Green Coverage Assessment System

Abstract

The object of the present invention is to solve the conventional problems, that is, to provide a vegetation coverage condition determination system that compares images from two periods after adjusting for seasons, and determines areas where vegetation coverage has changed using NDVI and GSI. [Solution] The green cover condition determination system of the present invention is a system that determines the green cover condition of the ground surface based on a first image and a second image, and is equipped with a vegetation index calculation means, a granularity index calculation means, an index difference calculation means, a green cover area selection means, and a changed area extraction means. The index difference calculation means is a means for calculating a first index difference and a second index difference by subtracting the granularity index from the vegetation index. The changed area extraction means is a means for extracting an area of ​​the first image that satisfies a change condition as a changed area from an area that corresponds to the green cover area of ​​the second image.

Description

[Technical Field]

[0001] The present invention relates to a technology for determining the green coverage state of the ground surface, and more specifically, to a green coverage state determination system that can determine areas where the green coverage state has changed using images acquired at two different times in the same season. [Background technology]

[0002] Embankments are sometimes built to construct roads and houses. They are also sometimes built to temporarily store surplus soil generated during tunnel excavation and cutting work at disposal sites. Normally, when building embankments, sufficient compaction is carried out under appropriate construction management. This creates a solid soil structure that does not easily collapse. On the other hand, if proper embankment construction is not carried out, such as simply piling up soil and sand for disposal, the soil and sand may easily collapse due to heavy rain, etc. In fact, disasters caused by the collapse of discarded soil and sand have occurred in various places in the past, and inappropriate embankments have been seen as a problem.

[0003] In July 2021, an embankment in Izuyama, Atami City, Shizuoka Prefecture, collapsed due to prolonged rainfall, causing a mudslide and resulting in the loss of many lives and serious damage. The embankment was built to fill in a former valley, and was located in an area where groundwater flowed in from the Aihaji River and Narusawa River. Furthermore, it was constructed improperly. Therefore, heavy rain over a long period of time caused a large amount of soil and sand to flow out all at once, resulting in a major disaster.

[0004] This disaster prompted the enactment of the "Land Development and Specific Embankment Regulation Act (Embankment Regulation Act)" in May 2023. This law allows local governments to designate areas where dangerous embankments are restricted (areas where land development, etc. is restricted and areas where specific embankments, etc. are restricted), and requires permission from the local government to carry out embankments in designated areas.

[0005] Even if a restricted area for embankments is designated under the Embankment Control Act, it is entirely possible that embankments have already been constructed within that restricted area. Furthermore, it is possible that those unaware of the law or malicious actors may build embankments without permission from the local government. While it is desirable for managers (such as local governments) to be aware of embankments constructed within such restricted areas, the area typically spans a large area, making mobile monitoring impractical. To detect target features over a wide area, aerial images are typically used. In particular, when extracting green areas covered with vegetation, such as wooded areas, grasslands, and farmlands, the Normalized Difference Vegetation Index (NDVI) obtained by analyzing satellite images is often used. For example, Patent Document 1 proposes a technology that calculates a vegetation index for each pixel in an image and detects dead and damaged tree areas based on the variability of the vegetation index for each pixel in a segmented area and the average vegetation index for the segmented area. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2023-151082 Summary of the Invention [Problem to be solved by the invention]

[0007] The technology disclosed in Patent Document 1 can detect dead and damaged tree areas as they currently exist, but because it is based on a single image, it is not possible to grasp changes in the dead and damaged tree areas. In other words, it is not possible to determine whether the dead and damaged tree areas have continued to exist since last year, or whether there was no dead and damaged tree area last year but the situation has changed this year. Therefore, even if there is an area where earthworks have been built without permission from the local government in a restricted area for earthworks, it is not easy to find it.

[0008] To check the change in green cover over time, it is possible to use images from two different periods. For example, by comparing the NDVI based on last year's satellite imagery with the NDVI based on this year's satellite imagery, areas that have lost their green cover this year (such as embankments) can be extracted.

[0009] However, because conventional satellites rarely photograph the same area, there was no practice of matching the seasons when comparing satellite images from two different periods. As a result, it was necessary to manually remove seasonal vegetation changes and then compare satellite images from two different periods. Even if images from two different periods were compared after matching the seasons, there are limits to extracting subtle seasonal changes in vegetation and crops using NDVI alone, so manual work by humans, based on visual interpretation and other information, is still necessary.

