JP6883829B1 - Tree discrimination program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To discriminate a type of a tree in a forest with higher accuracy. SOLUTION: In a tree discrimination program for discriminating a tree type in a forest, an information acquisition step of acquiring aerial image information of a forest for discriminating a new tree type from the air, and an image of the forest from the air in the past. The degree of association between the reference aerial image information and the type of tree in the forest is referred to, and the degree of association is based on the reference aerial image information according to the aerial image information acquired in the above information acquisition step. It is characterized in that a computer is made to perform a discrimination step of discriminating the type of a tree by giving priority to a higher one. [Selection diagram] Fig. 4

Description

The present invention relates to a tree discrimination program for discriminating the types of trees in a forest.

In recent years, a technique for analyzing the state of a forest based on an aerial image of the forest taken by an artificial satellite or a drone has attracted attention. If the state of the forest can be analyzed, it is possible to formulate a forest cultivation plan, predict the amount of timber production, and formulate each work plan for felling and lumbering.

However, in the past, there has been no proposal for a technique for determining the type of tree with higher accuracy among the forest conditions. If the type of tree cannot be identified, it will be difficult to formulate a forest growth plan and predict the amount of timber produced.

Therefore, the present invention has been devised in view of the above-mentioned problems, and an object of the present invention is to provide a tree discrimination program capable of discriminating the types of trees in a forest with higher accuracy. It is in.

The tree discrimination program according to the present invention is a tree discrimination program for discriminating the type of a tree in a forest, in which aerial image information of a forest for discriminating a new type of tree is captured from the air and a ground image of the forest captured from the ground. A combination having an information acquisition step for acquiring information, reference aerial image information obtained by imaging a forest from the air in the past, and reference ground image information obtained by imaging a forest from the ground in the past, and a type of tree in the forest. Based on the reference aerial image information according to the aerial image information acquired in the above information acquisition step and the reference ground image information according to the ground image information, the above association degree is referred to. It is characterized in that a computer is made to perform a discrimination step of discriminating the type of a tree by giving priority to a higher one.

Even if you do not have any special skills or experience, it is possible to determine the type of forest with high accuracy.

It is a block diagram which shows the whole structure of the system to which this invention is applied. It is a figure which shows the specific configuration example of a search device. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention.

Hereinafter, a tree discrimination program for discriminating the type of tree to which the present invention is applied will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing an overall configuration of a tree discrimination system 1 in which a tree discrimination program to which the present invention is applied is implemented. The tree discrimination system 1 includes an information acquisition unit 9, a search device 2 connected to the information acquisition unit 9, and a database 3 connected to the search device 2.

The information acquisition unit 9 is a device for a person who uses this system to input various commands and information, and specifically, is composed of a keyboard, buttons, a touch panel, a mouse, a switch, and the like. The information acquisition unit 9 is not limited to a device for inputting text information, and may be configured by a device such as a microphone that can detect voice and convert it into text information. Further, the information acquisition unit 9 may be configured as an image pickup device capable of capturing an image of a camera or the like. The information acquisition unit 9 may be configured by a scanner having a function of recognizing a character string from a paper-based document. Further, the information acquisition unit 9 may be integrated with the discrimination device 2 described later. The information acquisition unit 9 outputs the detected information to the determination device 2. Further, the information acquisition unit 9 may be configured by means for specifying the position information by scanning the map information. Further, the information acquisition unit 9 may be composed of an illuminance sensor for measuring a temperature sensor, a humidity sensor, and a wind direction sensor. Further, the information acquisition unit 9 may be composed of a communication interface for acquiring data about the weather from the Japan Meteorological Agency or a private weather forecast company. Further, the information acquisition unit 9 may be composed of a body sensor that is attached to the body to detect body data, and this body sensor detects, for example, body temperature, heart rate, blood pressure, number of steps, walking speed, and acceleration. It may be composed of a sensor for the purpose. Further, the body sensor may acquire biological data of animals as well as humans. Further, the information acquisition unit 9 may be configured as a device that acquires information such as drawings by scanning or reading it from a database. In addition to these, the information acquisition unit 9 may be configured by an odor sensor that detects an odor or scent.

The database 3 stores various information necessary for discriminating trees. Database 3 refers to reference aerial image information in which forests were imaged from the air in the past, reference ground image information in which forests were imaged from the ground in the past, and land classification in which forests were imaged in the past. Land classification information, reference terrain information about the terrain of the land where the forest has imaged reference aerial image information in the past, distance from the measurement point of the forest where reference aerial image information was imaged in the past is measured by a distance measuring sensor Reference distance information, reference shape information consisting of the thickness and height of trees by measuring the distance from the measurement point for trees in the forest where the aerial image information for reference was imaged in the past, in the past A data set of reference climate information regarding the climate of the area where the forest is located, which is an image of reference aerial image information, and the types of trees identified in the past is stored.

That is, any one of such reference aerial image information, reference ground image information, reference land classification information, reference terrain information, reference distance information, reference shape information, and reference climate information is stored in the database 3. One or more and the types of trees determined in the past are stored in association with each other.

The search device 2 is composed of, for example, an electronic device such as a personal computer (PC), but is embodied in any other electronic device such as a mobile phone, a smartphone, a tablet terminal, a wearable terminal, etc., in addition to the PC. It may be converted. The user can obtain a search solution by the search device 2.

