JP2017163934A - Methods of calculating damage classification of pine wilt disease and apparatus for calculating damage classification of pine wilt disease - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide methods of calculating damage classification of pine wilt disease and apparatus for calculating damage classification of pine wilt disease which make the production of materials showing the damage situation of pine wilt disease in a Japanese red pine forest achievable in a short time and with high accuracy.SOLUTION: A method of calculating damage classification of pine wilt disease in which the number of Japanese red pines are calculated by the damage classification using the image data of a Japanese red pine forest produced based on spectral reflection properties of tree species included in photographed image data comprises a Japanese red pine forest tree apex imaging data production step S31 of producing imaging data of Japanese red pine forest tree apex, a Japanese red pine forest tree crown imaging data production step S32 of producing imaging data of Japanese red pine forest tree crown, a tree crown specific damage classification image data production step S33 of producing imaging data of damage classification by tree crown, a tree apex specific damage classification imaging data production step S34, a damage classification specific number calculating step S40 of calculating the number of Japanese red pine trees by a damage classification, and a damage ratio classification figure production step S50 of producing a damage ratio classification figure.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a method for calculating a damage category for pine worms and a device for calculating a damage category for pine worms.

  The damage of pine weevil is the largest damage caused by forest pests in Japan. The amount of pine wilt damage nationwide has been decreasing since its peak in 1979, but the amount of damage has increased in Nagano Prefecture and Tohoku region in high latitudes and high altitude areas, and forest damage has become serious. Due to the spread of damage, forest owners' willingness to forestry and matsutake production decline, danger due to dead trees and fallen trees, reduced road disaster prevention function, local pine forests and shrines and temples, public facilities such as parks and environmental forests Countermeasures such as decline are extremely important issues.

  The damage caused by the pine weevil and the changes in the leaves of the pinewood nematode are as follows. In the summer, it grows in the summer, the infected pine is weakened from autumn to winter, and the leaves turn from yellowish green to reddish brown and wither. From the summer of the following year, the color fades and turns brown and eventually leaves. The pine wood beetle uses dead trees to lay eggs, and when the beetle becomes a cocoon in April-June, the pine wood nematode attaches to the body of the cocoon and moves symbiotically when it emerges. The damage is spread by being transported to other pine trees.

  A countermeasure against pine weevil has been proposed by the Forestry Agency (see, for example, Non-Patent Document 1). According to Non-Patent Document 1, prefectural governors, etc., define pine forests with high public interest functions as “pine forests to be preserved” and pine forests located in the vicinity as “peripheral pine forests”, and focus on these pine forests. And comprehensive measures will be taken. In addition, as a local measure, based on the damage situation, in the pine forests to be preserved, “preventive measures” such as drugs and “extermination measures” such as felling and fumigating the damaged trees, etc. It is supposed to promote conversion.

  The above measures need to be taken by the investigator to see how the damage occurred during ground surveys, but in prefectures and municipalities where measures against forest damage caused by pine weevil are being promoted, both budgetary and personal It is not limited, and only limited pine forests to be preserved and surrounding pine forests are limited (chemical spraying and felling and fumigation treatment). The municipalities that deal with the forefront of damage have a small number of staff, and there are often no specialized technical staff, so it is not possible to grasp the state of damaged forests away from wide area forest areas and roads Have difficulty.

  On the other hand, in order to grasp the detailed distribution of pine worm damage by municipalities and to prevent the spread of the damage, it is necessary to capture and accurately control the group of infected and dead trees that have been damaged by wormwood. It is desired. However, in a situation where the budget is limited, it is limited to measures in a narrow area such as a residential neighborhood or a pine forest along the road, and accurate control and measures for a wide area on a municipal level basis have been achieved. There is no problem.

  In order to capture and accurately control infected and dead tree groups without relying on surveys by survey committee members, etc., the current status of the Pinus densiflora forest to be surveyed is ascertained using information processing technology, and pine trees in the Pinus densiflora forest. It is necessary to be able to create materials that can easily understand the damage situation of caterpillars visually. “Infected trees” refers to damaged trees of the current year, whose leaves turn yellowish green, and are not energetic, and “dead trees” are damaged trees of the previous year. This refers to red pine with brown leaves.

  On the other hand, as a technique for grasping the current state of the forest area of the predetermined range of red pine forest to be investigated by information processing technology, from the forest image information obtained based on the captured image data obtained by photographing the forest area from the sky, Techniques for grasping the current state of forests have been proposed (see, for example, Patent Document 1).

  Patent Document 1 is not limited to red pine forests, but forests related to forest areas extracted based on photographed image data obtained by photographing regions including forests to be surveyed from above and geographical information of the forests. Data on tree vertices is extracted based on area image data, and the number of trees is calculated with high accuracy.

Forestry Agency About pine weevil damage and pine weevil countermeasures [Search February 6, 2016], Internet <URL: <http://www.rinya.maff.go.jp/j/hogo/higai/matukui. html >＞

Japanese Patent No. 4858793

  In the pine worm countermeasures described in Non-Patent Document 1, the investigator walks through the red pine forest to be surveyed and looks around the state of the damage to find out the damage situation of the pine worm. There is a problem that it is necessary for the staff to prepare a document indicating the damage situation, which requires a lot of expenses such as labor costs and a lot of time for the investigation. In addition, in the method for calculating the number of trees described in Patent Document 1, it is possible to calculate the number of trees with high accuracy by using information processing technology, but it shows the damage situation of pine weevil in the red pine forest to be investigated. You cannot create materials.

  Therefore, the present invention provides a method for calculating the damage category of pine worms and the damage category of pine worms, which enables the creation of data indicating the damage status of pine worms in a red pine forest to be realized in a short time and with high accuracy at a low cost. The purpose is to provide a calculation device.

  The inventor of the present invention has an infected leaf, a dead leaf from each crown of an infected tree, a dead tree, and a healthy red pine tree (a healthy tree) that is not damaged by the insect bite in a real red pine forest. We collected healthy leaves and measured them with a spectroreflectometer that can measure the same observation wavelength as an artificial satellite or an aircraft sensor. Because there are differences in the spectral reflection characteristics of infected leaves, dead leaves, and healthy leaves, damage classification ( It was discovered that infected trees, dead trees, and healthy trees can be extracted. The inventor of the present invention paid attention to this point and completed the present invention by conducting intensive studies. The present invention comprises the following elements.