[0010] The objective of the present invention is to solve the conventional problems, that is, to provide a vegetation coverage condition determination system that compares images from two periods after adjusting for seasons, and determines areas of change in vegetation coverage condition using NDVI and Grain Size Index (GSI). [Means for solving the problem]

[0011] The present invention focuses on the use of images taken at two different times during the same season, and extracts changes in vegetation coverage by comparing the differences in NDVI and GSI between the two times, and is an invention based on an idea that has not been seen before.

[0012] The vegetation coverage condition determination system of the present invention is a system for determining the vegetation coverage condition of the ground surface based on a first image and a second image, and includes a vegetation index calculation means, a granularity index calculation means, an index difference calculation means, a vegetation coverage area selection means, and a changed area extraction means. The first image is taken at a first time period, and the second image is taken at a second time period that is older (past) than the first time period, and both images are taken during the same determination period (a period into which a year is divided, such as January to December). Among these, the vegetation index calculation means is means for calculating a "first vegetation index (NDVI at the first time period)" based on the first image and a "second vegetation index (NDVI at the second time period)" based on the second image, and the granularity index calculation means is means for calculating a "first granularity index (GSI at the first time period)" based on the first image and a "second granularity index (GSI at the second time period)" based on the second image. The index difference calculation means calculates a "first index difference" by subtracting the first vegetation index from the first vegetation index, and calculates a "second index difference" by subtracting the second vegetation index from the second vegetation index. The green cover area selection means selects an area in the second image where the second index difference exceeds a predetermined "green cover threshold" as a "green cover area." The changed area extraction means extracts an area in the first image corresponding to the green cover area that satisfies a "change condition" as a "change area" where the green cover state has changed between the first time period and the second time period. The index difference calculation means calculates a first index difference based on the first vegetation index and the first vegetation index for the same position, and calculates a second index difference based on the second vegetation index and the second vegetation index for the same position. The change condition is a condition in which the value obtained by subtracting the first index difference from the second index difference exceeds a predetermined "difference threshold."

[0013] The green coverage state determination system of the present invention may also be configured such that the change condition is that the first index difference is equal to or less than a green coverage threshold value.

[0014] The vegetation coverage state determination system of the present invention can also calculate the first index difference and the second index difference for each pixel. In this case, the vegetation index calculation means calculates the first vegetation index and the second vegetation index for each pixel constituting the image, and the granularity index calculation means calculates the first granularity index and the second granularity index for each pixel. The changed area extraction means sets pixels that satisfy the change condition as "changed pixels" and extracts areas where a number of changed pixels exceeding a predetermined "pixel number threshold" is gathered as changed areas.

[0015] The vegetation coverage state determination system of the present invention can also be configured to calculate the first index difference and the second index difference for each "divided area (area obtained by dividing the first image or the second image into multiple areas)." In this case, the vegetation index calculation means calculates the first vegetation index and the second vegetation index by statistically processing the vegetation index for multiple pixels that make up the divided area, and the granularity index calculation means calculates the first granularity index and the second granularity index by statistically processing the granularity index for multiple pixels that make up the divided area. In addition, the changed area extraction means extracts divided areas that satisfy a change condition as changed areas.

[0016] The green coverage state determination system of the present invention may further include a definite change area selection means for selecting, as a definite change area, an area that has been determined as a change area in a predetermined number of consecutive determination periods (for example, the past two months and this month, a total of three months).

[0017] The green coverage condition determination system of the present invention may also be configured such that the green coverage threshold value is set to a different value for each determination period depending on the characteristics of the season.