FIG. 2 shows a specific configuration example of the search device 2. The search device 2 performs wired communication or wireless communication with a control unit 24 for controlling the entire search device 2 and an operation unit 25 for inputting various control commands via operation buttons, a keyboard, or the like. A communication unit 26 for the purpose, an estimation unit 27 for making various judgments, and a storage unit 28 for storing a program for performing a search to be executed represented by a hard disk or the like are connected to the internal bus 21, respectively. .. Further, a display unit 23 as a monitor that actually displays information is connected to the internal bus 21.

The control unit 24 is a so-called central control unit for controlling each component mounted in the search device 2 by transmitting a control signal via the internal bus 21. Further, the control unit 24 transmits various control commands via the internal bus 21 in response to the operation via the operation unit 25.

The operation unit 25 is embodied by a keyboard or a touch panel, and an execution command for executing a program is input from the user. When the execution command is input by the user, the operation unit 25 notifies the control unit 24 of the execution command. Upon receiving this notification, the control unit 24, including the estimation unit 27, executes a desired processing operation in cooperation with each component. The operation unit 25 may be embodied as the information acquisition unit 9 described above.

The estimation unit 27 estimates the search solution. The estimation unit 27 reads out various information stored in the storage unit 28 and various information stored in the database 3 as necessary information when executing the estimation operation. The estimation unit 27 may be controlled by artificial intelligence. This artificial intelligence may be based on any well-known artificial intelligence technique.

The display unit 23 is composed of a graphic controller that creates a display image based on the control by the control unit 24. The display unit 23 is realized by, for example, a liquid crystal display (LCD) or the like.

When the storage unit 28 is composed of a hard disk, predetermined information is written to each address based on the control by the control unit 24, and is read out as needed. Further, the storage unit 28 stores a program for executing the present invention. This program is read and executed by the control unit 24.

The operation in the tree discrimination system 1 having the above-described configuration will be described.

In the tree discrimination system 1, for example, as shown in FIG. 3, it is premised that three or more levels of association between the reference aerial image information and the tree type are set and acquired in advance.

The reference aerial image information is an image obtained by capturing a forest from the air or the sky. In order to acquire this reference aerial image information, it is possible to obtain an image of the forest from the sky by an aircraft, a helicopter, or an unmanned aerial vehicle such as a drone. Further, the reference aerial image information may be composed of satellite images captured by artificial satellites. The reference aerial image information may be composed of a so-called spectrum image color-coded for each spectrum in addition to the normal RGB image.

The type of tree is composed of the actual type name of the tree such as Sugi, Quercus acutissima, Japanese cypress, and pine.

That is, through this reference aerial image information and the tree type data set, the relationship between the characteristics of various images generated in the reference aerial image information and the tree type can be understood. That is, the image features of the reference aerial image information and the types of trees are a data set. Therefore, by collecting the reference aerial image information and the data set of the tree type, it is possible to know what kind of image features were present and how the tree type was determined in the past. Become.
When constructing this data set, an aerial image for reference is acquired by taking an image from the sky of a forest area whose tree type is known in advance, and the known tree type and data set are created.

In the example of FIG. 3, it is assumed that the input data is, for example, reference aerial image information P01 to P03. The reference aerial image information as such input data is linked to the output. In this output, the type of tree as the output solution is displayed.

The reference aerial image information is associated with each other through three or more levels of association with the type of tree as the output solution. Reference aerial image information is arranged on the left side via this degree of association, and tree types are arranged on the right side via this degree of association. The degree of association indicates the degree of relevance to which tree type is highly related to the reference aerial image information arranged on the left side. In other words, this degree of association is an index indicating what kind of tree the reference aerial image information is likely to be associated with, and the most probable tree type is selected from the reference aerial image information. It shows the accuracy in doing so. In the example of FIG. 3, w13 to w19 are shown as the degree of association. These w13 to w19 are shown in 10 stages as shown in Table 1 below, and the closer to 10 points, the higher the degree of relevance of each combination as an intermediate node to the type of tree as an output. On the contrary, the closer to one point, the lower the degree of relevance of each combination as an intermediate node to the type of tree as an output.

The search device 2 acquires in advance the degree of association w13 to w19 of three or more stages shown in FIG. That is, the search device 2 accumulates past data as to which of the reference aerial image information and the type of tree in that case was adopted in determining the actual search solution, and analyzes these. By analyzing, the degree of association shown in FIG. 3 is created.

For example, it is assumed that a certain reference aerial image information is determined to be a tree type A (for example, Japanese cypress). In such a situation, it is assumed that the reference aerial image information of a similar pattern is similarly determined to be tree type A. In such a case, the degree of association of the tree type A becomes stronger. On the other hand, in the exact same pattern (classification) of the aerial image information for reference, many were determined to be tree type B (for example, Sugi), and few were determined to be tree type A. To do. In such a case, the degree of association of the tree type B becomes stronger, and the degree of association of the tree type A becomes lower.