  [1] The method for calculating the damage classification of pine worms according to the present invention is created based on spectral reflectance characteristics of tree species included in photographed image data obtained by photographing an area including a red pine forest to be investigated from the sky. A method for calculating a damage category of a pine pine beetle by calculating the number of Pinus densiflora included in the Pinus densiflora forest according to the damage category using the Pinus densiflora image data, and each tree of the Pinus densiflora based on the Pinus densiflora image data A pine forest tree vertex image data creating step for extracting vertices and creating a red pine forest tree vertex image data for each tree vertex, and extracting each pine tree crown based on the pine forest image data; The pine forest canopy image data creating step for creating the red pine forest canopy image data, and the pine forest canopy image data, Spectral reflection characteristics of infected trees, dead trees and healthy trees in the area of each crown based on the spectral reflection characteristics of damaged leaves collected from red pine and healthy leaves collected from healthy red pine, The damage classification image data creation step by crown for identifying the damage classification by crown based on the spectral reflection characteristics of the created infected tree, dead tree and healthy tree, and creating the damage classification image data by crown, and the damage by the crown Based on the classification image data and the red pine forest tree vertex image data, the damage classification image data for each tree vertex is specified by identifying the damage classification for each tree vertex and creating the damage classification image data for each tree vertex; Damage zone that calculates the number of red pine contained in the whole area of the pine forest or in any area based on the image data and geographical information by vertex Based on the branch number calculation step and the damage category number calculation result, the total number of damaged trees in the entire area of the red pine forest or the infected trees and dead trees included in any area is calculated as the total area of the red pine forest or A damage rate classification diagram creation step of creating a damage rate classification diagram by calculating a damage rate obtained by dividing by the total number of red pine trees included in an arbitrary area.

  In the present invention, using red pine forest image data relating to red pine forest extracted based on the spectral reflection characteristics of the tree species included in the photographed image data obtained by photographing the region including the forest to be investigated from the sky, By implementing each of the above steps, it is possible to calculate the damage category of pine worms in the real red pine forest and to create a damage rate classification diagram. Note that the damage rate classification chart can be used as a material indicating the damage situation of pine weevil. As described above, according to the present invention, the current state of the red pine forest to be investigated can be grasped by the information processing technology. Therefore, it is possible to shorten the creation of the material indicating the pine weevil damage state in the red pine forest at a low cost. Realized in time and with high accuracy.

  In addition, since the damage rate classification diagram created in the present invention can be color-coded in the legend of the damage rate, the area where the damage is occurring can be grasped at a glance. The damage rate classification chart is an output chart required by prefectures and municipalities that are promoting countermeasures against forest damage caused by pine needles. It is effective in grasping the detailed distribution of pine weevil damage at the municipal level, and in order to prevent the spread of damage, it is necessary to plan for the prevention and proper control of the infected and dead tree groups at the forefront. It has an effective function for forest management.

  In this specification, the term “tree crown” refers to a portion composed of branches and leaves of each tree when the trees constituting the forest are viewed from above in a substantially vertical direction. If there are no other trees, the branches and leaves of one tree may be extracted as one crown, but if there are multiple trees in close proximity, the branches and leaves of multiple trees Therefore, a wide range composed of a plurality of tree branches and leaves is extracted as one crown (see FIG. 8 described later). Further, the “tree vertex” is a pointed tip portion in the tree crown, and this corresponds to the vertex portion of each tree (see FIG. 8 described later).

  [2] In the method for calculating the damage category of pine beetle of the present invention, red pine forest image data is extracted by extracting red pine forest from the photographed image data based on the spectral reflection characteristics of the tree species included in the photographed image data. A red pine forest image data creation step, and after performing the red pine forest image data creation step, for each red pine forest, the red pine forest tree vertex image data creation step and the red pine forest crown image data creation It is preferable to perform a step, the damage category image data creation step, the tree vertex-specific damage category image data creation step, the damage category number calculation step, and the damage rate category diagram creation step.

  Thereby, red pine forest image data can be created, and the above steps can be performed using the created red pine forest image data. The red pine forest image data creation step is performed by classifying the reflection luminance values included in the photographed image data obtained by photographing the region including the red pine forest to be investigated from the sky into classes. It can be automatically extracted using land cover classification by the known maximum likelihood method.

  [3] In the pine weevil damage classification calculation method of the present invention, it is preferable to use image data taken from autumn to winter as the photographed image data.

  In this way, by using image data taken from autumn to winter, when deciduous trees are mixed in the red pine forest to be investigated, the deciduous trees are deciduous. It becomes easy to classify trees and healthy trees, and damage classification is possible with high accuracy.

  [4] In the method for calculating the damage classification of pine worms according to the present invention, the classification of the infected tree and the dead tree among the infected tree, the dead tree and the healthy tree is affected by the pine worm. The infected tree is reflected in the red region of visible light compared to the dead tree and the healthy tree. It is distinguished from the dead tree and the healthy tree by a high rate, and the dead tree has a lower reflectance than the infected tree and the healthy tree in the near infrared region, and thus the infected tree and the healthy tree. It is preferable to categorize.

  In this way, by classifying infected trees, dead trees, and healthy trees based on the spectral reflection characteristics of damaged leaves and healthy leaves, it is possible to accurately classify infected trees, dead trees, and healthy trees. In particular, it is possible to accurately distinguish between infected trees and dead trees. The “damaged leaf” refers to both an infected leaf collected from an infected tree and a dead leaf collected from a dead tree.

  [5] In the method for calculating the damage category of a pine pine beetle according to the present invention, the step of creating the pine forest tree vertex image data uses a local maximum value filter that automatically extracts a point having a maximum luminance value. It is preferable to extract vertices.

  As a result, the tree vertices of Pinus densiflora can be extracted with high accuracy. Note that the pinus of the red pine is the pointed portion of the crown that strongly reflects sunlight, and since the reflection value is maximum in the red pine forest image data, it is automatically extracted by local maximum value filtering. Can do. In addition, by performing such processing, tree vertices of various crowns can be extracted from the entire forest area with high accuracy.

  [6] In the pine beetle damage classification calculation method according to the present invention, it is preferable that the step of creating the red pine forest canopy image data extracts the crown using the fact that the edge of the crown is darker than the crown portion.

  Thereby, a canopy (red pine canopy) can be extracted with high accuracy. In addition, the extraction of the tree crown is performed by utilizing the fact that the edge of the tree crown (also referred to as the tree crown edge) is darker than the tree crown part in the red pine forest image data, and this is represented by the valley following algorithm (Va11ey fo11owing a1gorithm) and automatic mask processing, tree crowns can be extracted accurately from Japanese red pine forest image data. Note that “the edge of the crown” and “the crown portion” will be described later with reference to FIG.

  [7] In the method for calculating the damage category of a pine beetle according to the present invention, the damage category image data creation step for each crown includes the three types of the damage category as an infected tree, a dead tree, and a healthy tree. When infected trees, dead trees, and healthy trees are extracted from the above, the occupied areas of the infected trees, dead trees, and healthy trees extracted from the respective crowns are calculated, and the calculated infected trees, dead trees, and healthy trees are calculated. Of these occupying areas, it is preferable to regard a damage category having a large occupancy area as a damage category of the crown.

