WO2022014586A1

WO2022014586A1 - Inspection method

Info

Publication number: WO2022014586A1
Application number: PCT/JP2021/026276
Authority: WO
Inventors: 登三崎; 治樹岩佐
Original assignee: マルイチエアリアルエンジニア株式会社
Priority date: 2020-07-16
Filing date: 2021-07-13
Publication date: 2022-01-20
Also published as: JP6785023B1; JP2022018783A; JP2022019497A

Abstract

[Problem] The purpose of the present disclosure is to realize a method for efficiently determining risks such as fallen trees as pertains to a plurality of trees. [Solution] One embodiment of the present disclosure is a method for inspecting trees, wherein: image data relating to each tree of a plurality of trees is acquired, the image data being captured by an imaging unit provided to a moving body moving on a prescribed route; the volumes of the trees are calculated on the basis of the captured image data of the trees; the weights of the trees are calculated on the basis of the volumes and specific densities of the trees; and degrees of risk of the trees are determined on the basis of the calculated weights.

Description

Inspection method

The present disclosure relates to tree inspection methods and inspection systems, and in particular, determines the risk of fallen trees, withering, etc. of multiple trees located in a predetermined area, and identifies trees to be subject to preventive pruning and preventive logging. Regarding that.

Conventionally, it has been necessary to manually inspect trees one by one in order to determine the risk of fallen trees and withering due to natural disasters such as typhoons and multiple trees located in a wide area.

Manual diagnosis of each tree is laborious and costly, and by inspecting multiple trees, there is a risk of fallen trees due to variations in the ability of the inspector and fatigue. The accuracy for identifying is also reduced.

Therefore, as a method for diagnosing the state of a tree, for example, the technique disclosed in Patent Document 1 can be mentioned.

JP-A-2005-0436161

However, the technique disclosed in Patent Document 1 diagnoses the corrosion state of a tree using a diagnostic agent, and it is necessary to diagnose each tree using a diagnostic agent, which requires manpower and cost. It takes.

Therefore, the purpose of this disclosure is to realize a method for efficiently determining the risk of fallen trees of a plurality of trees.

One embodiment of the present disclosure is a tree inspection method, in which image data relating to each tree of a plurality of trees captured by an imaging unit provided on a moving body traveling a predetermined route is acquired. , The volume of the tree is calculated based on the image data of the captured tree, the weight of the tree is calculated based on the volume and specific gravity of the tree, and the weight of the tree is calculated based on the calculated weight. Judge the degree of danger of.

According to the present disclosure, it is possible to realize a method for efficiently determining the danger of a plurality of trees such as fallen trees.

It is a conceptual diagram explaining the tree inspection method and system by 1st Embodiment of this disclosure. It is a conceptual diagram explaining the inspection method and system of a plurality of trees according to 1st Embodiment of this disclosure. It is a functional block diagram of the moving body and the analysis apparatus in the inspection system by 1st Embodiment of this disclosure. It is a functional block diagram of the control part of the analysis apparatus by 1st Embodiment of this disclosure. It is a flowchart explaining the inspection method by 1st Embodiment of this disclosure. It is a conceptual diagram explaining the risk degree determination of the inspection method by 1st Embodiment of this disclosure. It is a conceptual diagram explaining the output method of the risk degree determination result of the inspection method by 1st Embodiment of this disclosure.

The contents of the embodiments of the present disclosure will be listed and described. The transport method according to the embodiment of the present disclosure is as follows.
[Item 1]
It ’s a tree inspection method.
The image data related to each tree of a plurality of trees captured by the image pickup unit provided in the moving body moving on a predetermined route is acquired, and the image data is acquired.
Based on the image data of the captured tree, the volume of the tree is calculated.
Based on the volume and specific gravity of the tree, the weight of the tree is calculated.
Judging the risk of the tree based on the calculated weight,
Inspection method.
[Item 2]
The inspection method according to item 1, wherein the actual weight data of the tree is acquired and the specific weight of the tree is calculated based on the actual weight data.
[Item 3]
The inspection method according to item 1, wherein the type of the tree is determined based on machine learning, and the specific weight of the tree is determined based on the determined type of the tree.
[Item 4]
The inspection method according to item 1, wherein it is determined whether or not the tree is a dead tree based on the image data.
[Item 5]
The inspection method according to item 1, wherein the determination result of the degree of danger of the tree is output on a predetermined route included in the map displayed on the user terminal.