[0018] The green coverage state determination system of the present invention may be configured such that the difference threshold is set to a different value for each determination period depending on the characteristics of the season. [Effects of the Invention]

[0019] The green coverage state determination system of the present invention has the following effects. (1) By comparing images taken at two different times in the same season, the manual work of removing seasonal changes in vegetation can be eliminated. (2) Since the judgment is based on reference to GSI in addition to NDVI, seasonal changes in crops, etc. can be extracted more accurately than with conventional technology. (3) It eliminates the need for manual judgment based on visual interpretation and other information, improving efficiency and objectivity. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a block diagram showing the main configuration of a green coverage state determination system according to the present invention; [Figure 2] FIG. 1 is a model diagram schematically showing an image in which a plurality of divided regions are set. [Figure 3] A model diagram that schematically shows a "first index difference image" generated based on a first index difference, a "second index difference image" generated based on a second index difference, and an image of a change area extracted based on the first index difference image and the second index difference image. [Figure 4] FIG. 10 is a model diagram that schematically shows a changed region in which changed pixels exceeding a pixel number threshold are gathered. [Figure 5] A model diagram illustrating an example in which 12 types of change regions are extracted. [Figure 6] A model diagram illustrating an example in which an area that was determined to be a change area three times in a row in September, October, and November was selected as a confirmed change area. [Figure 7] 1 is a flow chart showing the main processing flow of the green coverage determination system of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0021] An example of an embodiment of the green coverage state determination system of the present invention will be described with reference to the drawings.

[0022] 1.Definition Before describing the embodiments of the present invention, definitions of terms used herein will be provided.

[0023] (1st and 2nd images) The green cover condition determination system of the present invention extracts areas where vegetated areas, such as wooded areas, grasslands, and farmlands, have changed (no longer green cover areas) due to embankments or other factors (hereinafter referred to as "changed areas"). One of its technical features is the use of images taken at two different times. However, both images taken at two different times are images taken during the same "determination period." Here, the determination period refers to a period set by dividing a year into periods. For example, the determination period can be set to January to December, or to spring (March to May), summer (June to August), autumn (September to November), and winter (December to February). For convenience, an image taken at the first of the two periods will be referred to as the "first image," and an image taken at a second period older than the first period will be referred to as the "second image." For example, an image taken in July of this year (the determination period) will be referred to as the first image, and an image taken in July of last year will be referred to as the second image. The images (first and second images) used in the present invention are images of the target area taken from the sky, such as satellite photographs.

[0024] (Vegetation index and grain size index) One of the technical features of the vegetation coverage determination system of the present invention is that it extracts changed areas based on NDVI (Normalized Difference Vegetation Index) and GSI (Granularity Index) obtained from images. For convenience, the NDVI obtained from the first image will be referred to as the "first vegetation index," and the NDVI obtained from the second image will be referred to as the "second vegetation index." Similarly, the GSI obtained from the first image will be referred to as the "first granularity index," and the GSI obtained from the second image will be referred to as the "second granularity index."

[0025] (index difference) Both NDVI and GSI are indices that can estimate the state of the earth's surface. It is known that a higher NDVI value indicates a higher probability of vegetation coverage, while a higher GSI value indicates a higher probability of bare ground. Therefore, the larger the value obtained by subtracting GSI from NDVI (hereinafter referred to as the "index difference"), the higher the likelihood of vegetation coverage, and the smaller the index difference, the higher the likelihood of non-vegetation coverage (a state without vegetation coverage). Therefore, the present invention utilizes this index difference (NDVI-GSI). For convenience, the index difference (first vegetation index - first granularity index) calculated based on the first image will be referred to as the "first index difference," and the index difference (second vegetation index - second granularity index) calculated based on the second image will be referred to as the "second index difference." Furthermore, the value obtained by subtracting the first index difference from the second index difference (second index difference - first index difference) will be referred to as the "two-time difference."

[0026] 2. Green Coverage Assessment System Next, the vegetation coverage condition determination system 100 of the present invention will be described in detail. Fig. 1 is a block diagram showing the main components of the vegetation coverage condition determination system 100 of the present invention. As shown in this figure, the vegetation coverage condition determination system 100 of the present invention is configured to include vegetation index calculation means 101, granularity index calculation means 102, index difference calculation means 103, vegetation coverage area selection means 104, and changed area extraction means 105, and can also be configured to include confirmed changed area selection means 106, output means 107, image storage means 108, etc.