  Thereby, one crown can be classified as one damage category. For example, in a certain crown, as a result of calculating each occupation area of the infected tree, dead tree and healthy tree, in the crown, if the occupation area of the infected tree is larger than each occupation area of the dead tree and healthy tree, The damage classification of the tree crown is to be regarded as an “infected tree”. In this way, one tree crown is classified as one damage category, so that it is possible to quickly grasp the infected tree and select the pine forest and the pest control area to be protected for each municipality in a wide area and in detail. it can.

  [8] In the method for calculating a damage category of a pine beetle according to the present invention, the step of creating the damage category image data for each vertex of the tree uses the top image data of the red pine forest tree as the damage category for each crown based on the geographical information. It is preferable to specify the damage category for each tree vertex by superimposing the image data.

  Thus, by specifying the damage classification of each tree vertex, it is possible to specify the damage classification for each individual tree. In addition, it is possible to create a distribution image of individual tree positions by color-coding the damage classification on the photographed image obtained by photographing the survey area from above, so it is possible to visually grasp the state of forest damage. The damage category refers to infected trees, dead trees, and healthy trees.

  [9] In the method for calculating the damage category of a pine beetle according to the present invention, the step of calculating the number of damage categories is based on the damage category image data classified by tree vertices in the entire region or an arbitrary region of the red pine forest. It is preferable to extract a subcompartment section from the boundary data of the forest owner defined as a subcompartment based on the target information, and to calculate the number of trees for each of the damage categories in the extracted subcompartment.

  As a result, the number of trees can be calculated for each subdivision, for each damage category, that is, for each infected tree, dead tree, and healthy tree.

[10] The pine beetle damage classification calculating apparatus according to the present invention is created based on spectral reflection characteristics of tree species included in photographed image data obtained by photographing an area including a red pine forest to be investigated from the sky. An apparatus for calculating the damage category of a pine weevil that uses the red pine forest image data to calculate the number of red pine trees contained in the red pine forest according to the damage category, wherein each tree of the red pine tree is based on the red pine forest image data. A pine forest tree vertex image data creation unit that extracts vertices and creates red pine forest tree vertex image data relating to each tree vertex, and extracts each of the red pine tree crowns based on the red pine forest image data, Japanese pine forest canopy image data creation unit for creating red pine forest canopy image data, and the red pine forest canopy image data Based on the spectral reflection characteristics of the damaged leaves collected from the tree and the healthy leaves collected from the healthy red pine, the spectral reflection characteristics of the infected tree, the dead tree and the healthy tree in the area of each crown are created and created. Damage classification image data creation unit for identifying damage classification by crown based on the spectral reflection characteristics of infected infected trees, dead trees and healthy trees, and creating damage classification image data by crown, and said damage classification by crown Based on the image data and the red pine forest tree vertex image data, the damage classification image data creating unit for each tree vertex identifying the damage category for each tree vertex and creating the damage category image data for each tree vertex, and the tree vertex A damage category number calculating unit for calculating the number of red pine trees included in the entire area of the red pine forest or an arbitrary area based on the different damage category image data and geographical information; Based on the result of calculating the number of damage categories, the total number of damaged trees in the entire pine forest or in any area of the red pine forest is calculated as the total number of damaged trees in the entire pine forest or in any area. A damage rate classification diagram creation unit that calculates the damage rate obtained by dividing by the total number of
It is characterized by having.

  According to the present invention, the same effect as the method for calculating the damage category of pine worms described in [1] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [11] In the pine beetle damage classification calculating apparatus according to the present invention, red pine forest image data is created by extracting red pine forest from the photographed image data based on the spectral reflection characteristics of the tree species. And generating the red pine forest tree vertex image data by the red pine forest tree vertex image data creating unit for each of the red pine forests after the red pine forest image data is created by the red pine forest image data creation unit. Creation of the red pine forest canopy image data by the red pine forest canopy image data creation unit, creation of the damage classification image data by the crown by the crown specific damage classification image data creation unit, and damage classification image data by the tree apex Creation of the damage classification image data for each tree vertex by the creation section and the damage section by the damage classification number calculation section And the number calculation, it is preferred to carry out the creation of the damage rate levels view by the damage rate levels diagram creating unit.

  Thereby, the effect similar to the damage classification calculation method of the pine weevil described in the above [2] is obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [12] In the pine weevil damage category calculating apparatus according to the present invention, it is preferable that the photographed image data is image data photographed from autumn to winter.

  Thereby, the same effect as the method for calculating the damage category of pine worms described in [3] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [13] In the pine weevil damage classification calculating apparatus of the present invention, the classification of the infected tree and the dead tree among the infected tree, the dead tree, and the healthy tree is affected by the pine worm. The infected tree is reflected in the red region of visible light compared to the dead tree and the healthy tree. It is distinguished from the dead tree and the healthy tree by a high rate, and the dead tree has a lower reflectance than the infected tree and the healthy tree in the near infrared region, and thus the infected tree and the healthy tree. It is preferable to categorize.

  Thereby, the same effect as the method for calculating the damage category of pine worms described in [4] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [14] In the pine beetle damage classification calculating apparatus according to the present invention, the red pine forest tree vertex image data creating unit uses a local maximum value filter that automatically extracts a point having a maximum luminance value to determine a tree vertex. It is preferable to extract.

  Thereby, the same effect as the method for calculating the damage category of pine worms described in [5] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [15] In the pine beetle damage classification calculating apparatus according to the present invention, the red pine forest canopy image data creation unit is performed by utilizing the fact that the edge of the canopy is darker than the canopy portion. It is preferable that the extraction is automatically performed by the valley following algorithm and the automatic mask process.

  Thereby, the same effect as the method for calculating the damage category of pine worms described in [6] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [16] In the pine beetle damage classification calculating apparatus according to the present invention, the crown-specific damage classification image data creating step includes setting the damage classification as an infected tree, a dead tree and a healthy tree, and each crown of the red pine. When infected trees, dead trees, and healthy trees are extracted from the above, the occupied areas of the infected trees, dead trees, and healthy trees extracted from the respective crowns are calculated, and the calculated infected trees, dead trees, and healthy trees are calculated. Of these occupying areas, it is preferable to regard a damage category having a large occupancy area as a damage category of the crown.

  Thereby, the same effect as the method for calculating the damage category of pine worms described in [7] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [17] In the pine beetle damage category calculating apparatus according to the present invention, the tree-based damage category image data creating step includes converting the red pine forest tree vertex image data into the tree-based damage category based on the geographical information. It is preferable to specify the damage category for each tree vertex by superimposing the image data.

  Thereby, the same effect as the method for calculating the damage category of pine worms described in [8] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

  [18] In the pine beetle damage category calculating apparatus according to the present invention, the damage category number calculating unit is configured to perform the geography in the entire area of the red pine forest or an arbitrary area based on the damage category image data classified by tree vertices. It is preferable to extract a subcompartment section from the boundary data of the forest owner defined as a subcompartment based on the target information, and to calculate the number of trees for each of the damage categories in the extracted subcompartment.

