Site preparation area determination system

Abstract

The object of the present invention is to solve the problems of the conventional method, namely, to provide a site preparation area determination system that evaluates whether site preparation has been carried out using images that can be obtained at low cost. [Solution] The site preparation area determination system of the present invention is a system that determines an area where "site preparation" has been performed based on a first image and a second image, and comprises a green cover index calculation means, a target area selection means, and a site preparation area extraction means. Of these, the site preparation area extraction means is a means for extracting the target area in the first period as a site preparation area when the target area satisfies a change condition. The change condition is that the value obtained by subtracting the first green cover index from the second green cover index exceeds a predetermined difference threshold.

Description

Technical Field

[0001] The present invention relates to a technique for determining the greening state of the ground surface. More specifically, it relates to a system for determining an earthwork area that can detect an area where "earthwork" has been performed based on changes in indices such as the vegetation index (NDVI: Normalized Difference Vegetation Index) and the grain size index (GSI: Grain Size Index).

Background Art

[0002] Timber such as columns and boards is an essential material for wooden buildings including detached houses. In the past, a large amount of timber could be imported from overseas, but since the 1980s, movements to stop deforestation have become active in various countries, and domestic production has been attracting attention recently. Currently, the self-sufficiency rate of timber is said to be 35% (27 million m 3 ), but it is expected that the maturation of domestic timber will further progress in the future, and forestry is expected to be a growing industry in our country.

[0003] Naturally, logging is carried out to obtain timber, and reforestation may be carried out at the logging site for the purpose of obtaining timber again. In this case, although contractors are supposed to report to forest managers (such as local governments) about reforestation, it is a reality that there are contractors who forget to report or are unaware of the system.

[0004] Assuming that the reporting of reforestation by contractors is uncertain, forest managers need to investigate the reforestation sites themselves. For this purpose, it is conceivable for forest managers to visit the site for visual confirmation, but usually, it is not realistic to conduct on-site confirmation by people because their management area is extremely large. When detecting target ground objects over a wide area, it is also conceivable to use images taken from above. For example, Patent Document 1 proposes a technique for managing the logging status of a forest using a cloud-removed image obtained by removing clouds from a forest image. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2023-39648 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] Normally, when using satellite imagery for assessment, high-resolution images are required to ensure the accuracy of the assessment. However, reforestation is an irregular event, and detecting all areas where reforestation is carried out requires frequent observation, which incurs considerable costs. On the other hand, in recent years, there are organizations that provide satellite imagery at low cost (for example, free of charge), although the image quality is not as high.

[0007] The inventors of the present invention conducted extensive research on a technique for detecting reforestation sites using low-cost, but not high-resolution, satellite imagery, and discovered that "site preparation" serves as a clue. Site preparation involves removing tree trunks and branches left in the logging area, as well as shrubs and herbaceous plants that have grown after logging, to create an environment suitable for reforestation. Consequently, when site preparation is carried out, the green cover area decreases compared to the logging site. The inventors found that changes in images triggered by site preparation can be confirmed even with low-resolution images. In other words, it is possible to evaluate whether site preparation has been carried out using low-resolution images, and thereby estimate that reforestation will soon be carried out in that area.

[0008] The object of the present invention is to solve the problems of the conventional method, namely, to provide a site preparation area determination system that evaluates whether site preparation has been carried out using images that can be obtained at low cost. [Means for solving the problem]

[0009] The present invention focuses on the idea of ​​evaluating that land preparation has been carried out based on a change in an index (however, an index capable of determining vegetation conditions) based on images of an area designated as a logging site, and is an invention based on a novel concept.

[0010] The site preparation area determination system of the present invention is a system that determines an area where "site preparation" has been carried out based on a "first image" taken in a first period and a "second image" taken in a second period (an earlier period than the first period), and comprises a green cover index calculation means, a target area selection means, and a site preparation area extraction means. Of these, the green cover index calculation means is a means for obtaining a "first green cover index (an index that can determine the vegetation status)" from the first image and a "second green cover index (an index that can determine the vegetation status)" from the second image. The target area selection means is a means for selecting an area in the second image where the second green cover index is below a predetermined green cover threshold as the "target area," and the site preparation area extraction means is a means for extracting the target area in the first period as the "site preparation area" when the target area satisfies the change condition. The change condition is that "the value obtained by subtracting the first green cover index from the second green cover index exceeds a predetermined difference threshold." Then, based on the site preparation area extracted by the site preparation area extraction means, it can be evaluated that site preparation has been carried out.