Hereinafter, the transport method and the system according to the first embodiment of the present disclosure will be described with reference to the drawings.

FIG. 1 is a conceptual diagram illustrating a tree inspection method and a system according to the first embodiment of the present disclosure.

In FIG. 1, first, the inspection system is composed of a mobile body 10 and an analysis device 11 described later. Here, the moving body 1 passing in the vicinity of the tree 10 captures the appearance of the tree 10 by the imaging unit 2, and the image data of the imaged tree is analyzed by the communication unit 3 via the network, which will be described later. Send to 11.

The moving body 1 can be, for example, a car, railroad, or robot traveling on land, or an aircraft or drone (unmanned aerial vehicle) flying in the sky, and can be manned or unmanned traveling / flying. , Regardless. Hereinafter, a case where the automobile travels on land as the moving body 1 will be described as an example.

The image pickup unit 2 is composed of one or a plurality of cameras, and can be composed of a monocular camera and / or a stereo camera. Assuming a car traveling on a predetermined route as a moving body 1, it is conceivable that the image pickup unit 2 is installed on the roof of the car and one or more cameras shoot a moving image of trees around the car. Will be.

The communication unit 3 connects the mobile body 1 to a network such as the Internet or mobile communication. The communication unit 3 may be provided with a short-range communication interface of Bluetooth (registered trademark) and BLE (Bluetooth Low Energy). The communication unit 3 transmits the image data of the tree imaged by the image pickup unit 2 to the analysis device 11 via the network.

FIG. 2 is a conceptual diagram illustrating a plurality of tree inspection methods and systems according to the first embodiment of the present disclosure.

In the tree inspection method and system of the present embodiment, the moving body 1 moves on a predetermined route shown on the map, and the appearance of the tree 10 around the route is imaged by the imaging unit 2. In FIG. 2, a camera, which is an image pickup unit 2, is installed on the roof of an automobile, which is a mobile body 1. There may be one or more cameras, and a plurality of cameras can image trees located in each direction, and one camera can image trees in one direction. Alternatively, by rotating one camera 360 degrees, trees in the corresponding directions can be imaged, and by using a 360 degree camera, trees in all directions can be imaged.

Further, as shown in FIG. 2, a user terminal (not shown) displays map information, and the current location information of the moving body 1 traveling on the route on the map is used as GPS (Global Positioning System: Global Positioning System). ) Can be used to determine and display. In addition, the tree to be inspected can be displayed as an illustration (icon). It is also possible to hide information about the mobile body 1 from anyone other than the user who is the administrator.

FIG. 3 is a functional configuration diagram of a mobile body and an analysis device in the inspection system according to the first embodiment of the present disclosure.

As shown in FIG. 3, the inspection system has an analysis device 11 in addition to the mobile body 1. The analysis device 11 can be, for example, an information processing device such as a personal computer or a tablet terminal, but may be configured by a smartphone, a mobile phone, a PDA, or the like. The analysis device 11 has at least a communication unit 12, a control unit 13, and a storage unit 14.

The communication unit 12 connects the analysis device 11 to a network such as the Internet or mobile communication. The transmission / reception unit 18 may be provided with a short-range communication interface of Bluetooth (registered trademark) and BLE (Bluetooth Low Energy). The communication unit 12 receives the image data of the tree imaged by the image pickup unit 2 of the mobile body 1 from the mobile body 1 via the network. Further, the communication unit 12 transmits the calculation result and the control information calculated by the control unit 13 to the mobile terminal or other terminal via the network.

The control unit 13 is an arithmetic unit that controls the operation of the entire system, controls the transmission and reception of data between each element, and performs information processing necessary for application execution and authentication processing. For example, the control unit 13 is a CPU (Central Processing Unit), and executes each information processing by executing a program or the like expanded in the storage unit 14.

The storage unit 14 stores a main storage composed of a volatile storage device such as a DRAM (Dynamic Random Access Memory) and an auxiliary storage composed of a non-volatile storage device such as a flash memory or an HDD (Hard Disk Drive). include. The storage unit 14 is used as a work area or the like of the control unit 13, and also stores various setting information and the like.

Further, although not shown, the analysis device 11 can also have a storage. The storage stores various programs such as application programs. A database (not shown) storing data used for each process may be built in the storage.