  Thereby, the same effect as the method for calculating the damage category of pine worms described in [9] above can be obtained. Note that the pine worm damage category calculating apparatus of the present invention has the functions of each of the above-described components included in the pine worm damage category calculating device installed as a computer program. By giving this data, the function of each component is executed on the software of the computer.

It is a figure which shows the structure of the damage classification calculation apparatus 100 of the pine weevil which concerns on embodiment. It is a flowchart explaining the damage classification calculation method of the pine weevil which concerns on embodiment. It is a figure which shows the group of the infected tree and the group of dead trees in a pine stake insect damage forest. It is a figure which shows an example of the equipment used when measuring the healthy leaf, an infected leaf, and a dead leaf by a spectral reflectometer, and a measurement object leaf. It is a figure which shows the spectral reflection characteristic of a healthy leaf, an infected leaf, and a dead leaf. It is a figure which shows the picked-up image displayed on the display. It is a figure which shows the pine forest image displayed on the display. It is a figure explaining the tree vertex and tree crown of a tree. It is a figure which shows the pine forest tree vertex image data extracted from the pine forest image displayed on the display with a white point. It is a figure which shows the pine forest canopy image displayed on the display. It is a figure which shows the damage classification image classified by tree crown displayed on the display. It is a figure which shows the damage classification image classified by tree | vertice displayed on the display. It is a figure which shows the damage classification calculation image of the pine weevil displayed on the display. It is a figure which shows an example of the damage condition database displayed on the display. It is a figure which shows the damage rate division figure displayed on the display.

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

  FIG. 1 is a diagram illustrating a configuration of a pine worm damage category calculation apparatus 100 according to the embodiment. As shown in FIG. 1, a pine beetle damage category calculating apparatus 100 according to the embodiment includes a photographed image data input unit 10, a red pine forest image data creation unit 20, a red pine forest tree vertex image data creation unit 31, There is a red pine forest canopy image data creation unit 32, a damage classification image data creation unit 33 by crown, a damage classification image data creation unit 34 by tree apex, a damage number calculation unit 40, and a damage rate classification diagram creation unit 50 doing. The red pine forest tree vertex image data creation unit 31, the red pine forest crown image data creation unit 32, the damage classification image data creation unit 33 by crown, and the damage classification image data creation unit 34 by tree vertex provide a red pine forest tree. The damage classification system 30 according to vertex is comprised.

  The photographed image data input unit 10 has a function of inputting photographed image data obtained by photographing a region including a red pine forest to be surveyed (referred to as a survey target region) from above.

  The red pine forest image data creating unit 20 has a function of creating red pine forest image data by extracting red pine forest from the photographed image data based on the spectral reflection characteristics of the tree species included in the photographed image data.

  The red pine forest tree vertex image data creation unit 31 has a function of extracting each tree vertices of red pine based on the red pine forest image data and creating red pine forest tree vertex image data regarding each tree vertex. Specifically, the red pine forest tree vertex image data creation unit 31 extracts the red pine tree vertices using a local maximum value filter that automatically extracts a point having a maximum luminance value.

  The red pine forest canopy image data creation unit 32 has a function of extracting each red pine canopy based on the red pine forest image data and creating red pine forest canopy image data relating to each canopy. Specifically, the red pine forest canopy image data creation unit 32 extracts the crown using the fact that the edge (crown edge) of the crown is darker than the crown portion.

  The damage classification image data creation unit 33 according to canopy includes a red pine forest canopy image data, spectral reflection characteristics of damaged leaves collected from red pine that have been damaged by pine needles, and healthy leaves collected from healthy red pine. Based on the above, the spectral reflection characteristics of infected trees, dead trees and healthy trees in the area of each canopy are created, and damage classification (infected trees) by crown based on the spectral reflection characteristics of the infected trees, dead trees and healthy trees created. , Withered trees and healthy trees), and has a function of creating damage classification image data by crown.

  Specifically, the damage classification image data creation unit 33 classified by tree crown uses three types of damage categories of pine worms: infected tree, dead tree, and healthy tree. If extracted, calculate the occupied area of each infected tree, dead tree, and healthy tree extracted from each crown, and among the calculated occupied area of the infected tree, dead tree, and healthy tree, the occupied area is large. By considering the damage classification as the damage classification of the crown, the damage classification image data for each crown is created.

  The damage classification image data creation unit 34 classified by tree vertices identifies the damage classification (infected tree, dead tree and healthy tree) for each tree vertex based on the damage classification image data by tree crown and the red pine forest tree vertex image data, It has a function to create tree-based damage classification image data. Specifically, the tree-based damage classification image data creation unit 34 superimposes the red pine forest tree vertex image data on the tree-specific damage classification image data based on the geographical information, thereby determining the damage classification for each tree vertex. The damage classification image data classified by tree vertices is created. In this case, the geographical information refers to the entire area of the forest area or an arbitrary area registered in the forest management ledger or the like maintained in the prefecture. The geographic information is registered in the geographic information system (referred to as GIS) 60 shown in FIG.

  The damage classification number calculation unit 40 determines the number of red pine trees contained in the whole or arbitrary area of the red pine forest based on the damage classification image data for each vertex and geographical information. It has a function to calculate the number of damage categories by calculating each tree. Specifically, the damage category number calculation section uses three types of pine worm damage categories as infected trees, dead trees and healthy trees, and pine forests for each damage category (infected trees, dead trees and healthy trees). In all or any area of the above, subdivisions are extracted based on the boundary data of forest owners defined as subdivisions, and the number of trees is calculated for each damage category related to the extracted subdivisions.

  Based on the result of calculating the number of damage categories, the damage rate category map creation unit 50 calculates the total number of damaged trees in the entire area of the red pine forest by adding the total number of infected trees and dead trees contained in any area of the red pine forest. Or, it has a function to create a damage rate classification chart by calculating the damage rate obtained by dividing by the total number of red pine contained in an arbitrary area.

  In preparing the damage rate classification diagram, the total number of damage categories (infected trees, dead trees and healthy trees) calculated by the damage category number calculation unit 40, and the total number of infected trees and dead trees Create a database of the number of trees, total number (total), damage rate, etc., based on the contents of the database, create a forest survey database (in this case, forest management ledgers etc. maintained in prefectures, etc.) (For example, see FIG. 14) or the geographic information system (G1S). Then, a damage rate classification diagram is created based on the tabled contents and the contents registered in the forest survey book database or G1S.

  The pine worm damage category calculating apparatus 100 shown in FIG. 1 has the functions of the components included in the pine worm damage category calculating apparatus 100 installed as a computer program. By giving predetermined data to each component, the function of each component is executed on the software of the computer.