[0011] The site preparation area determination system of the present invention may also be configured such that the change condition is "the first green cover index falls below a predetermined site preparation threshold."

[0012] The site preparation area determination system of the present invention can also be configured such that the first green cover index and the second green cover index are both green cover indexes.

[0013] In the site preparation area determination system of the present invention, the first green cover index and the second green cover index can also be defined as values ​​obtained by subtracting the particle size index from the green cover index.

[0014] The site preparation area determination system of the present invention may further include a site preparation area determination means for determining a site preparation area as a "confirmed site preparation area." In this case, the site preparation area extraction means extracts site preparation areas from each of a plurality of first images. The site preparation area determination means then determines a site preparation area as a confirmed site preparation area when it has been determined to be a site preparation area a predetermined number of times consecutively. [Effects of the Invention]

[0015] The site preparation area determination system of the present invention has the following effects: (1) Since there is no need to visually interpret the image and no other information is required, work efficiency and objectivity are improved. (2) Because images are used, changes in the work area can be grasped over a wide area. (3) Even if images are acquired at a high frequency, costs can be reduced compared to conventional methods. [Brief explanation of the drawing]

[0016] [Figure 1] A graph showing the change in the green cover index over time in areas A, B, and C. [Figure 2] A block diagram showing the main components of the invention domain determination system. [Figure 3] A schematic model diagram showing an image with multiple divided regions. [Figure 4] A schematic model diagram showing a target region where the number of pixels exceeds a pixel count threshold. [Figure 5] A flowchart showing the main processing flow of the site preparation area determination system of the present invention. [Modes for carrying out the invention]

[0017] An example of an embodiment of the site preparation area determination system of the present invention will be described with reference to the figures.

[0018] The invention of this application focuses on "site preparation" as a clue for estimating the reforestation work area, and extracts an area where site preparation can be evaluated (hereinafter referred to as "site preparation area"). As described above, site preparation is an operation that removes the trunks and branches of trees left on the logged area, shrubs and herbs that grew after logging, etc., and prepares an environment that is easy to plant during reforestation. Therefore, when site preparation is carried out, the green coverage area will decrease compared to after logging. Thus, the invention of this application first selects an area that can be evaluated as logged (hereinafter referred to as "target area"), and further extracts the site preparation area from the target area.

[0019] In addition, the invention of this application is information obtained from an image (such as a satellite photo, etc.) taken from above the investigation target range (for example, the management range of forest managers), and detects the target area and the site preparation area based on an index that can determine the vegetation status (hereinafter referred to as "green coverage index"). Note that this green coverage index indicates that the larger the value, the more the state is covered by vegetation (hereinafter referred to as "green coverage state"), and conversely, the smaller the value, the less the state is covered by vegetation (hereinafter referred to as "non-green coverage state"). For example, NDVI (vegetation index) can be used as the green coverage index, or the difference between NDVI and GSI (granularity index) (NDVI - GSI) can be used as the green coverage index.

[0020] It is known that the higher the value of NDVI, the higher the probability of being covered by vegetation. On the other hand, it is known that the higher the value of GSI, the higher the probability of being bare land. Therefore, it is considered that the larger the value obtained by subtracting GSI from NDVI (hereinafter referred to as "index difference"), the higher the possibility of being in the green coverage state, and the smaller the index difference, the higher the possibility of being in the non-green coverage state (not being covered by vegetation).

[0021] Figure 1 is a graph showing the changes in the green cover index in areas A, B, and C, with the horizontal axis representing time and the vertical axis representing the green cover index. Areas B and C, shown in this figure, continue to have high green cover index values, meaning that the green cover state is maintained and therefore they can be evaluated as areas where logging has not taken place. On the other hand, in area A, a significant change in the green cover index is observed at each time period. Specifically, the green cover index for "Image A2" has decreased significantly from the green cover index for "Image A1," and it can be evaluated that logging took place during this period, meaning that area A can be selected as a target area. Subsequently, for a while (during the period of "Image A3" and "Image A4"), the green cover index is at a similar level to that for Image A2, but the green cover index for Image A5 is considerably smaller than that for Image A2. Therefore, it can be evaluated that site preparation took place immediately before Image A5 was acquired, meaning that area A at this point can be selected as a site preparation area.