Although not shown, the analysis device 11 can also include an input / output unit. The input / output device is, for example, a keyboard / mouse for inputting an instruction to operate the system for a worker to carry a load, an information input device such as a touch panel, and an output device such as a display.

Although not shown, the analysis device 11 can be provided with a bus that is commonly connected to each of the above elements and transmits, for example, an address signal, a data signal, and various control signals.

FIG. 4 is a functional configuration diagram of the control unit of the analysis device according to the first embodiment of the present disclosure.

As shown in FIG. 4, the control unit 13 has an information acquisition unit 21, an image analysis unit 22, a weight analysis unit 23, and a determination unit 23.

The information acquisition unit 21 acquires various data from the mobile body 1 and other devices and / or terminals. For example, the image data of the tree captured by the moving body 1 can be acquired via the communication unit 12. Alternatively, the actual weight data of the tree can be acquired.

The image analysis unit 22 executes the image analysis process. For example, the image analysis unit 22 can extract the shape of the tree from the acquired image and calculate the volume of the tree based on the extracted shape. In extracting the shape of a tree, for example, an image recognition method such as CNN (Convolutional Neural Network) is used to calculate the feature amount related to the color and / or shape of the tree, and the image data obtained by capturing another tree. Using the learning model generated based on, the recognition and shape of the tree (more specifically, the type of tree) contained in the image is determined, the size of the tree is estimated, and the tree is based on the estimated size. The volume can be estimated.

The weight analysis unit 23 can derive the specific gravity of the tree based on the acquired actual weight data or the data related to the analyzed tree type, and can calculate the weight of the tree based on the above-estimated volume and specific weight.

The determination unit 24 can determine the degree of risk of fallen trees, withering, etc. of trees based on the image data of the trees and / or the calculated weight of the trees.

FIG. 5 is a flowchart illustrating an inspection method according to the first embodiment of the present disclosure.

First, as step S101, the information acquisition unit 21 of the control unit 13 of the analysis device 11 acquires the image data of the tree captured by the moving body 1 via the communication unit 12. Here, the information acquisition unit 21 can also acquire the position information of the imaged tree, assigns an identification number to the imaged tree, and stores the image data together with the position information in the storage unit 14. can.

Next, in step S102, the image analysis unit 22 of the control unit 13 analyzes the acquired image data. For example, as described above, the image analysis unit 22 can extract the shape of the tree from the acquired image by using a known image recognition method, and calculate the volume of the tree based on the extracted shape. Here, the image analysis unit 22 can also determine the type of the tree based on the shape and color of the leaves of the tree included in the image. The image analysis unit 22 stores the calculated information on the volume of the tree (information on the type of the tree, if necessary) as information related to the tree corresponding to the already assigned identification number, together with other information. It can be stored in 14. Further, the image analysis unit 22 can also determine whether or not the tree is a dead tree based on the color of the trunk, branches, and / or leaves of the tree. Regardless, it is possible to immediately propose logging or pruning of trees. In addition, the amount of leaves, the crotch generated on the trunk, and the like can be referred to in the method of determining the dead tree.

Next, as step S103, the information acquisition unit 21 acquires the actual weight data of the tree via the communication unit 12. Here, the actual weight data can be a numerical value measured by using a measuring tool such as a load cell using a predetermined part of the tree as a sample. The actual weight data can be acquired from the mobile body 1 or can be received from a terminal different from the mobile body 1. Here, the information acquisition unit 21 can store the actual weight data in the storage unit 14 together with other information as information related to the tree corresponding to the already assigned identification number.

In the above-mentioned shape data analysis process, if the type of the tree can be determined, the specific gravity of the tree can be stored in the storage unit 14 in advance as specific gravity data, or can be acquired as external data. Therefore, it is not necessary to acquire the actual weight data as in step S103.

Next, as step S104, the weight analysis unit 23 of the control unit 13 calculates the weight of the tree based on the volume of the tree calculated in step S102 and the actual weight data acquired in step S103. The weight analysis unit 23 can calculate the specific weight of the tree (weight per unit volume) based on the actual weight data, and can calculate the total weight of the tree as the estimated value from the total capacity of the tree calculated as the estimated value. .. Here, as described above, the total weight of the tree can be estimated based on the specific weight data of the tree and the volume of the tree, not based on the actual weight data. The weight analysis unit 23 uses the weight of the tree calculated as an estimated value as the strength, and uses the calculated volume of the tree as information related to the tree corresponding to the already assigned identification number, together with other information, in the storage unit 14. Can be stored in.