This analysis and analysis may be performed by artificial intelligence. In such a case, for example, in the case of reference aerial image information P01, analysis is performed from the data of the determination result of the past tree type. In the case of reference aerial image information P01, if there are many cases of tree type A, the degree of association connected to this tree type A is set higher, and if there are many cases of tree type B, this is used. Set a higher degree of association that leads to tree type B. For example, in the example of the reference aerial image information P01, the tree type A and the tree type B are linked, but from the previous case, the degree of association of w13 connected to the tree type A is set to 7 points, and the tree type B is used. The degree of association of w14 that leads to is set to 2 points.

Further, the degree of association shown in FIG. 3 may be composed of the nodes of the neural network in artificial intelligence. That is, the weighting coefficient for the output of the node of this neural network corresponds to the above-mentioned degree of association. Further, the network is not limited to a neural network, and may be composed of all decision-making factors constituting artificial intelligence.

In such a case, as shown in FIG. 4, reference aerial image information is input as input data, each tree type is output as output data, and at least one or more hidden layers are formed between the input node and the output node. It may be provided and machine-learned. On the contrary, the tree type may be input and the reference aerial image information may be output.

Such a degree of association is what is called learned data in artificial intelligence. After creating such trained data, the type of tree will be predicted using the above-mentioned trained data when actually determining the type of tree from now on. In such a case, the aerial image information of the forest for which the type of the tree is to be newly determined is acquired. The method of acquiring the aerial image information is the same as the above-mentioned reference aerial image information.

The newly acquired aerial image information is input by the information acquisition unit 9 described above. The information acquisition unit 9 may acquire such aerial image information as electronic data.

Based on the aerial image information newly acquired in this way, the type of tree that is likely to be actually determined is searched for the aerial image information. In such a case, the degree of association shown in FIG. 3 (Table 1) acquired in advance is referred to. For example, when the newly acquired aerial image information is the same as or similar to P02, the tree type B is associated with the association degree w15 and the tree type C is associated with the association degree w16 through the association degree. .. In such a case, the tree type B having the highest degree of association is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the type C of the tree in which the degree of association is low but the association itself is recognized may be selected as the optimum solution. In addition to this, it goes without saying that an output solution to which the arrows are not connected may be selected, and any other output solution may be selected in any other priority as long as it is based on the degree of association.

In this way, it is possible to search for the type of tree to be determined from the newly acquired aerial image information and display it to the user. By seeing this search result, the user can formulate a tree felling plan and a growing plan based on the type of the searched tree. By the way, in the process of outputting this tree type, it is advised not only to display only the searched tree type, but also to display the specific time when logging should be done based on this tree type. May be configured.

In the example of FIG. 5, it is assumed that the input data is, for example, reference aerial image information P01 to P03 and reference ground image information P14 to 17. The intermediate node shown in FIG. 5 is a combination of the reference aerial image information and the reference ground image information as such input data. Each intermediate node is further linked to the output. In this output, the type of each tree as an output solution is displayed.

In the example of FIG. 5, it is premised that a combination of the reference aerial image information and the reference ground image information is formed. The reference ground image information is an image obtained by actually capturing a forest from the ground. The reference ground image information may be composed of a so-called spectrum image color-coded for each spectrum in addition to the normal RGB image. Further, the ground image information for reference may be an image of a branch or leaf portion of a tree, but may be an image of a trunk portion or a root portion of a tree.

In the example of FIG. 5, it is assumed that the input data is, for example, reference aerial image information P01 to P03 and reference ground image information P14 to 17. The intermediate node shown in FIG. 5 is a combination of the reference aerial image information and the reference ground image information as such input data. Each intermediate node is further linked to the output. In this output, the type of tree as the output solution is displayed.

Each combination (intermediate node) of the reference aerial image information and the reference ground image information is associated with each other through three or more levels of association with the type of tree as the output solution. The reference aerial image information and the reference ground image information are arranged on the left side through this degree of association, and the tree types are arranged on the right side through this degree of association. The degree of association indicates the degree of relevance to each tree type with respect to the reference aerial image information and the reference ground image information arranged on the left side. In other words, this degree of association is an index indicating what kind of tree the reference aerial image information and the reference ground image information are likely to be associated with, and the reference aerial image information and the reference site. It shows the accuracy in selecting the most probable tree type from the above image information. In addition to the aerial image information, it is possible to improve the accuracy by discriminating according to the image actually captured from the ground. Therefore, the optimum type of tree is searched for by combining the reference aerial image information and the reference ground image information.

In the example of FIG. 5, w13 to w22 are shown as the degree of association. As shown in Table 1, these w13 to w22 are shown in 10 stages, and the closer to 10 points, the higher the degree of relevance of each combination as an intermediate node to the output, and conversely, 1 point. The closer they are, the less relevant each combination as an intermediate node is to the output.

The search device 2 acquires in advance the degree of association w13 to w22 of three or more stages shown in FIG. That is, the search device 2 accumulates past data as to which of the reference aerial image information, the reference ground image information, and the type of tree in that case was suitable for determining the actual search solution. By analyzing and analyzing these, the degree of association shown in FIG. 5 is created.