  FIG. 2 is a flowchart for explaining a pine worm damage category calculation process in the pine worm damage category calculation method according to the embodiment. As shown in FIG. 2, the pine weevil damage classification calculation procedure is as follows. The photographed image (color image) corresponding to the photographed image data obtained by photographing the area including the red pine forest to be investigated from the sky is shown in FIG. Display is performed (step S10). Also, red pine forest image data relating to red pine forest extracted based on the spectral reflection characteristics of the tree species included in the photographed image data is created (step S20). Then, based on the red pine forest image data, each of the red pine tree vertices is extracted, and the red pine forest tree vertex image data relating to the respective tree vertices is created (step S31). Tree crowns are extracted, and pine forest crown image data relating to each tree crown is created (step S32).

  After that, based on the image data of the red pine forest canopy and the spectral reflection characteristics of the damaged leaves collected from the red pine and the healthy leaves collected from the healthy red pine, Spectral reflection characteristics of the infected tree, dead tree, and healthy tree are created, the damage classification (infected tree, dead tree, and healthy tree) is specified for each tree crown, and the damage classification image data for each tree crown is created (step S33).

  Then, based on the damage classification image data by tree crown and the red pine forest tree vertex image data, the damage classification (infected tree, dead tree and healthy tree) is identified by tree vertex and the damage classification image data by tree vertex is created ( Step S34). After that, based on the damage classification image data by tree vertex and geographical information, the number of red pine contained in the whole area of the pine forest or any area is calculated for each damage classification (infected tree, dead tree and healthy tree). The damage category number calculation data is created (step S40).

  After that, based on the data for calculating the number of damage categories, the total number of damaged trees and infected trees in all areas of the pine forest or any area is included in all areas or any area of the pine forest. The damage rate obtained by dividing by the total number of red pine is calculated and a damage rate classification diagram is created (step S50).

  FIG. 3 is a diagram showing a group of infected trees and a group of dead trees in a pine scabies damaged forest. In FIG. 3, a region 1 indicates a group of infected trees, and a region 2 indicates a group of dead trees. Since FIG. 3 is a monochrome image, from FIG. 3, it is difficult to read the group of infected trees (region 1) and the group of dead trees (region 2) by color. On the original color image of 3, the infected tree population (region 1) is shown in a color that is changing from dull green to brown, and the dead tree population (region 2) is shown in brown overall Therefore, the group of infected trees (area 1) and the group of dead trees (area 2) can be easily read on the captured image (color image) displayed on the display.

  FIG. 4 is a diagram illustrating an example of equipment and measurement target leaves used when measuring healthy leaves, infected leaves, and dead leaves with a spectroreflectometer. As equipment used for the measurement, a spectral reflectometer 3 and a standard white plate 4 are used as shown in FIG. Further, in FIG. 4, the healthy leaf 5 is shown as the measurement target leaf, but the infected leaf and the dead leaf (these illustrations are omitted) are also used as the measurement countermeasure leaf. Spectral reflection characteristics represent the intensity of reflection of the electromagnetic wave radiated from the sun for each wavelength from the visible range to the near infrared range. The reflection intensity of the standard white plate 4 is 100, and the ratio is the spectral reflectance. And In addition, healthy leaves are leaves collected from healthy trees, infected leaves are leaves collected from infected trees, and dead leaves are collected leaves. Since FIG. 4 is a monochrome image, it is difficult to read from FIG. 4 that the healthy leaves of red pine are green, but in reality, the healthy leaves are dark green.

  FIG. 5 is a diagram showing the spectral reflection characteristics of damaged leaves and healthy leaves of a pine weevil. Note that the damaged leaves of pine worms refer to infected leaves and dead leaves. Spectral reflection characteristics of damaged and healthy leaves of pine weevil are from infected trees of red pine trees in the area where they are actually infected (current year damaged trees: pine trees with discolored leaves that turn yellow-green), dead trees (previous year) Damaged trees: pine trees with brown leaves), infected leaves, dead leaves, and healthy leaves from each crown of healthy trees, and measured the same observation wavelength as satellites and aircraft sensors as shown in Fig. 4. As a result of measurement with the spectroreflectometer 3, the spectral reflection characteristics as shown in FIG. 5 were obtained.

  In FIG. 5, a curved line L1 indicated by a thick solid line indicates the spectral reflection characteristic of healthy leaves, a curved line L2 indicated by a broken line indicates the spectral reflection characteristics of dead leaves, and a curved line L3 indicated by a white outline indicates the spectral reflection characteristics of infected leaves. Is shown. The spectral reflection characteristic represents the intensity of reflection of each wavelength of the electromagnetic wave radiated from the sun from the visible range to the near infrared range. The standard white plate is defined as 100, and the ratio is defined as the reflectance.

  As shown in FIG. 5, since there is a difference in the spectral reflection characteristics of the three types of leaves (infected leaves, dead leaves, and healthy leaves), it was found that infected leaves, dead leaves, and healthy leaves can be classified. That is, as shown in FIG. 5, according to the spectral reflection characteristics (curve L2) of the dead leaves, the dead leaves have a very low reflectance in the near-infrared region (770 nm to 830 nm) 7. Can be easily segmented. On the other hand, according to the spectral reflection characteristics of the infected leaves (curve L3), the infected leaves are inferior in photosynthetic ability due to a decrease in chlorophyll and water in the visible red region (670 nm to 710 nm) 6, thereby reducing the absorption of electromagnetic waves. Reflectivity is higher than healthy and dead leaves. For this reason, infected leaves can be distinguished from dead leaves and healthy leaves.

  As shown in FIG. 5, this result shows that dead leaves are infected and healthy in the near-infrared region of 4 channels (ch) at the observation wavelengths of ordinary satellites (blue, green, red, near-infrared region). As shown in Fig. 5, the latest 8-channel satellite can be divided into 3 channels and multiple wavelengths in the red region of the visible region. Thus, it can be classified with high accuracy.

  Thus, by classifying infected leaves, dead leaves, and healthy leaves depending on the spectral reflection characteristics of the three types of leaves (infected leaves, dead leaves, and healthy leaves), infected trees, dead trees, and healthy trees Classification can be performed with high accuracy. In particular, it is possible to accurately classify infected trees and dead trees.

  FIG. 6 is a diagram illustrating a captured image displayed on the display. The photographed image shown in FIG. 6 is an image corresponding to photographed image data obtained by photographing an area including a red pine forest to be investigated from the sky. In the photographed image, the area shown in black as a whole is a forest area, and the area shown in white is a village, a field, and the like. Also in the forest area, there are areas of red pine forest and other trees.

  Since FIG. 6 is a monochrome image, it is difficult to read the red pine forest area and the other tree areas separately by color from FIG. 6, but in practice it is the origin of FIG. On the color image, the red pine forest is shown in purple, and the forest area of other trees is shown in green. Therefore, on the photographed image (color image) displayed on the display, the red pine forest and other trees Can be easily distinguished.