[0027] Of the components of the vegetation coverage state determination system 100, the vegetation index calculation means 101, the granularity index calculation means 102, the index difference calculation means 103, the vegetation coverage area selection means 104, the changed area extraction means 105, and the confirmed changed area selection means 106 can be manufactured as dedicated components or can be implemented using general-purpose computers. That is, the specific processing of each component is performed by causing the computer to execute calculations using a predetermined program. The computer can be implemented as a personal computer (PC), a tablet PC such as an iPad (registered trademark), a mobile device such as a smartphone, or a personal data assistant (PDA). The computer includes a processor such as a CPU, memory such as ROM and RAM, and some computers also include input means such as a mouse and keyboard, and a display. If the computer includes a display, the display can also be used as the output means 107.

[0028] The image storage means 108 can be a storage device of a general-purpose computer (for example, a personal computer), or can be built in a database server. When built in a database server, it can be placed on a local network (LAN: Local Area Network), or can be a cloud server that stores data via the Internet.

[0029] Below, each of the main elements constituting the green coverage state determination system 100 of the present invention will be described in detail.

[0030] (Image storage means) The image storage means 108 constituting the green coverage state determination system 100 stores images taken at multiple periods, and stores images taken for each determination period in each period. For example, if the determination period is set to January to December and five years' worth of images are stored, the image storage means 108 will store 60 (5 x 12) types of images.

[0031] (Vegetation index calculation method) The vegetation index calculation means 101 constituting the green coverage state determination system 100 is means for calculating a first vegetation index based on a first image and for calculating a second vegetation index based on a second image. This vegetation index calculation means 101 can be configured to calculate the first vegetation index and the second vegetation index for each pixel constituting the images (first image and second image), or can be configured to calculate the first vegetation index and the second vegetation index for each region obtained by dividing the image into a plurality of regions (hereinafter simply referred to as "divided regions").

[0032] Each of the multiple divided regions set in an image is made up of two or more pixels. These divided regions can be set with the same shape and size, or they can be set with different shapes and sizes, as shown in Figure 2. For example, divided regions can be set by grouping pixels that have similar spectral information (pixel values) and are located nearby.

[0033] When the vegetation index calculation means 101 calculates the first vegetation index and the second vegetation index for each divided area, it is preferable to calculate the first vegetation index and the second vegetation index that represent the divided area by statistically processing the NDVI (vegetation index) of the multiple pixels that make up each divided area. For example, if a divided area is made up of 100 pixels, the statistical value of the NDVI of 100 pixels is used as the first vegetation index and the second vegetation index. The statistical value can be the average, median, mode, maximum, minimum, etc.

[0034] (Particle size index calculation means) The granularity index calculation means 102 constituting the green coverage condition determination system 100 is means for calculating a first granularity index based on a first image and for calculating a second granularity index based on a second image. Like the vegetation index calculation means 101, this granularity index calculation means 102 can be configured to calculate the first granularity index and the second granularity index for each pixel constituting an image, or can be configured to calculate the first granularity index and the second granularity index for each divided area. However, when the vegetation index calculation means 101 calculates the first vegetation index and the second vegetation index for each pixel, the granularity index calculation means 102 also calculates the first granularity index and the second granularity index for each pixel, and when the vegetation index calculation means 101 calculates the first vegetation index and the second vegetation index for each divided area, the granularity index calculation means 102 also calculates the first granularity index and the second granularity index for each divided area.

[0035] When the granularity index calculation means 102 calculates the first granularity index and the second granularity index for each divided area, it is preferable to calculate the first granularity index and the second granularity index by statistically processing the GSI (granularity index) of the multiple pixels that make up each divided area, as in the case of the vegetation index calculation means 101. For example, if a divided area is made up of 100 pixels, the statistical value of the 100 GSIs is used as the first granularity index and the second granularity index. The statistical value can be the mean value, median value, mode value, maximum value, minimum value, or the like.

[0036] (Indicator difference calculation means) The index difference calculation means 103 constituting the green coverage condition determination system 100 is a means for calculating a first index difference by subtracting a first vegetation index from a first vegetation index, and for calculating a second index difference by subtracting a second vegetation index from a second vegetation index. However, the index difference calculation means 103 calculates the first index difference based on the first vegetation index and the first granularity index relating to the same planar position, and similarly calculates the second index difference based on the second vegetation index and the second granularity index relating to the same planar position. Furthermore, when the vegetation index calculation means 101 and the granularity index calculation means 102 calculate their respective values ​​for each pixel, the index difference calculation means 103 also calculates the first index difference and the second index difference for each pixel, and when the vegetation index calculation means 101 and the granularity index calculation means 102 calculate their respective values ​​for each divided area, the index difference calculation means 103 also calculates the first index difference and the second index difference for each divided area.