[0022] As shown in Figure 1, the present invention uses multiple images (such as satellite images) acquired at different times within the same area. For convenience, the image acquired when the target area was selected (corresponding to image A2 in Figure 1) will be specifically referred to as the "second image," the image acquired before the second image (corresponding to image A1 in Figure 1) will be specifically referred to as the "third image," and the images acquired after the second image (corresponding to images A3 to A5 in Figure 1) will be specifically referred to as the "first image." Furthermore, the time when the first image was acquired will be referred to as the "first period," the time when the second image was acquired as the "second period," and the time when the third image was acquired as the "third period." Note that logging may occur at multiple locations at different times within the survey area. Therefore, the second image may be set differently for each logging site (i.e., the target area).

[0023] Figure 2 is a block diagram showing the main components of the site preparation area determination system 100 of the present invention. As shown in this figure, the site preparation area determination system 100 of the present invention is configured to include a green cover index calculation means 101, a target area selection means 102, and a site preparation area extraction means 103, and may also be configured to include a site preparation area determination means 104, an output means 105, an image storage means 106, and the like.

[0024] Of the various means constituting the site preparation area determination system 100, the green cover index calculation means 101, the target area selection means 102, the site preparation area extraction means 103, and the site preparation area determination means 104 can be manufactured as dedicated components or a general-purpose computer device can be used. In other words, each means performs its own specific processing by having the computer device execute calculations using a predetermined program. This computer device can consist of a personal computer (PC), a tablet PC such as an iPad (registered trademark), a mobile terminal including a smartphone, or a PDA (Personal Data Assistance). The computer device is equipped with a processor such as a CPU, memory such as ROM or RAM, and may also include input means such as a mouse or keyboard and a display. If the computer device includes a display, this display can also be used as an output means 105.

[0025] Furthermore, the image storage means 106 can utilize the storage device of a general-purpose computer (for example, a personal computer), or it can be built on a database server. When built on a database server, it can be located on a local network (LAN: Local Area Network), or it can be a cloud server that stores data via the internet.

[0026] The following describes in detail each of the main elements constituting the site preparation area determination system 100 of the present invention.

[0027] (Image storage means) The image storage means 106, which constitutes the site preparation area determination system 100, stores images of the survey area taken at multiple times. As described above, logging may be carried out at multiple locations at different times within the survey area (for example, the area managed by a forest manager), and therefore a different second image may be selected for each logging site (i.e., the target area). Accordingly, the image storage means 106 stores each image without assigning attributes such as first image, second image, third image, etc., but stores each image while associating it with the shooting time (the actual time, not the first or second period, etc.).

[0028] (Method for calculating green coverage index) The green cover index calculation means 101, which constitutes the site preparation area determination system 100, is a means for determining the green cover index based on each image. As described above, it can be specified to determine the NDVI as the green cover index, or it can be specified to determine the index difference (NDVI-GSI) as the green cover index. When a target area is selected, a second image for that target area is set, as well as the first and third images. For convenience, the green cover index determined based on the first image will be specifically referred to as the "first green cover index marker," the green cover index determined based on the second image will be specifically referred to as the "second green cover index marker," and the green cover index determined based on the third image will be specifically referred to as the "third green cover index marker."

[0029] Furthermore, the green cover index calculation means 101 can be configured to calculate the green cover index for each pixel that makes up the image, or it can be configured to calculate the green cover index for each of the multiple regions into which the image is divided (hereinafter simply referred to as "divided regions"). Each of the multiple divided regions set in the image is composed of two or more pixels. When setting these divided regions, they can be set with the same shape and size, or they can be set with different shapes and sizes, as shown in Figure 3. For example, divided regions can be set by grouping together pixels that have similar spectral information (pixel values) and are located in close proximity to each other.

[0030] When the green cover index calculation means 101 calculates the green cover index for each divided region, it is preferable to obtain a green cover index that represents that divided region by statistically processing the green cover indices of multiple pixels that constitute each divided region. For example, if a divided region consists of 100 pixels, the statistical value obtained from the green cover index of 100 pixels is used as the green cover index for that divided region. This statistical value can be the mean, median, mode, maximum, minimum, etc.

[0031] (Means for selecting the target area) The target area selection means 102, which constitutes the site preparation area determination system 100, is a means for selecting a target area (an area evaluated as a logging site) based on an image. More specifically, it selects an area from the image in which the green cover index is below a predetermined threshold (hereinafter referred to as the "green cover threshold"). Alternatively, it is also possible to use images from two periods and select an area as a target area in which the difference between the green cover index for the older image (hereinafter referred to as the "old green cover index") and the green cover index for the newer image (hereinafter referred to as the "new green cover index") exceeds the green cover threshold. However, the difference here is the value obtained by subtracting the new green cover index from the old green cover index (old green cover index - new green cover index).