  FIG. 7 shows a red pine forest image displayed on the display. The red pine forest image shown in FIG. 7 is an image corresponding to the red pine forest image data created by the red pine forest image data creation unit 20. The red pine forest image is obtained by expressing the red pine forest area in a specific color (in orange) in order to extract the red pine forest area in the photographed image shown in FIG. Since FIG. 7 is a monochrome image, it is difficult to read red pine forest as orange from FIG. 7, but actually, the red pine forest is shown in orange on the original color image of FIG. Therefore, the red pine forest can be easily read on the photographed image (color image) displayed on the display.

  FIG. 8 is a diagram illustrating tree vertices and tree crowns. FIG. 8A is a diagram schematically illustrating a case where the forest is viewed from the side (almost horizontal direction) on the ground, and FIG. 8B is a diagram schematically illustrating a case where the forest is viewed from the sky almost vertically. FIG. In FIG. 8, the tree apex is represented by “Pe”, and the tree crown is represented by “Ti”. When there are a plurality of trees close to each other, as seen from above, the crowns of the three trees overlap each other, so that as shown in FIG. 8B, one crown (in FIG. 8B) Taken as shown by the solid line). In each crown Ti, the contour line TiL of the crown Ti is defined as “the edge of the crown (crown edge)”, and the inside of the contour line TiL (the portion surrounded by the contour line TiL) is defined as the “crown portion”. . In addition, the edge of the crown and the inside of the crown are collectively referred to as a “crown area”.

  FIG. 9 is a diagram showing a red pine forest tree vertex image extracted from the red pine forest image displayed on the display. The red pine forest tree vertex image shown in FIG. 9 is an image corresponding to the red pine forest tree vertex image data created by the red pine forest tree vertex image data creation unit 31. The peak image of the red pine forest tree reflects the sunlight strongly from the image that has been processed as described in FIG. 7 (process for extracting the red pine forest), and each peak of the maximum value in the crown of the red pine forest Can be obtained automatically by local maximum filtering. By performing such processing, tree vertices difficult to read with the naked eye can be extracted with high accuracy. FIG. 9 is an enlarged view of a predetermined area of the image shown in FIG.

  Since FIG. 9 is a monochrome image, the red pine forest is shown in black in FIG. 9, and the tree vertices of each red pine are shown as white dots. For this reason, from FIG. 9, it is difficult to read the pine forest area and other tree areas as different colors, but in reality, the pine forest is a specific area on the original color image of FIG. In FIG. 9, it is shown in purple (unlike in FIG. 7), and the forest area by other trees is shown in green. Therefore, on the photographed image (color image) displayed on the display, red pine forest and others Can be easily distinguished from other trees.

  FIG. 10 is a diagram showing a red pine forest canopy image displayed on the display. The red pine forest canopy image shown in FIG. 10 is an image corresponding to the red pine forest canopy image data created by the red pine forest canopy image data creation unit 32. The red pine forest canopy image shown in FIG. 10 is similar to the red pine forest tree apex image shown in FIG. 9. The red pine forest is shown in black, and the small areas surrounded by white represent the respective crowns.

  Since FIG. 10 is a monochrome image, it is difficult to read the red pine forest region and the other tree region from each other according to color from FIG. 10, but actually, the color image that is the original of FIG. Above, the red pine forest is shown in purple as in FIG. 9, and the forest area of the other trees is shown in green. Therefore, on the photographed image (color image) displayed on the display, It can be easily distinguished from other trees.

  By the way, the extraction of the canopy is performed by utilizing the fact that the edge of the canopy is darker than the canopy portion, and the technique for extracting the canopy in this way is possible by using a well-known bray following algorithm and automatic mask processing For example, the following publicly known document 1 is known. In the known document 1, tree crown extraction is obtained using a valley following algorithm.

  Known Document 1: Masato Katoh, Francois Gougeon, Don Leckie “Application of high-resolution airborne data using individual tree crowns in Japanese conifer plantations”, Journal of Forestry Research 14 (1), 2009, P, 10-19

  FIG. 11 is a diagram showing a damage classification image by tree crown displayed on the display. The damage classification image shown in FIG. 11 is an image corresponding to the damage classification image data for each crown created by the damage classification image data creation unit 33 for each crown.

  In the damage classification image by crown shown in FIG. 11, there are a plurality of regions 8 where the infected tree group exists and a region 9 where the dead tree group exists, but FIG. 11 is a monochrome image. Therefore, from FIG. 11, it is difficult to read the area 8 where the infected tree population exists and the area 9 where the dead tree population exist as different colors. However, in practice, in the original color image in FIG. 11, it is easy to read the region 8 where the infected tree population exists and the region 9 where the dead tree population exist as different colors. In this case, in the original color image of FIG. 11, the area 8 where the infected tree group exists is shown in light brown, and the area 9 where the dead tree group exists is shown in dark brown. On the photographed image (color image) displayed above, it is possible to easily read the region 8 where the infected tree group exists and the region 9 where the dead tree group exists.

  FIG. 12 is a diagram showing a damage classification image classified by tree vertices displayed on the display. The damage classification image classified by tree vertices shown in FIG. 12 is an image corresponding to the damage classification image data classified by tree vertices created by the damage classification image data creation part 34 classified by tree vertices.

  In FIG. 12, the tree vertices of each tree (Pinus densiflora) have X and Y coordinates of position information for each Pinus densiflora. For this reason, on-site confirmation of infected trees, dead trees and healthy trees can be performed by GPS (Global Positioning System).

  Since FIG. 12 is a monochrome image, an infected tree has a tree vertex shown in gray, a dead tree has a tree vertex shown in black, and a healthy tree has a tree vertex shown in white. For this reason, it is difficult to clearly distinguish an infected tree, a dead tree, and a healthy tree from FIG. However, it is easy to read an infected tree, a dead tree, and a healthy tree on the original color image of FIG. In this case, in the original color image of FIG. 12, the infected tree is shown with a light brown tree vertex, the dead tree is shown with a red tree vertex, and a healthy tree is shown with a white tree vertex. On the captured image (color image) displayed on the display, the infected tree, the dead tree, and the healthy tree can be easily read.

  FIG. 13 is a diagram showing an image of calculating the number of damage categories of pine worms displayed on the display. Note that FIG. 13 shows a forest area in a wider range than the forest area shown in FIG. The damage category number calculation unit 40 calculates the number of damage categories based on the damage category number calculation image shown in FIG.