[0037] (Green area selection means) The green coverage area selection means 104 that constitutes the green coverage state determination system 100 is a means for selecting an area in the second image where the second index difference exceeds a predetermined "green coverage threshold" (hereinafter referred to as a "green coverage area"). Of course, when the index difference calculation means 103 calculates the second index difference for each pixel, the green coverage area selection means 104 also selects a green coverage area for each pixel, and when the index difference calculation means 103 calculates the second index difference for each divided area, the green coverage area selection means 104 also selects a green coverage area for each divided area.

[0038] (Varied area extraction means) The changed area extraction means 105 constituting the green coverage state determination system 100 is a means for extracting areas that became non-green coverage areas (e.g., bare areas) in the first period from areas that were estimated to be green coverage areas in the second period. That is, the changed area extraction means 105 selects areas in the first image that correspond to green coverage areas (hereinafter referred to as "candidate areas"), and then extracts areas (hereinafter referred to as "change areas") that satisfy predetermined conditions (hereinafter referred to as "change conditions") from the candidate areas. Figure 3 shows a "first index difference image" generated based on the first index difference and a "second index difference image" generated based on the second index difference, and shows images in which changed areas are extracted based on these first index difference image and second index difference image.

[0039] The change condition for extracting a changed area can be a condition set based on the degree of change between the first index difference and the second index difference. The second index difference corresponding to a candidate area indicates that the area is a green-covered area, and if the first index difference is significantly reduced from that second index difference, the area is considered to have become a non-green-covered area. Therefore, the change condition can be a condition that the difference between the two periods (second index difference - first index difference) exceeds a predetermined threshold (hereinafter referred to as the "difference threshold"). In other words, the change condition in this case is a condition that the first index difference is smaller than the second index difference, and the absolute value of the difference between the first index difference and the second index difference is greater than the difference threshold.

[0040] Alternatively, the change condition may be such that non-green areas are extracted directly based on the first index difference. In other words, the change condition in this case is a condition that the first index difference is equal to or less than the green coverage threshold value.

[0041] When the green covered area selection means 104 selects a green covered area for each divided area, the changed area extraction means 105 also extracts a changed area for each divided area. On the other hand, when the green covered area selection means 104 selects a green covered area for each pixel, the changed area extraction means 105 may also be designed to extract a changed area for each pixel, or to extract only areas having a certain area (hereinafter referred to as "aggregate areas") as changed areas.

[0042] When extracting a cluster region as a changed region, the changed region extraction means 105 sets pixels in the candidate region that satisfy the change condition as "changed pixels," and can extract a region where the number of changed pixels gathered exceeds a predetermined threshold (hereinafter referred to as the "pixel count threshold") as a changed region. For example, in the case of Figure 4, changed pixels (pixels shown in gray in the figure) are set in approximately three locations, but because the pixel count threshold is set to 10, only the cluster of changed pixels in the upper left of the first image (13 changed pixels) is extracted as a changed region.

[0043] The changed areas extracted by the changed area extraction means 105 can be stored in association (linked) with divided areas or pixels converted into raster data. Alternatively, a point can be set at the center of a pixel, and the changed areas 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 is easier to indicate, and attribute data is easier to manage.

[0044] (Definite change area selection means) If the evaluation period is set to January to December, it is possible to extract 12 changed areas in the first period (for example, this year). In other words, as shown in Figure 5, 12 types of green areas and candidate areas (the images in the middle row in the figure) can be selected based on 12 pairs of first and second images (the images in the top row in the figure), and 12 types of changed areas (the images in the bottom row in the figure) can be extracted from those candidate areas based on the change conditions.

[0045] However, it is conceivable that an area that is determined to be a changed area in November, for example, may not be determined to be a changed area in October or December. In this case, it is not appropriate to immediately determine the area as a changed area based on the November determination. Therefore, it is advisable to position the extraction of changed areas by the changed area extraction means 105 as a "first screening," and then perform a "second screening" to confirm the changed area. In other words, candidate changed areas are extracted through the first screening, and a changed area with a higher probability is confirmed from among the candidate changed areas through the second screening. Of course, the green coverage condition determination system 100 of the present invention can also be configured to confirm the changed area as a changed area using only the first screening, i.e., the changed area extracted by the changed area extraction means 105, as is.

