is not limited to the type as long as it can transmit data.

The unmanned aerial vehicle 10 autonomously flies, collects image data obtained by photographing areas suspected of pine wilt nematodes, and analyzes the collected image data to detect suspicious and infected trees. Here, the unmanned aerial vehicle 10 is a vehicle manufactured to perform a designated mission without a pilot on board, and can autonomously fly by being supported by aerodynamic force.

Hereinafter, the configuration of the unmanned aerial vehicle 10 of the forest pest detection system 1 using the unmanned aerial vehicle will be described with reference to FIG. 2 .

FIG. 2 is a diagram schematically showing sub-configurations of the unmanned aerial vehicle of FIG. 1 .

Referring to FIG. 2 , the unmanned aerial vehicle 10 may include a collection unit 110, a memory 120, a processor 130, and a communication unit 140. However, not all of the components shown in FIG. 2 are essential components of the unmanned aerial vehicle 10, the unmanned aerial vehicle 10 may be implemented with more components, and the unmanned aerial vehicle 10 may be implemented with fewer components ( 10) may be implemented.

The collection unit 110 of the unmanned aerial vehicle 10 is a means for collecting data, and the camera 111 collects image data on pine trees in the area suspected of pine wilt nematode, and the change in angular velocity of the unmanned aerial vehicle 10 is sensed. It may include a gyro sensor 112 that measures the altitude of the unmanned aerial vehicle 10 and a height sensor 113 that measures the altitude.

The camera 111 may collect image data by capturing images of pine trees. In this case, the camera 111 may be an infrared camera, but the type of camera 111 is not limited thereto. Collected image data may be transmitted to the processor 130 and analyzed by the processor 130 .

Under the control of the processor 130, the camera 111 may collect primary image data at a first altitude, collect secondary image data at a second altitude, and collect a third image at a third altitude. data can be collected. In this case, the first altitude may be an altitude of a predetermined height from the ground, and the second altitude may be an altitude of a height lower than the first altitude but spaced apart from the top of the suspicious tree by a predetermined vertical distance. Also, the third altitude may be an altitude lower than the second altitude but lower than the top of the suspicious tree. In this case, the first image data and the second image data may be orthographic images photographed in a horizontal direction with respect to the longitudinal direction of the pine tree, and the third image data may be a side image of the suspect tree photographed in a lateral direction. That is, the first image data and the second image data may be images of a pine tree including a suspect tree photographed from above, that is, the first image data is at a first altitude higher than the second altitude at which the second image data is collected. As the altitude is collected, it may include more pine trees than the second image data.

The first image data may be used to detect a suspicious tree among the pine trees included in the data, and the second image data and the third image data may be used to detect an infected tree among the detected suspicious trees.

The gyro sensor 112 may sense yaw, pitch, and roll values of the unmanned aerial vehicle 10, and the height sensor 113 may measure the height of the unmanned aerial vehicle 10 from the ground and The relative height from the highest point of a specific forest to the unmanned aerial vehicle 10 may be sensed.

Sensor data sensed and collected from the gyro sensor 112 and the height sensor 113 may be transferred to the processor 130 and used for controlling a Flight Controller (FC) of the unmanned aerial vehicle 10 .

The memory 120 is a computer-readable recording medium, and includes a volatile storage device such as RAM (Random Access Memory) and a non-volatile mass storage device such as ROM (Rean Only Memory) and a disk drive. can do. For example, the memory 120 may be a flash memory type, a hard disk type, a multimedia card micro type, or a card type memory (eg SD or XD memory). etc.), RAM (RAM, Random Access Memory), SRAM (Static Random Access Memory), ROM (ROM, Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), It may include at least one type of storage medium among a magnetic memory, a magnetic disk, and an optical disk.

At least one program code capable of processing data or calculating an algorithm for detecting pine wilt nematodes according to the present embodiment may be stored in the memory 120 based on the data. That is, the memory 120 may store programs for processing and control of the processor 130 . The processor 130 may load a program code stored in the memory 120 to implement a preset algorithm. In this embodiment, it is described as an example that such a program code is transmitted from the server 20 and stored in the memory 120, but the spirit of the present invention is not limited thereto. As another example, the program code may be loaded from a recording medium (eg, DVD, memory card, etc.) readable by a separate computer.

The memory 120 may be physically provided in plurality or may be provided as a physical device integrated with the processor 130 . In addition, the memory 120 is provided as a physically or logically independent device, and may constitute a single device as a whole by being connected to a device in which the processor 130 is mounted.

A database 121 capable of storing various data and programs as described above may be provided in the memory 120 . That is, the memory 120 may be understood as including the database 121 . The database 121 may be any type of storage capable of storing data as well as a database independently built in terms of software.

Processor 130 may be configured to process instructions of a program by performing basic arithmetic, logic, and operations. The processor 130 may perform various functions such as data input/output, data processing, data management, and communication through the network 30 .