This analysis and analysis may be performed by artificial intelligence. In such a case, for example, in the case of the reference aerial image information P01 and the reference ground image information P16, the type of the tree is analyzed from the past data. When there are many cases of tree type A, the degree of association connected to this tree type A is set higher, and when there are many cases of tree type B and there are few cases of tree type A, there are few cases of tree type A. The degree of association connected to type B is set high, and the degree of association connected to type A of trees is set low. For example, in the example of the intermediate node 61a, the output of the tree type A and the tree type B is linked, but from the previous case, the degree of association of w13 connected to the tree type A is set to 7 points, and the tree type B is set. The degree of association of the connected w14 is set to 2 points.

Further, the degree of association shown in FIG. 5 may be composed of the nodes of the neural network in artificial intelligence. That is, the weighting coefficient for the output of the node of this neural network corresponds to the above-mentioned degree of association. Further, the network is not limited to a neural network, and may be composed of all decision-making factors constituting artificial intelligence.

In the example of the degree of association shown in FIG. 5, the node 61b is a node in which the reference ground image information P14 is combined with the reference aerial image information P01, the degree of association of the tree type C is w15, and the tree type. The degree of association of E is w16. The node 61c is a node that is a combination of the reference ground image information P15 and P17 with respect to the reference aerial image information P02, and the degree of association of the tree type B is w17 and the degree of association of the tree type D is w18. ing.

Such a degree of association is what is called learned data in artificial intelligence. After creating such learned data, when actually performing a search for determining the type of tree from now on, the above-mentioned learned data will be used. In such a case, in addition to the aerial image information, the ground image information is acquired from the company that newly determines the type of the tree. This ground image information corresponds to the above-mentioned reference ground image information, and the acquisition method thereof is also the same.

Based on the newly acquired aerial image information and ground image information in this way, the optimum tree type is searched for. In such a case, the degree of association shown in FIG. 5 (Table 1) acquired in advance is referred to. For example, when the newly acquired aerial image information is the same as or similar to P02 and the ground image information is the same as or similar to P17, the node 61d is determined via the degree of association. The node 61d is associated with the tree type C by w19 and the tree type D by the degree of association w20. In such a case, the tree type C having the highest degree of association is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the type D of the tree in which the degree of association is low but the association itself is recognized may be selected as the optimum solution. In addition to this, it goes without saying that an output solution to which the arrows are not connected may be selected, and any other output solution may be selected in any other priority as long as it is based on the degree of association.

In addition, Table 2 below shows examples of the degrees of association w1 to w12 extending from the input.

The intermediate node 61 may be selected based on the degree of association w1 to w12 extending from this input. That is, the larger the degree of association w1 to w12, the heavier the weighting in the selection of the intermediate node 61 may be. However, the degrees of association w1 to w12 may all have the same value, and the weights in the selection of the intermediate node 61 may all be the same.

FIG. 6 shows an example in which a combination of the above-mentioned reference aerial image information and reference land classification information and a degree of association of three or more levels of the tree type with respect to the combination are set.

The reference land classification information is composed of, for example, land classification survey data published and published by the Ministry of Land, Infrastructure, Transport and Tourism, and is information on the geology and soil of the surface layer, the land use status, and the disaster history. This reference land classification information is composed of information on the geology and soil of the surface layer.

In the example of FIG. 6, it is assumed that the input data is, for example, reference aerial image information P01 to P03 and reference land classification information P18 to 21. The intermediate node shown in FIG. 6 is a combination of the reference land classification information and the reference aerial image information as such input data. Each intermediate node is further linked to the output. In this output, the type of tree as the output solution is displayed.

Each combination (intermediate node) of the reference aerial image information and the reference land classification information is associated with each other through three or more levels of association with the tree type as this output solution. The reference aerial image information and the reference land classification information are arranged on the left side through this degree of association, and the tree types are arranged on the right side through this degree of association. The degree of association indicates the degree of relevance to the type of tree with respect to the reference aerial image information and the reference land classification information arranged on the left side. In other words, this degree of association is an index showing what kind of tree the reference aerial image information and the reference land classification information are likely to be associated with, and the reference aerial image information and the reference land. It shows the accuracy in selecting the most probable tree type from the classification information. In addition to aerial image information, the types can be narrowed down according to the relationship with the actual soil. Therefore, the optimum tree type is searched for by combining the reference aerial image information and the reference land classification information.

In the example of FIG. 6, w13 to w22 are shown as the degree of association. As shown in Table 1, these w13 to w22 are shown in 10 stages, and the closer to 10 points, the higher the degree of relevance of each combination as an intermediate node to the output, and conversely, 1 point. The closer they are, the less relevant each combination as an intermediate node is to the output.

The search device 2 acquires in advance the degree of association w13 to w22 of three or more stages shown in FIG. That is, the search device 2 accumulates past data as to which of the reference aerial image information, the reference land classification information, and the type of tree in that case was suitable for determining the actual search solution. By analyzing and analyzing these, the degree of association shown in FIG. 6 is created.

This analysis and analysis may be performed by artificial intelligence. In such a case, for example, in the case of the reference aerial image information P01 and the reference land classification information P20, the type of the tree is analyzed from the past data. For example, in the example of the intermediate node 61a, the output of the tree type A and the tree type B is linked, but from the previous case, the degree of association of w13 connected to the tree type A is set to 7 points, and the tree type B is used. The degree of association of w14 that leads to is set to 2 points.