[0046] The confirmed change area selection means 106 included in the green coverage state determination system 100 is a means for finally determining a change area from the candidate change areas extracted by the change area extraction means 105, and is, in a sense, a means for performing secondary screening. The change area extraction means 105 selects an area (hereinafter referred to as a "confirmed change area") that has been determined as a change area for a predetermined number of consecutive determination periods (hereinafter referred to as the "determined number of determinations"). For example, in the case shown in FIG. 6, the number of determinations is set to three, and an area that has been determined as a change area three consecutive times in September, October, and November is selected as a confirmed change area. Note that when extracting a cluster area (an area where changed pixels are clustered by a pixel count threshold) as a change area, it is possible that the consecutive change areas for the number of determinations do not perfectly match. In this case, it is possible to select the part of the change area that matches the number of determinations as a confirmed change area, or to assign a degree of certainty to each pixel depending on the number of times it has been determined as a change area.

[0047] (Green cover threshold and difference threshold) The green coverage threshold and the difference threshold can be set to the same values ​​regardless of the judgment period, or can be set to different values ​​for each judgment period depending on the characteristics of the season. For example, the green coverage threshold and the difference threshold can be set to relatively large values ​​for the judgment period corresponding to summer, and relatively small values ​​for the judgment period corresponding to winter.

[0048] (Processing flow) The main processing of the green coverage condition determination system 100 of the present invention will be described in detail below with reference to Fig. 7. Fig. 7 is a flow diagram showing an example of the flow of the main processing of the green coverage condition determination system 100 of the present invention, in which the central column shows the processing to be performed, the left column shows input information required for that processing, and the right column shows output information resulting from that processing.

[0049] When the vegetation coverage state determination system 100 determines the vegetation coverage state of the ground, first, the first image and the second image are read from the image storage means 108, and the vegetation index calculation means 101 is used to calculate the first vegetation index based on the first image and the second vegetation index based on the second image (Step 201 in FIG. 7). Furthermore, the granularity index calculation means 102 is used to calculate the first granularity index based on the first image and the second vegetation index based on the second image (Step 202 in FIG. 7). Next, the index difference calculation means 103 is used to calculate the first index difference from the first vegetation index and the first granularity index (Step 203 in FIG. 7), and to calculate the second index difference from the second vegetation index and the second granularity index (Step 204 in FIG. 7).

[0050] Once the second index difference is obtained, the green area selection means 104 is used to select a green area from the second image (Step 205 in FIG. 7), and an area in the first image corresponding to the green area is selected as a candidate area (Step 206 in FIG. 7). In addition, a two-time difference (second index difference - first index difference) is calculated based on the first index difference and the second index difference for the same evaluation period (Step 207 in FIG. 7). At this time, the two-time difference can be calculated over the entire range of the images (first image and second image), or the two-time difference can be calculated limited to the candidate area.

[0051] Once the candidate region and the two-time difference are obtained, the changed region is extracted using the changed region extraction means 105 (Step 208 in FIG. 7). Also, a definite changed region can be selected from the changed region using the definite changed region selection means 106 (Step 209 in FIG. 7). Then, once the definite changed region is obtained, the definite changed region is output to the output means 107 such as a display or printer (Step 210 in FIG. 7). [Industrial Applicability]

[0052] The green cover condition determination system of the present invention can be used in various cases, particularly to detect areas where green cover has changed to bare ground, such as detecting improper embankments, improper tree cutting, etc. According to the present invention, it is possible to quickly detect unexpected embankments in restricted areas, and by taking prompt measures, disasters caused by embankment collapse can be avoided. Considering this, it can be said that this invention is not only applicable to industry, but is also expected to make a great contribution to society. [Explanation of symbols]

[0053] 100 Green Coverage Determination System of the Present Invention 101 Vegetation index calculation method (for vegetation coverage assessment system) 102 (Green Coverage Assessment System) Grain Size Index Calculation Method 103 (Green Coverage Assessment System) Index Difference Calculation Means 104 (of the green coverage state determination system) green coverage area selection means 105 (Green Coverage Assessment System) Change Area Extraction Means 106 (Green Coverage Determination System) Definite Change Area Selection Means 107 (Output means of green coverage determination system) 108 Image storage means (of green coverage determination system)