[0032] When a target region is selected, a second image is set for that region, and simultaneously, the first and third images are set. As described above, in the case where a target region is selected using one image, that image becomes the second image for the target region, the image newer than the second image becomes the first image, and the image older than the second image becomes the third image. On the other hand, in the case where a target region is selected using two images from different time periods, the newer image becomes the second image for the target region, the image newer than the second image becomes the first image, and the image older than the second image becomes the third image.

[0033] Incidentally, when the green cover index calculation means 101 calculates the green cover index for each divided region, the target region selection means 102 also selects a target region for each of those divided regions. On the other hand, when the green cover index calculation means 101 calculates the green cover index for each pixel, the target region selection means 102 may be configured to select a target region for each pixel, or it may be configured to select a region with a certain area (hereinafter referred to as a "collected region") as the target region. In other words, for a pixel designated as a target region by the target region selection means 102 (hereinafter referred to as a "target pixel"), multiple neighboring target pixels are collected to form a target region.

[0034] When selecting a set region as a target region, the target region selection means 102 can select a region as a target region in which the number of target pixels constituting the set region exceeds a predetermined threshold (hereinafter referred to as the "pixel count threshold"). For example, in the case of Figure 4, set regions are formed in approximately three places, but since the pixel count threshold is set to 10, only the set region in the upper left (13 pixels) will be selected as the target region.

[0035] (Ground preparation area extraction means) The site preparation area extraction means 103, which constitutes the site preparation area determination system 100, is a means for extracting a "site preparation area" from the target area selected by the target area selection means 102. More specifically, it is a means for extracting a site preparation area based on the first image acquired after the second time period in which the second image was acquired (i.e., the time when logging took place). The process by which the site preparation area extraction means 103 extracts a site preparation area will be described in detail below.

[0036] First, the site preparation area extraction means 103 focuses on a target area based on the second image and sets that target area in the first image. Of course, the target area is set in the same position as in the second image, and its size and shape are also set in the same way. Then, when the target area in the first image satisfies the "change condition", the site preparation area extraction means 103 extracts that target area as a site preparation area.

[0037] Here, the change conditions can be set as conditions based on the first and second green cover indices for the target area (hereinafter referred to as the "comparison condition"), or as conditions based only on the first green cover index for the target area (hereinafter referred to as the "single condition"). The comparison condition is the condition that the value obtained by subtracting the first green cover index from the second green cover index for the target area (second green cover index - first green cover index) exceeds a predetermined threshold (hereinafter referred to as the "difference threshold"). The single condition, on the other hand, is the condition that the first green cover index for the target area falls below a predetermined threshold (hereinafter referred to as the "ground preparation threshold").

[0038] The preparation regions extracted by the preparation region extraction means 103 can be stored in association with (linked to) the divided regions or pixels that are converted into raster data. Alternatively, a point can be set at the center of each pixel, and the preparation region can be stored in association with this point. In this case, since raster data is not used, the background image becomes easier to see, its location becomes easier to indicate, and attribute data becomes easier to manage.

[0039] (Method for determining area of ​​land preparation) Incidentally, when the site preparation area extraction means 103 performs site preparation area extraction processing on multiple first images, it is possible that even if a target area is initially extracted as a site preparation area, it may not be extracted as a site preparation area in subsequent processing. In this case, it is not appropriate to immediately designate that target area as a site preparation area with a single extraction process. Therefore, it is preferable to position the extraction by the site preparation area extraction means 103 as a "primary screening," and then perform a "secondary screening" to determine the site preparation area. In other words, the primary screening extracts candidate site preparation areas, and the secondary screening determines the site preparation area with higher accuracy from among the candidate site preparation areas. Of course, the site preparation area determination system 100 of the present invention can also be designed to determine the site preparation area solely by primary screening, that is, by determining the site preparation area extracted by the site preparation area extraction means 103 as a site preparation area.

[0040] The site preparation area determination means 104, which constitutes the site preparation area determination system 100, is a means for finally determining the site preparation area from the candidate site preparation areas extracted by the site preparation area extraction means 103, and is, so to speak, a means for performing secondary screening. The site preparation area extraction means 103 determines the target area that has been designated as a site preparation area for a predetermined number of times (hereinafter referred to as the "number of determinations") as a "determined site preparation area".