Further, the degree of association shown in FIG. 6 may be composed of the nodes of the neural network in artificial intelligence. That is, the weighting coefficient for the output of the node of this neural network corresponds to the above-mentioned degree of association. Further, the network is not limited to a neural network, and may be composed of all decision-making factors constituting artificial intelligence.

In the example of the degree of association shown in FIG. 6, the node 61b is a node in which the reference land classification information P18 is combined with the reference aerial image information P01, the degree of association of the tree type C is w15, and the tree type. The degree of association of E is w16. The node 61c is a node that is a combination of the reference land classification information P19 and P21 with respect to the reference aerial image information P02, and the degree of association of the tree type B is w17 and the degree of association of the tree type D is w18. ing.

Such a degree of association is what is called learned data in artificial intelligence. After creating such learned data, when actually determining the type of a new forest tree from now on, the above-mentioned learned data will be used. In such a case, in addition to the above-mentioned aerial image information, the land classification information of the forest for newly determining the type of tree is acquired. The land classification information corresponds to the reference land classification information, and may be acquired from data stored by, for example, the Ministry of Land, Infrastructure, Transport and Tourism or other organizations.

Based on the newly acquired aerial image information and land classification information in this way, the type of tree is searched for. In such a case, the degree of association shown in FIG. 6 (Table 1) acquired in advance is referred to. For example, when the newly acquired aerial image information is the same as or similar to P02 and the land classification information is P21, the node 61d is associated via the degree of association. The node 61d is associated with the tree type C by w19 and the tree type D by the degree of association w20. In such a case, the tree type C having the highest degree of association is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the type D of the tree in which the degree of association is low but the association itself is recognized may be selected as the optimum solution. In addition to this, it goes without saying that an output solution to which the arrows are not connected may be selected, and any other output solution may be selected in any other priority as long as it is based on the degree of association.

The embodiment shown in FIG. 6 described above uses the reference land classification information as the reference information, but the present invention is not limited to this, and the reference topography information is used as a substitute for the reference land classification information. May be trained together with the reference aerial image information. This reference terrain information can also be obtained from data held by the Ministry of Land, Infrastructure, Transport and Tourism, and is all information regarding the terrain of the land where the forest is an image of the reference aerial image information. Examples of this reference terrain information include elevation data and contour data, slopes of trees growing on mountain slopes, and positional relationships with rivers if there are rivers. Contains data.

In such a case, in FIG. 6, a combination having reference topographical information and reference aerial image information as a substitute for reference land classification information and a degree of association of three or more levels with the type of tree in the forest are formed. I will do it. Then, as input data, in addition to the aerial image information, new topographical information is acquired. This terrain information is information on the terrain of the forest that captures the aerial image information, and the content of the information corresponds to the terrain information for reference. Based on the same or similar reference topographical information as the acquired topographical information, the one with a higher degree of association is prioritized to determine the type of tree.

The embodiment shown in FIG. 6 described above uses the reference land classification information as the reference information, but is not limited to this, and is used as a substitute for the reference land classification information in the past. Reference climate information regarding the climate of the area where the forest where the aerial image information is captured may be learned together with the reference aerial image information. This reference climate information can be obtained from data held by, for example, the Japan Meteorological Agency or a private weather forecasting company, and is all information about the climate in the area where the forest is imaged with the reference aerial image information. Examples of this reference climate information are information on past temperature and humidity, weather, precipitation, wind direction, wind speed, etc. in the area. The type of trees in the forest may correlate with the climate in the area, so this is added to the training data.

In such a case, in FIG. 6, a combination having reference climate information and reference aerial image information as a substitute for reference land classification information and a degree of association of three or more levels with the type of tree in the forest are formed. I will do it. Then, as input data, in addition to the aerial image information, new climate information is acquired. This climate information is information on the climate in the forest area where the aerial image information is captured, and the content of the information depends on the reference climate information. Based on the same or similar reference climate information as the acquired climate information, the one with the higher degree of association is prioritized to determine the type of tree.

FIG. 7 shows an example in which a combination of the above-mentioned reference aerial image information and the reference distance information and a degree of association of three or more levels of the tree type with respect to the combination are set.

The reference distance information is information obtained by measuring the distance from the measurement point of the forest where the reference aerial image information was captured by the distance measurement sensor in the past. The distance measuring sensor may be, for example, a known distance measuring sensor using infrared rays, or a known laser range finder may be used. Further, an optical radar or a laser radar such as LiDAR (Light Detection And Ranging) may be used. That is, distance sensing and two-dimensional or three-dimensional spatial imaging may be acquired via a laser image by using laser light instead of electromagnetic waves.

When measuring a distance with such a distance measuring sensor, sensing is performed from a certain measurement point on the ground. The target of sensing may focus on one tree, or may capture a plurality of trees. When focusing on one tree, it may be decided in advance whether the target is the branches and leaves of the tree or the trunk of the tree.

In the example of FIG. 7, it is assumed that the input data is, for example, reference aerial image information P01 to P03 and reference distance information P18 to 21. The intermediate node shown in FIG. 7 is a combination of reference distance information and reference aerial image information as such input data. Each intermediate node is further linked to the output. In this output, the type of tree as the output solution is displayed.