  As described above, the damage category number calculation unit 40 determines the number of red pine trees contained in the whole area of the pine forest or an arbitrary area based on the damage category image data classified by tree vertices and geographical information. , Dead and healthy trees). In FIG. 13, a subcompartment section (a section surrounded by a white line in FIG. 13) is extracted based on the boundary data of the forest owner defined as a subcompartment, and a tree is formed for each damage category related to the extracted subcompartment section. Calculate the number of

  Since FIG. 13 is a monochrome image, it is difficult to read an infected tree, a dead tree, and a healthy tree as different colors from FIG. 13, but actually, on the color image that is the original of FIG. It is easy to read infected trees, dead trees and healthy trees. In this case, in the original color image of FIG. 13, the infected tree is shown in light brown, the dead tree is shown in red, and the healthy tree is shown in white. On (color image), it is possible to easily read infected trees, dead trees, and healthy trees.

  FIG. 14 is a diagram illustrating an example of a damage status database displayed on the display. For example, the number of infected trees, dead trees and healthy trees calculated by the damage classification number calculation unit 40, the total number of damaged trees and the total number of damaged trees, the total number of damaged trees, the damage rate, etc. Record in association with each small group in the forest group. The damage rate can be obtained by dividing the total number of damaged trees by the total number (total) of infected trees and dead trees in each subdivision.

  Then, based on the contents of the damage situation database as shown in FIG. 14, it can be created, forest survey database (in this case, a database such as a forest management ledger maintained in prefectures) or a geographic information system ( G1S). This makes it possible to grasp the number of trees by damage category in the entire area of the pine forest or an arbitrary area, which is effective in diagnosing and reporting the damage situation.

  FIG. 15 is a diagram showing a damage rate classification diagram displayed on the display. The damage rate classification diagram is an output diagram required in prefectures and municipalities that are promoting forest damage countermeasures due to pine weevil. The damage rate classification chart can be used as a material for easily understanding the state of damage of pine weevil in a red pine forest. In addition, in Fig. 15, the legend indicating the damage rate is color-coded every 5 to 10%, and it is possible to grasp at a glance which forest area and sub-compartment is damaged, and it can be used for measures against pine weevil damage. it can.

  Since FIG. 15 is a monochrome image, it is difficult to read the damage rate for each color from FIG. 15, but actually, the damage rate is displayed for each color on the original color image of FIG. It is easy to read as. For example, the area shown in dark green has a damage rate of 0%, the area shown in yellow green has a damage rate of 6% to 10%, and the area shown in dark red is damaged. The rate is 31% to 62%.

  By using this damage rate classification chart as a material to show the damage situation of pine weevil in Pinus densiflora forest, forest managers and engineers can grasp the detailed distribution of pine weevil damage by municipality. In order to prevent the spread of damage, it will be effective for planning and forest management to capture the group of infected and dead trees at the forefront and to control them accurately. Therefore, it is no longer necessary for the investigator to walk through the forest and actually look at the damage as in the past, so it is no longer necessary to grasp the damage situation of the pine weevil, thus reducing the labor cost required for the investigation. In addition, the time required for the survey can be shortened, and the damage classification of pine worms can be calculated with high accuracy and at a low cost in a short time.

  DESCRIPTION OF SYMBOLS 1 ... Area | region where the group of infected trees exists, 2 ... Area | region where the group of dead trees exist, 3 ... Branch radiometer, 4 ... Standard white board, 5 ... It becomes a measuring object Leaf (healthy leaf), 10 ... photographed image data input unit, 20 ... red pine forest image data creation unit, 30 ... damage classification system by red pine forest tree vertices, 31 ... red pine forest tree vertex image data Creation part, 32 ... Red pine forest canopy image data creation part, 33 ... Damage classification image data creation part by crown, 34 ... Damage classification image data creation part by tree apex, 40 ... Damage classification number calculation Part, 50 ... damage rate classification diagram creation part, 60 ... Geographic Information System (GIS), Pe ... tree vertex, Ti ... tree crown

Claims (18)