[0041] (Process flow) The main processes of the site preparation area determination system 100 of the present invention will be described in detail below with reference to Figure 5. Figure 5 is a flowchart showing an example of the main processing flow of the site preparation area determination system 100 of the present invention, with the processes to be performed shown in the center column, the input information required for those processes shown in the left column, and the output information generated from those processes shown in the right column.

[0042] When the site preparation area determination system 100 extracts a site preparation area, first an image is read from the image storage means 106, and the green cover index is calculated based on that image using the green cover index calculation means 101 (Step 201 in Figure 5). Once the green cover index is calculated, the target area selection means 102 is used to select a target area from the image (Step 202 in Figure 5). At this time, if a target area is selected, that image becomes the second image for that target area. Next, an image newer than the second image (i.e., the first image) is read from the image storage means 106, and the site preparation area extraction means 103 is used to determine whether the target area satisfies the change condition. If the target area satisfies the change condition, that target area is extracted as a site preparation area (Step 203 in Figure 5). Once a site preparation area is extracted, that site preparation area is output to an output means 105 such as a display or printer (Step 204 in Figure 5). [Industrial applicability]

[0043] The land preparation area determination system of the present invention can be effectively used in cases where the conditions of various forests, including natural forests and artificial forests, are managed. According to the present invention, appropriate forest management will be carried out, promoting reforestation, thereby reducing the risk of extinction of wild animals due to deforestation, and considering that it will have the effect of mitigating global warming, which is a pressing issue today, the present invention can be said to be an invention that can be expected to make a great contribution not only to industry but also to society. [Explanation of symbols]

[0044] 100 The present invention's site preparation area determination system 101 (Means for calculating the green cover index of the site preparation area determination system) 102 (Target area selection means of the site preparation area determination system) 103 (Site preparation area extraction means of the site preparation area determination system) 104 (Means for determining the site preparation area of ​​the site preparation area determination system) 105 Output means (of the site preparation area determination system) 106 Image storage means (of the site preparation area determination system)

Claims

1. A system for determining the area where "land preparation" has been carried out, based on a first image taken during the first period and a second image taken during the second period which is earlier than the first period. A green cover index calculation means that calculates a first green cover index, which is an index that can determine the vegetation status from the first image, and a second green cover index, which is an index that can determine the vegetation status from the second image, A target area selection means for selecting a region in the second image in which the second green coverage index is below a predetermined green coverage threshold, The system includes a ground preparation area extraction means that extracts the target area as a ground preparation area in the first period when the target area satisfies the change conditions, The aforementioned change condition is that the value obtained by subtracting the first green cover index from the second green cover index exceeds a predetermined difference threshold. Based on the site preparation area extracted by the site preparation area extraction means, it can be evaluated that site preparation has been carried out. A site preparation area determination system characterized by the following.

2. A system for determining the area where "land preparation" has been carried out, based on a first image taken during the first period and a second image taken during the second period which is earlier than the first period. A green cover index calculation means that calculates a first green cover index, which is an index that can determine the vegetation status from the first image, and a second green cover index, which is an index that can determine the vegetation status from the second image, A target area selection means for selecting a region in the second image in which the second green coverage index is below a predetermined green coverage threshold, The system includes a ground preparation area extraction means that extracts the target area as a ground preparation area in the first period when the target area satisfies the change conditions, The aforementioned change condition is that the first green cover index falls below a predetermined site preparation threshold. Based on the site preparation area extracted by the site preparation area extraction means, it can be evaluated that site preparation has been carried out. A site preparation area determination system characterized by the following.

3. The first green cover index and the second green cover index are vegetation indicators. A site preparation area determination system according to claim 1 or 2, characterized in that it is the same as the one described in claim 2.

4. The first green cover index and the second green cover index are values ​​obtained by subtracting the particle size index from the vegetation index. A site preparation area determination system according to claim 1 or 2, characterized in that it is the same as the one described in claim 2.

5. The system further includes means for determining the aforementioned site preparation area as a determined site preparation area, The aforementioned site preparation area extraction means extracts the site preparation area from each of the multiple first images, The means for determining the site preparation area determines the site preparation area as the determined site preparation area when it has been determined to be the site preparation area for a predetermined number of consecutive times. A site preparation area determination system according to claim 1 or 2, characterized in that it is the same as the one described in claim 2.

Images