Each combination (intermediate node) of the reference aerial image information and the reference distance information is associated with each other through three or more levels of association with the type of tree as this output solution. The reference aerial image information and the reference distance information are arranged on the left side via this degree of association, and the tree types are arranged on the right side via this degree of association. The degree of association indicates the degree of relevance to the type of tree with respect to the reference aerial image information and the reference distance information arranged on the left side. In other words, this degree of association is an index indicating what kind of tree the reference aerial image information and the reference distance information are likely to be associated with, and the reference aerial image information and the reference distance information. It shows the accuracy in selecting the most probable tree type from. Two-dimensional and three-dimensional shapes of trees can also be detected via distance information, and the types of trees can be narrowed down through this information. Therefore, the optimum tree type is searched for by combining the reference aerial image information and the reference distance information.

In the example of FIG. 7, w13 to w22 are shown as the degree of association. As shown in Table 1, these w13 to w22 are shown in 10 stages, and the closer to 10 points, the higher the degree of relevance of each combination as an intermediate node to the output, and conversely, 1 point. The closer they are, the less relevant each combination as an intermediate node is to the output.

The search device 2 acquires in advance the degree of association w13 to w22 of three or more stages shown in FIG. 7. That is, the search device 2 accumulates past data as to which of the reference aerial image information, the reference distance information, and the type of tree in that case was suitable for determining the actual search solution. By analyzing and analyzing these, the degree of association shown in FIG. 7 is created.

This analysis and analysis may be performed by artificial intelligence. In such a case, for example, in the case of the reference aerial image information P01 and the reference distance information P20, the type of the tree is analyzed from the past data. For example, in the example of the intermediate node 61a, the output of the tree type A and the tree type B is linked, but from the previous case, the degree of association of w13 connected to the tree type A is set to 7 points, and the tree type B is used. The degree of association of w14 that leads to is set to 2 points.

Further, the degree of association shown in FIG. 7 may be composed of the nodes of the neural network in artificial intelligence. That is, the weighting coefficient for the output of the node of this neural network corresponds to the above-mentioned degree of association. Further, the network is not limited to a neural network, and may be composed of all decision-making factors constituting artificial intelligence.

In the example of the degree of association shown in FIG. 7, the node 61b is a node in which the reference distance information P18 is combined with the reference aerial image information P01, the degree of association of the tree type C is w15, and the tree type E. The degree of association is w16. The node 61c is a node in which the reference distance information P19 and P21 are combined with respect to the reference aerial image information P02, and the degree of association of the tree type B is w17 and the degree of association of the tree type D is w18. There is.

Such a degree of association is what is called learned data in artificial intelligence. After creating such learned data, when actually determining the type of a new forest tree from now on, the above-mentioned learned data will be used. In such a case, in addition to the above-mentioned aerial image information, the distance information of the forest for newly determining the type of tree is acquired. The distance information corresponds to the reference distance information, and may be acquired from data stored by, for example, the Ministry of Land, Infrastructure, Transport and Tourism or other organizations.

Based on the newly acquired aerial image information and distance information in this way, the type of tree is searched. In such a case, the degree of association shown in FIG. 7 (Table 1) acquired in advance is referred to. For example, when the newly acquired aerial image information is the same as or similar to P02 and the land classification information is P21, the node 61d is associated via the degree of association. The node 61d is associated with the tree type C by w19 and the tree type D by the degree of association w20. In such a case, the tree type C having the highest degree of association is selected as the optimum solution. However, it is not essential to select the one with the highest degree of association as the optimum solution, and the type D of the tree in which the degree of association is low but the association itself is recognized may be selected as the optimum solution. In addition to this, it goes without saying that an output solution to which the arrows are not connected may be selected, and any other output solution may be selected in any other priority as long as it is based on the degree of association.

In the form shown in FIG. 7, the reference distance information is not used as it is, but the reference shape information consisting of the shape of the tree acquired based on the data measured by the distance measuring sensor is used. You may. The shape of the tree referred to here includes information on all shapes such as the thickness of the trunk of the tree, the height of the trunk, the degree of branching, the shape of the epidermis of the tree, the height of the tree, and the position where the leaves grow.

In such a case, the relationship between the distance data measured by the distance measuring sensor and the shape data of the tree is investigated in advance, and a table of the shape data of the tree with respect to the distance data measured by the distance measuring sensor is created. This is stored in the database 3. Then, when the distance data is newly measured by the distance measuring sensor, the table in which the shape data of the corresponding tree is stored in the database 3 is referred to, and the shape data of the tree corresponding to this is read out.

At this time, as shown in FIG. 8, a machine learning model in which the reference distance information and the shape of the tree are previously trained as a data set may be used. For the shape of the tree judged by humans or the shape of the tree judged from the image, a data set is created with the reference distance information based on the distance data by the distance measuring sensor, and this is learned. Then, when the distance information is newly acquired as the input information, the shape of the tree corresponding to this is solved and searched. The specific method of this solution search will be omitted below by quoting the explanations of FIGS. 3 and 4.

FIG. 9 shows three or more stages of the reference shape information obtained by converting the shape of the tree acquired through the reference distance information into data or directly acquired through the distance measuring sensor, and the type of tree for the combination of the reference aerial image. An example is shown in which the degree of association of is set.