Next, as step S105, the determination unit 24 of the control unit 13 determines the risk of fallen or dead trees. For example, as the degree of danger, the load applied to the tree (strong wind withstand load) when a strong wind having a wind speed of 10 to 60 meters (m) is applied from the horizontal direction of the tree is calculated in advance, and the determination unit 24 determines the strong wind. The degree of danger can be determined by comparing the withstand load with the calculated strength (total weight of the tree).

FIG. 6 is a conceptual diagram illustrating the risk determination of the inspection method according to the first embodiment of the present disclosure.

For example, when a wind with a wind speed of v meters is applied to a tree, the wind pressure P is as follows.

Here, ρ is the air density (1.225 kg / m ³ under the condition of 1 atmospheric pressure 15 ° C., v is the wind speed (m / s), and g is the gravitational acceleration (9.8 m / s ² ), which is 60 meters. When the wind speed is applied, the wind pressure P becomes 225 kg / m ² .

In particular, when a strong wind having a wind speed of 60 meters is applied to 10 m above the ground of the tree, the bending moment applied to the trunk of the tree becomes ^{225 kg / m 2} x 10 m = 2250 ^{kg / m 3.}

If the total weight of the tree calculated in step S105 is 2000 kg, there is a risk of the tree falling in consideration of the bending moment, and the determination unit 24 determines that there is a risk of the tree falling.

FIG. 7 is a conceptual diagram illustrating a method of outputting a risk level determination result of the inspection method according to the first embodiment of the present disclosure.

When the determination unit 24 determines the degree of danger for one or more trees and determines that there is a degree of danger, the image generation unit (not shown) of the analysis device 11 displays on the display of the user terminal shown in FIG. 7. It can be output so as to indicate the degree of danger and / or the degree of danger by the icon or the like by superimposing or adjoining the icon of the corresponding tree on the predetermined route included in the map. .. For example, if the total weight of a tree exceeds the strong wind withstand load by a predetermined amount or more, the danger notification icon is not displayed for the tree, and the total weight of the tree exceeds the strong wind withstand load but is less than the predetermined amount. Can display an icon, indicate its level as "1", and inform that it is preferable to cut down. In addition, if the total weight of the tree is less than the load capacity for strong winds, an icon can be displayed to indicate that the level is "2", which indicates that logging will be performed immediately.

From the above, according to the inspection method and system of the present embodiment, it is possible to efficiently inspect the risk of fallen trees of a plurality of trees.

In the above embodiment, the method of determining the risk level of the tree based on the actual weight data of the tree is described, but for example, the risk level can also be determined based on the elasticity data of the tree. For example, the elasticity data is calculated by measuring the winding distance of the rope with respect to the load by pulling the tree from an arbitrary direction with a rope with a load cell or the like and returning it. Here, the values at the time of weighting and the value at the time of weighting are compared, and if there is a proportional relationship (or within an appropriate value), it can be determined that the degree of danger is low.

The above-described embodiment is merely an example for facilitating the understanding of the present disclosure, and is not intended to limit the interpretation of the present disclosure. It goes without saying that the present disclosure may be modified or improved without departing from the spirit thereof, and the present disclosure includes the equivalent thereof.

1 Mobile 2 Imaging unit 3 Communication unit 10 Tree 11 Analysis device 12 Communication unit 13 Control unit 14 Storage unit

Claims

It ’s a tree inspection method.
The image data related to each tree of a plurality of trees captured by the image pickup unit provided in the moving body moving on a predetermined route is acquired, and the image data is acquired.
Based on the image data of the captured tree, the volume of the tree is calculated.
Based on the volume and specific gravity of the tree, the weight of the tree is calculated.
Judging the risk of the tree based on the calculated weight,
Inspection method.
The inspection method according to claim 1, wherein the actual weight data of the tree is acquired and the specific weight of the tree is calculated based on the actual weight data.
The inspection method according to claim 1, wherein the type of the tree is determined based on machine learning, and the specific weight of the tree is determined based on the determined type of the tree.
The inspection method according to claim 1, wherein it is determined whether or not the tree is a dead tree based on the image data.
The inspection method according to claim 1, wherein the determination result of the degree of danger of the tree is output on a predetermined route included in the map displayed on the user terminal.