The processor 130 may include an image data pre-processing unit 131, a suspicious item detection unit 132, an FC control unit 133, and an infection history management unit 135. In this embodiment, the image data pre-processing unit 131, suspicious item detection unit 132, FC control unit 133, infection history management unit 135 may be included.

The image data pre-processing unit 131 may pre-process the image data by removing features unrelated to the detection of the pine wilt nematode from the collected image data and processing the image of the pine tree to be analyzed. The image data preprocessed by the image data preprocessor 131 may be at least one of first image data, second image data, and third image data.

The suspicious tree detection unit 132 may analyze the preprocessed first image data and detect a suspicious tree that may be determined to be infected with the pine wilt nematode. In addition, the suspicious item detector 132 may derive the location and degree of suspicion of the detected suspicious item as a result. The suspicious tree detection unit 132 may determine the pine tree as a suspicious tree when the derived degree of suspicion is greater than or equal to a reference value.

The FC control unit 133 is a component that controls the position, speed, etc. of the drone of the unmanned aerial vehicle 10, and the sensor data collected by the gyro sensor 112 and the height sensor 113 and the suspicious object detector 132 It is possible to control the flight over the unmanned aerial vehicle 10 based on the location and degree of suspicion derived by the above.

Specifically, when a specific pine tree is determined as a suspicious tree by the suspicious tree detection unit 132, the FC control unit 133 receives the location of the suspicious tree derived by the suspicious tree detection unit 132. , the unmanned aerial vehicle 10 may be moved to a position corresponding to the second altitude and the third altitude for the suspected item. The unmanned aerial vehicle 10 whose location has been moved by the FC control unit 133 may collect second image data from a location corresponding to the second altitude through the collection unit 110, and may collect second image data from a location corresponding to the third altitude. Third image data may be collected. Also, the FC control unit 133 may move the location of the unmanned aerial vehicle 10 to a location for collecting first image data of pine trees located in another location.

The infected tree detection unit 134 may detect an infected tree infected with the pine wood nematode by analyzing the second image data and the third image data that have been collected and preprocessed through the above process. In this case, the location of the detected infected tree may be stored in the memory 120 together with the detection time. The detection time may be information including season, date, time, and the like.

The infection history management unit 135 may manage the infection history of each infected tree detected by the infected tree detection unit 134 and stored in the memory 120 . In addition, it is possible to calculate the number of infected trees and the frequency of infection in the risk area where the infected trees are located. In addition, image data collection and analysis for a risk area with a high frequency of infection can be performed at a relatively shorter cycle.

Meanwhile, the suspicious tree detection unit 132 and the infected tree detection unit 134 may perform image features for detecting pine wilt nematodes through machine learning algorithms. The machine learning algorithm may be configured to detect suspicious and infected trees through stored cumulative information on detection of suspected and infected trees. As an example, a machine learning algorithm may be provided as an algorithm for extracting and comparing colors of image data, and may be provided to detect pine trees discolored by infection with pine wilt nematodes.

The machine learning algorithm may be transmitted from the server 20 through the unmanned aerial vehicle 10 .

The communication unit 140 of the unmanned aerial vehicle 10 may include one or more components that allow the unmanned aerial vehicle 10 to communicate with the server 20 . The communication unit 140 may include at least one of a short-distance communication unit, a mobile communication unit, and a broadcast reception communication unit. However, the type of the communication unit 140 is not limited thereto, and any configuration enabling communication with other devices including the server 20 is not limited to the type.

Hereinafter, the configuration of the server 20 of the forest pest detection system 1 using an unmanned aerial vehicle will be described with reference to FIG. 3 .

FIG. 3 is a diagram schematically illustrating sub-configurations of the server of FIG. 1 .

Referring to FIG. 3 , the server 20 may include a memory 220 , a processor 230 and a communication unit 240 .

First of all, the communication unit 240 of the server 20 is a component enabling communication with the unmanned aerial vehicle 10 through the network 30, and may include at least one of a short-range communication unit, a mobile communication unit, and a broadcast reception communication unit. . However, the type of the communication unit 240 is not limited thereto, and any configuration enabling communication with other devices including the unmanned aerial vehicle 10 is not limited to the type.

The unmanned aerial vehicle 10 and the server 20 may exchange data through respective communication units 140 and 240 . Specifically, the unmanned aerial vehicle 10 may receive information about a machine learning algorithm from the server 20, and the server 20 transmits image data that becomes learning data for learning such a machine learning algorithm to the unmanned aerial vehicle 10. ) can be received from

The memory 220 is a computer-readable recording medium, and includes a non-permanent mass storage device such as a volatile storage device such as RAM (Random Access Memory) and a ROM (Rean Only Memory) and a disk drive. can do. For example, the memory 120 may be a flash memory type, a hard disk type, a multimedia card micro type, or a card type memory (eg SD or XD memory). etc.), RAM (RAM, Random Access Memory), SRAM (Static Random Access Memory), ROM (ROM, Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), It may include at least one type of storage medium among a magnetic memory, a magnetic disk, and an optical disk.