The search device 2 acquires in advance the degree of association w13 to w22 of three or more stages shown in FIG. That is, the search device 2 accumulates past data as to which of the reference aerial image information, the reference shape information, and the type of tree in that case was suitable for determining the actual search solution. By analyzing and analyzing these, the degree of association shown in FIG. 9 is created. This analysis and analysis may be performed by artificial intelligence.

Such a degree of association is what is called learned data in artificial intelligence. After creating such learned data, when actually determining the type of a new forest tree from now on, the above-mentioned learned data will be used. In such a case, shape information is acquired in addition to the above-mentioned aerial image information. The shape information corresponds to the reference shape information, and the acquisition method thereof is the same as that of the reference shape information, and the above-mentioned table stored in the database 3 and the machine learning model shown in FIG. 8 are used.

Based on the newly acquired aerial image information and shape information in this way, the type of tree is searched for. In such a case, the solution is searched in the same manner as described above with reference to the degree of association shown in FIG. 9 (Table 1) acquired in advance.

In the above-described embodiment, the case where the relationship between the reference aerial image information and the type of the tree is learned in advance has been described as an example, but the present invention is not limited to this. As an alternative to the reference aerial image information, the relationship between the reference ground image information and the tree type may be learned and related via the above-mentioned degree of association, or the reference distance information and the tree type may be associated with each other. The relationship may be learned and related through the above-mentioned degree of association, or further, the relationship between the reference shape information and the type of tree may be learned and related through the above-mentioned degree of association. .. In such a case, when the input of information (any of ground image information, distance information, and shape information) corresponding to the learned reference information is accepted, the type of tree as a search solution corresponding to the input is selected. Ask. The specific method for obtaining the above will be omitted below by quoting the above-mentioned explanations of FIGS. 3 and 4.

In addition, when the reference information as the keynote is any of the reference ground image information, the reference distance information, and the reference shape information, this and other reference information (reference ground image information, reference land). Classification information, reference terrain information, reference distance information, reference shape information, reference climate information) may be combined to form three or more levels of association with the type of tree. Then, as input data, information (ground image information, distance information, shape information) according to the reference information (any of the reference ground image information, the reference distance information, and the reference shape information) as the keynote, and the keynote When information (ground image information, land classification information, terrain information, distance information, shape information, climate information) corresponding to other reference information combined with the reference information to be input is input, the search solution is the same as above. Find the type of tree as. The specific method for obtaining the above will be omitted below by quoting the above-mentioned explanations in FIGS. 5 to 7.

In addition to the basic information (aerial image information, ground image information, distance information, shape information), other information (ground image information, land classification information, topography information, distance information, shape information, climate information) When any two or more are acquired, the reference information (reference aerial image information, reference ground image information, reference) based on the two or more reference information corresponding to the acquired two or more information. By creating learning data consisting of three or more levels of association between the combination with distance information and reference shape information and the type of tree for the combination, it is possible to search for a solution to the credit rating in the same way. ..

In the above-mentioned degree of association, the degree of association is expressed by a 10-step evaluation, but it is not limited to this, and it may be expressed by a degree of association of 3 or more levels, and conversely, it may be expressed by 3 or more levels. For example, 100 steps or 1000 steps may be used. On the other hand, this degree of association does not include those expressed in two stages, that is, whether or not they are related to each other, either 1 or 0.

According to the present invention having the above-described configuration, anyone can easily search for the creditworthiness of a company considering financing without any special skill or experience. Further, according to the present invention, it is possible to make a judgment of this search solution with higher accuracy than a human being. Further, by configuring the above-mentioned degree of association with artificial intelligence (neural network or the like), it is possible to further improve the discrimination accuracy by learning this.

Since there are many cases where the input data and the output data described above do not exist exactly the same in the process of learning, the input data and the output data may be classified by type. That is, the information P01, P02, ... P15, 16, ... That constitute the input data are classified according to the criteria classified in advance on the system side or the user side according to the content of the information, and the classified inputs. A data set may be created between the data and the output data and trained.

Further, according to the present invention, it is characterized in that an optimum solution search is performed through the degree of association set in three or more stages. The degree of association can be described by, for example, a numerical value from 0 to 100% in addition to the above-mentioned 10 steps, but is not limited to this, and any step can be described as long as it can be described by a numerical value of 3 or more steps. It may be configured.

In a situation where there are multiple possible candidates for a search solution by determining the creditworthiness with higher credibility and lower misunderstanding based on the degree of association expressed by such numerical values of three or more levels. It is also possible to search and display in descending order of the degree of association.

In addition to this, according to the present invention, it is possible to judge without overlooking the discrimination result of the output having an extremely low degree of association such as 1%. It warns the user that even a judgment result with an extremely low degree of association is connected as a slight sign and may be useful as the judgment result once every tens or hundreds of times. be able to.

Further, according to the present invention, there is an advantage that the search policy can be determined by the method of setting the threshold value by performing the search based on the degree of association of three or more stages. If the threshold value is lowered, even if the above-mentioned degree of association is 1%, it can be picked up without omission, but it is unlikely that a more appropriate discrimination result can be detected favorably, and a lot of noise may be picked up. is there. On the other hand, if the threshold value is raised, there is a high possibility that the optimum search solution can be detected with high probability, but the degree of association is usually low and it is passed through, but it is suitable that it appears once every tens or hundreds of times. Sometimes the solution is overlooked. It is possible to decide which one to prioritize based on the ideas of the user side and the system side, but it is possible to increase the degree of freedom in selecting the points to be emphasized.