Claims

1. A system for determining the state of green coverage of the ground surface based on a first image of a target area taken during a first period and a second image of the target area taken during the same determination period as the first image, the second image being older than the first period and being older than the first period, the system comprising: a vegetation index calculation means for calculating a first vegetation index based on the first image and a second vegetation index based on the second image; a particle size index calculation means for calculating a first particle size index based on the first image and a second particle size index based on the second image; an index difference calculation means for calculating a first index difference by subtracting the first vegetation index from the first particle size index, and for calculating a second index difference by subtracting the second particle size index from the second vegetation index; a green coverage area selection means for selecting an area in the second image where the second index difference exceeds a predetermined green coverage threshold as a green coverage area; a change area extraction means for extracting an area in the first image that corresponds to the green coverage area and satisfies a change condition as a change area in which the green coverage state has changed between the first time period and the second time period; the index difference calculation means calculates the first index difference based on the first vegetation index and the first granularity index relating to the same position, and calculates the second index difference based on the second vegetation index and the second granularity index relating to the same position, the change condition is a condition that a value obtained by subtracting the first index difference from the second index difference exceeds a predetermined difference threshold value. A green coverage determination system characterized by:

2. A system for determining the state of green coverage of the ground surface based on a first image of a target area taken during a first period and a second image of the target area taken during the same determination period as the first image, the second image being older than the first period and being older than the first period, the system comprising: a vegetation index calculation means for calculating a first vegetation index based on the first image and a second vegetation index based on the second image; a particle size index calculation means for calculating a first particle size index based on the first image and a second particle size index based on the second image; an index difference calculation means for calculating a first index difference by subtracting the first vegetation index from the first particle size index, and for calculating a second index difference by subtracting the second particle size index from the second vegetation index; a green coverage area selection means for selecting an area in the second image where the second index difference exceeds a predetermined green coverage threshold as a green coverage area; a change area extraction means for extracting an area in the first image that corresponds to the green coverage area and satisfies a change condition as a change area in which the green coverage state has changed between the first time period and the second time period; the index difference calculation means calculates the first index difference based on the first vegetation index and the first granularity index relating to the same position, and calculates the second index difference based on the second vegetation index and the second granularity index relating to the same position, The change condition is a condition that the first index difference is equal to or less than the green coverage threshold. A green coverage determination system characterized by:

3. the vegetation index calculation means calculates the first vegetation index for each pixel constituting the first image, and calculates the second vegetation index for each pixel constituting the second image; the particle size index calculation means calculates the first particle size index for each pixel constituting the first image, and calculates the second particle size index for each pixel constituting the second image, the changed region extraction means sets pixels that satisfy the change condition as changed pixels; The changed region extraction means extracts, as the changed region, a region in which the changed pixels are gathered in a number exceeding a preset pixel number threshold.

3. The green coverage determination system according to claim 1 or 2.

4. A plurality of divided regions are set in the first image and the second image, the vegetation index calculation means calculates the first vegetation index by statistically processing vegetation indices related to a plurality of pixels that constitute the divided region of the first image, and calculates the second vegetation index by statistically processing vegetation indices related to a plurality of pixels that constitute the divided region of the second image; the granularity index calculation means calculates the first granularity index by statistically processing granularity indices related to a plurality of pixels that constitute the divided region of the first image, and calculates the second granularity index by statistically processing granularity indices related to a plurality of pixels that constitute the divided region of the second image, the changed region extraction means extracts the divided region that satisfies the change condition as the changed region; 3. The green coverage determination system according to claim 1 or 2.

5. The apparatus further includes a definite change area selection means for selecting, as a definite change area, an area that has been determined as the change area in each of the determination periods that lasts a predetermined number of times.

3. The green coverage determination system according to claim 1 or 2.

6. The green coverage threshold is set to a different value for each determination period depending on the characteristics of the season.

3. The green coverage determination system according to claim 1 or 2.

7. The difference threshold is set to a different value for each determination period depending on the characteristics of the period.

2. The green coverage determination system according to claim 1.

Images