The memory 220 may be provided with at least one program code capable of processing data or machine learning an algorithm for detecting pine wilt nematodes of the present embodiment based on the data. That is, the memory 220 may store a program for processing and controlling the processor 230 .

A database 221 capable of storing the above programs may be provided in the memory 220 . That is, the memory 220 may be understood as including the database 221 . The database 221 may be any type of storage capable of storing data as well as a database independently built in terms of software.

Also, image data received from the unmanned aerial vehicle 10 through the communication unit 240 may be stored in the memory 220 . These image data may be used as learning data for machine learning of a machine learning algorithm. Hereinafter, for convenience of description, image data received from the unmanned aerial vehicle 10 will be referred to as 'learning data'. In this case, the learning data may include all of the first image data, the second image data, and the third image data collected from the unmanned aerial vehicle 10 .

The processor 130 may include a learning data pre-processing unit 232 , a machine learning unit 233 and a lightweight unit 234 .

The learning data pre-processing unit 232 may pre-process the learning data by removing features unrelated to the detection of the pine wilt nematode from the learning data received from the unmanned aerial vehicle 10 and processing an image of a pine tree to be analyzed.

The machine learning unit 233 may analyze the preprocessed learning data to detect an infected tree, and machine learn a machine learning algorithm for detecting an infected tree.

The lightweight unit 234 may lighten the machine learning algorithm machine-learned by the machine learning unit 233 .

The lightweight machine learning algorithm may be transmitted to the unmanned aerial vehicle 10 through the communication unit 240 and stored in the memory 120 of the unmanned aerial vehicle 10 . Accordingly, the unmanned aerial vehicle 10 detects suspicious items and infected items through the machine learning algorithm stored in the memory 120, and autonomously controls the flight of the unmanned aerial vehicle 10 through the FC control unit 133 based on the detected results. can be controlled with

Although the forest pest detection system 1 using an unmanned aerial vehicle according to an embodiment of the present invention has been described as a specific embodiment, this is only an example and the present invention is not limited thereto, and the best light according to the basic ideas disclosed herein should be construed as having a range of A person skilled in the art may implement an embodiment that is not indicated by combining or substituting the disclosed embodiments, but this also does not deviate from the scope of the present invention. In addition, those skilled in the art can easily change or modify the disclosed embodiments based on this specification, and it is clear that such changes or modifications also fall within the scope of the present invention.

1: Forest pest detection system using unmanned aerial vehicles
10: drone
110: collection unit
111: camera
112: gyro sensor
113: height sensor
120: memory
121: database
130: processor
131: image data pre-processing unit
132: suspicious item detection unit
133: FC control unit
134: infection tree detection unit
135: infection history management department
140: communication department
20: server
220: memory
221: database
230: processor
232: training data pre-processing unit
233: machine learning unit
234: lightweight unit
240: communication department
30: network

Claims (9)

In the system for detecting forest pests using unmanned aerial vehicles,
The unmanned aerial vehicle,
a collection unit that collects image data about the forest; and
A suspicious item detection unit configured to analyze the collected image data and detect a suspicious item.
A forest pest detection system using unmanned aerial vehicles. According to claim 1,
The collection unit collects first image data for a forest at a first altitude,
The suspicious item detector detects a suspicious item by analyzing the first image data;
The collection unit collects second image data for the suspicious tree detected at a second altitude when a suspicious tree is detected by the suspicious tree detector.
A forest pest detection system using unmanned aerial vehicles. According to claim 2,
The second altitude is an altitude lower than the first altitude.
A forest pest detection system using unmanned aerial vehicles. According to claim 3,
The first altitude is an altitude of a predetermined height from the ground,
The second altitude is an altitude of a height spaced apart from the top of the suspect tree by a preset vertical distance.
A forest pest detection system using unmanned aerial vehicles. According to claim 2,
The collection unit additionally collects third image data for the suspicious tree detected at a third altitude when a suspicious tree is detected by the suspicious tree detector.
A forest pest detection system using unmanned aerial vehicles. According to claim 5,
The third altitude is an altitude lower than the second altitude.
A forest pest detection system using unmanned aerial vehicles. According to claim 6,
The third altitude is an altitude lower than the top of the suspect tree.
A forest pest detection system using unmanned aerial vehicles. According to claim 1,
The unmanned aerial vehicle further comprises an FC control unit for controlling the position of the unmanned aerial vehicle
A forest pest detection system using unmanned aerial vehicles. According to claim 1,
Further comprising a server for machine learning a suspicious item detection algorithm by the suspicious item detector
A forest pest detection system using unmanned aerial vehicles.

Images