Using the red pine forest image data created based on the spectral reflection characteristics of the tree species included in the photographed image data obtained by photographing the region including the red pine forest to be surveyed from above, the red pine forest included in the red pine forest is used. Is a method for calculating the damage category of pine beetles,
Based on the red pine forest image data, each of the red pine tree vertices is extracted, and a red pine forest tree vertex image data creating step for creating red pine forest tree vertex image data relating to each tree vertex;
Based on the red pine forest image data, each of the red pine canopies is extracted, and a red pine forest canopy image data creation step for creating a red pine forest canopy image data for each of the canopies,
Based on the image data of the red pine forest canopy and the spectral reflection characteristics of the damaged leaves collected from the red pine and the healthy leaves collected from the healthy red pine, Create spectral reflection characteristics of infected, dead, and healthy trees, identify damage classifications by tree crown based on the spectral reflection characteristics of the infected, dead, and healthy trees, and generate damage classification image data by crown. Step of creating damage classification image data by tree crown,
Based on the damage classification image data by tree crown and the pine forest tree vertex image data, the damage classification image data creation step by tree vertex which identifies the damage classification by tree vertex and creates damage classification image data by tree vertex When,
Based on the damage classification image data by tree vertex and geographical information, the number of damage classification number calculating step for calculating the number of red pine included in the whole area or arbitrary area of the red pine forest for each damage classification,
Based on the result of calculating the number of damage categories, the total number of damaged trees and infected trees in all areas or arbitrary areas of the red pine forest is included in all areas or arbitrary areas of the red pine forest. A damage rate classification diagram creation step of calculating a damage rate classification diagram by calculating the damage rate obtained by dividing by the total number of red pine,
A method for calculating a damage classification of a pine weevil characterized by having In the method for calculating the damage category of a pine weevil according to claim 1,
And further comprising a red pine forest image data creating step for creating red pine forest image data by extracting red pine forest from the photographed image data based on the spectral reflection characteristics of the tree species included in the photographed image data, After performing the image data creation step, for each of the red pine forests, the red pine forest tree vertex image data creation step, the red pine forest crown image data creation step, the damage classification image data creation step, and the damage by tree vertex A method for calculating a damage category of pine worms, comprising performing a step of creating a category image data, a step of calculating the number of damage categories, and a step of creating a damage rate category diagram. In the method for calculating the damage category of the pine beetle according to claim 1 or 2,
A method for calculating the damage classification of a pine weevil, wherein the photographed image data is image data photographed from autumn to winter. In the method for calculating the damage classification of pine beetles according to any one of claims 1 to 3,
Among the infected tree, the dead tree and the healthy tree, the infected tree and the dead tree are classified into a damaged leaf collected from the red pine damaged by the pine weevil and a healthy tree collected from the healthy red pine. Based on the spectral reflection characteristics of the leaves,
The infected tree has a higher reflectance than the dead tree and the healthy tree in the visible red region, and is classified as the dead tree and the healthy tree,
The dead tree has a lower reflectance than the infected tree and the healthy tree in the near-infrared region, and is classified from the infected tree and the healthy tree.
This is a method for calculating the damage classification of pine weevil. In the method for calculating the damage classification of a pine beetle according to any one of claims 1 to 4,
The pine pine forest tree vertex image data creating step includes extracting a pine tree tree vertex using a local maximum filter that automatically extracts a point having a maximum luminance value. . In the method for calculating the damage classification of a pine beetle according to any one of claims 1 to 5,
A method for calculating a damage classification of a pine weevil, wherein the step of creating the image data of a red pine forest canopy extracts the crown using the fact that the edge of the crown is darker than the crown portion. In the method for calculating the damage classification of a pine beetle according to any one of claims 1 to 6,
The damage classification image data creation step according to the tree crown includes the damage classification as an infected tree, a dead tree, and a healthy tree, and when the infected tree, the dead tree, and the healthy tree are extracted from each crown of the red pine, Calculate the occupation area of each infected tree, dead tree, and healthy tree extracted from the tree canopy, and select the damage category with the largest occupation area of the calculated infected tree, dead tree, and healthy tree. A method for calculating the damage category of a pine beetle characterized by being regarded as a damage category. In the method for calculating the damage classification of a pine beetle according to any one of claims 1 to 7,
The damage classification image data creation step for each tree vertex identifies the damage classification for each tree vertex based on the geographical information by superimposing the vertex image data for the red pine forest tree on the damage classification image data for each tree crown A method for calculating the damage classification of a pine weevil characterized by: In the method for calculating the damage classification of a pine beetle according to any one of claims 1 to 8,
The step of calculating the number of damage categories is based on the submarine data from the forest owner's boundary data defined by the geographical information in the entire area of the red pine forest or an arbitrary area based on the damage classification image data classified by tree vertices. And calculating the number of trees for each of the damage categories in the extracted subcompartment. Using the red pine forest image data created based on the spectral reflection characteristics of the tree species included in the photographed image data obtained by photographing the region including the red pine forest to be surveyed from above, the red pine forest included in the red pine forest is used. Is a pine weevil damage category calculation device that calculates the number of
Based on the red pine forest image data, each of the red pine tree vertices is extracted, and a red pine forest tree vertex image data creation unit for creating red pine forest tree vertex image data relating to each tree vertex;
Based on the red pine forest image data, each of the red pine canopies is extracted, and a red pine forest canopy image data creation unit for creating red pine forest canopy image data relating to each of the canopies;
Based on the image data of the red pine forest canopy and the spectral reflection characteristics of the damaged leaves collected from the red pine and the healthy leaves collected from the healthy red pine, Create spectral reflection characteristics of infected, dead, and healthy trees, identify damage classifications by tree crown based on the spectral reflection characteristics of the infected, dead, and healthy trees, and generate damage classification image data by crown. The damage classification image data creation part by tree to create,
Based on the damage classification image data by tree crown and the vertex image data of the red pine forest tree, the damage classification image data generation unit by tree apex that identifies the damage classification by tree vertex and creates damage classification image data by tree vertex When,
Based on the damage classification image data and geographical information according to the tree vertices, a damage classification number calculating unit that calculates the number of red pine contained in the whole area of the red pine forest or an arbitrary area for each damage classification,
Based on the result of calculating the number of damage categories, the total number of damaged trees and infected trees in all areas or arbitrary areas of the red pine forest is included in all areas or arbitrary areas of the red pine forest. A damage rate classification diagram creation unit that calculates the damage rate obtained by dividing by the total number of red pine and creates a damage rate classification diagram;
A device for calculating the damage classification of pine worms, characterized by comprising: In the pine weevil damage classification calculating apparatus according to claim 10,
It further comprises a red pine forest image data creation unit that creates red pine forest image data by extracting red pine forest from the photographed image data based on the spectral reflection characteristics of the tree species, and the red pine forest image data creation unit further comprises a red pine forest image data creation unit. After creating the data,
For each of the red pine forests, the red pine forest tree vertex image data creation unit by the red pine forest tree vertex image data creation unit, the red pine forest crown image data creation unit by the red pine forest crown image data creation unit, and the damage by the crown Creation of the damage classification image data for each crown by the classification image data creation section, creation of the damage classification image data for each tree vertex by the damage classification image data creation section by the tree apex, and the number of damage classification by the damage classification number calculation section An apparatus for calculating the damage category of pine worms, characterized in that the calculation and the creation of the damage rate category map by the damage rate category diagram creation unit are performed. In the pine weevil damage classification calculation apparatus according to claim 10 or 11,
An apparatus for calculating the damage category of a pine weevil, wherein the photographed image data uses image data photographed from autumn to winter. In the pine beetle damage classification calculation apparatus according to any one of claims 10 to 12,
Among the infected tree, the dead tree and the healthy tree, the infected tree and the dead tree are classified into a damaged leaf collected from the red pine damaged by the pine weevil and a healthy tree collected from the healthy red pine. Based on the spectral reflection characteristics of the leaves,
The infected tree has a higher reflectance than the dead tree and the healthy tree in the visible red region, and is classified as the dead tree and the healthy tree,
The dead tree has a lower reflectance than the infected tree and the healthy tree in the near-infrared region, and is classified from the infected tree and the healthy tree.
An apparatus for calculating the damage classification of pine weevil, which is characterized by that. In the pine beetle damage classification calculation apparatus according to any one of claims 10 to 13,
The Pinus densiflora damage category calculation apparatus, wherein the Pinus densiflora vertices image data creation unit extracts tree vertices using a local maximum value filter that automatically extracts a point having a maximum luminance value. In the pine weevil damage classification calculation apparatus according to any one of claims 10 to 14,
The pine pine beetle damage classification calculating apparatus, wherein the red pine forest canopy image data creation unit extracts a tree crown using the fact that the edge of the tree crown is darker than the crown portion. In the pine beetle damage classification calculation apparatus according to any one of claims 10 to 15,
The damage classification image data creation step according to the tree crown includes the damage classification as an infected tree, a dead tree, and a healthy tree, and when the infected tree, the dead tree, and the healthy tree are extracted from each crown of the red pine, Calculate the occupation area of each infected tree, dead tree, and healthy tree extracted from the tree canopy, and select the damage category with the largest occupation area of the calculated infected tree, dead tree, and healthy tree. An apparatus for calculating the damage category of pine beetles characterized by being regarded as a damage category. In the pine weevil damage classification calculation apparatus according to any one of claims 10 to 16,
The damage classification image data creation step for each tree vertex identifies the damage classification for each tree vertex based on the geographical information by superimposing the vertex image data for the red pine forest tree on the damage classification image data for each tree crown A device for calculating damage classification of pine weevil characterized by In the pine beetle damage classification calculation apparatus according to any one of claims 10 to 17,
The damage classification number calculation unit is configured to calculate subdivisions from the boundary data of forest owners defined as subdivisions by the geographical information in the whole area of the pine forest or in any area based on the damage classification image data classified by tree vertices. And calculating the number of trees for each damage category in the extracted sub-compartment.