Further, in the present invention, the above-mentioned degree of association may be updated. This update may reflect information provided via a public communication network such as the Internet. In addition to market information, when event information, external environmental information, household information, real estate information, expert opinion information, and natural environment information knowledge, information, and data are acquired, the degree of association is increased or decreased accordingly. Let me.

In other words, this update corresponds to learning in the sense of artificial intelligence. It can be said that it is a learning act because it acquires new data and reflects it in the learned data.

In addition, this update of the degree of association is done by the system side or the user side based on the contents of research data, treatises, conference presentations, newspaper articles, books, etc. by experts, except when it is based on information that can be obtained from the public communication network. It may be updated artificially or automatically. Artificial intelligence may be utilized in these update processes.

Further, the process of first creating the trained model and the above-mentioned update may use not only supervised learning but also unsupervised learning, deep learning, reinforcement learning, and the like. In the case of unsupervised learning, instead of reading and learning the data set of input data and output data, information corresponding to the input data is read and trained, and the degree of association related to the output data is self-formed from there. You may let it.

1 Tree discrimination system 2 Search device 21 Internal bus 23 Display unit 24 Control unit 25 Operation unit 26 Communication unit 27 Estimate unit 28 Storage unit 61 Node

Claims (7)

In a tree discrimination program that discriminates the types of trees in the forest
And aerial image information of the captured forest to determine the type of trees from the air newly, an information acquisition step of acquiring and ground image information obtained by imaging the forest from the ground,
Refer to the combination of the reference aerial image information obtained by capturing the forest from the air in the past and the reference ground image information obtained by capturing the forest from the ground in the past, and the degree of association of three or more levels between the types of trees in the forest. Then, based on the reference aerial image information according to the aerial image information acquired in the above information acquisition step and the reference ground image information according to the ground image information, the tree having a higher degree of association is prioritized and the tree is given priority. A tree discrimination program characterized in that a computer executes a discrimination step for discriminating the type. In a tree discrimination program that discriminates the types of trees in the forest
And aerial image information newly captured forest to determine the type of trees from the air, an information acquisition step of acquiring the land classification information regarding the classification of land there is the forest,
A combination having reference aerial image information obtained by imaging a forest from the air in the past and reference land classification information relating to the classification of a certain land in which a forest imaged reference aerial image information in the past, and the types of trees in the forest. Refer to the three or more levels of association, and use the reference aerial image information according to the aerial image information acquired in the above information acquisition step and the reference land classification information according to the land classification information acquired in the above information acquisition step. based, trees discrimination program characterized by giving priority to be higher in the association degree, to execute a determining step of determining the type of trees to the computer. In the above information acquisition step, topographical information regarding the topography of the land where the forest that newly determines the type of tree is located is acquired .
In the above discrimination step, reference is made to three or more levels of association between the above combination having the reference terrain information regarding the terrain of the land where the forest has captured the reference aerial image information in the past and the type of trees in the forest. Further, claim 1 or 2 , wherein the type of the tree is determined by giving priority to the one having a higher degree of association based on the reference terrain information corresponding to the terrain information acquired in the above information acquisition step. Tree discrimination program. In the above information acquisition step, the distance information obtained by measuring the distance from the measurement point of the forest, which newly determines the type of tree, by the distance measurement sensor is acquired.
In the above discrimination step, there are three stages: the above combination having the reference distance information obtained by measuring the distance from the measurement point of the forest where the reference aerial image information was captured by the distance measuring sensor in the past, and the type of trees in the forest. It is characterized by referring to the above degree of association and, based on the reference distance information according to the distance information acquired in the above information acquisition step, prioritizing the one with the higher degree of association to determine the type of tree. The tree discrimination program according to claim 1 or 2. In the above information acquisition step, shape information regarding the shape of the tree is acquired by measuring the distance from the measurement point of the tree whose type is newly determined by the distance measuring sensor.
In the above discrimination step, the above combination having the reference shape information regarding the shape of the tree by measuring the distance from the measurement point with respect to the tree in the forest where the reference aerial image information was imaged in the past and the forest thereof. Based on the reference shape information according to the shape information acquired in the above information acquisition step, priority is given to the tree having the higher degree of association with the tree type in the above. The tree discrimination program according to claim 1 or 2, wherein the type is discriminated. In the above information acquisition step, climate information on the climate of the area where there is a forest that newly determines the type of tree is acquired .
In the above discrimination step, reference is made to three or more levels of association between the above combination having the reference climate information regarding the climate of the area where the forest has captured the reference aerial image information in the past and the type of trees in the forest. Further, claim 1 or 2 , wherein the type of the tree is determined by giving priority to the one having a higher degree of association based on the reference climate information according to the climate information acquired in the above information acquisition step. Tree discrimination program. The tree discrimination program according to any one of claims 1 to 6 , wherein in the discrimination step, the degree of association corresponding to the weighting coefficient of each output of the node of the neural network in artificial intelligence is used.

Images