JP4553826B2 - Feature environment status providing method and program thereof - Google Patents

Description

  The present invention grasps the situation of rivers, coasts, wetlands, etc., and the conditions of rivers, rivers, rivers, or the size of gravel existing in the state, and displays them appropriately in color. The present invention relates to a feature environment situation providing method that provides information visually.

  In 1990, the “National Census of River Waterside” (hereinafter referred to as the “Water Country Survey”) started in the first-class rivers of 109 water systems nationwide and some of the second-class rivers managed by each prefecture. This survey is to grasp the organisms that inhabit and grow in rivers, such as plants, fish, and birds, by field confirmation. The survey has been conducted for 5 years as one cycle, and the survey for the third round is now underway.

  At the beginning of the survey, river managers had no information about living organisms, and began with investigating what organisms were present against the backdrop of growing public opinion on environmental issues. After that, the River Law was revised in 1997, and in addition to the conventional “Flood Control” and “Water Utilization” as the purpose of the river business, “maintenance and conservation of the river environment” was positioned as one of the three pillars of the project. As basic data that bears the burden, surveys have been conducted on rivers (including lakes) nationwide with a large budget each year.

  On the other hand, while the cost reduction is screamed throughout the public works, the value of data accumulated over the past ten years is being questioned. In other words, although the biological information has been accumulated, the fact that it is hardly utilized for river management (management of flood control while maintaining the biological environment) has become apparent.

  Since it is essential for river managers to acquire environmental information, the manual is being revised for the fourth round of surveys starting in 2006. There is no solution to the problem of how to utilize it.

  Factors that make it difficult for current water country survey data to be used for river management include survey sites that are heavily focused on living organisms, such as river channel topography, riverbed materials, trees in Nakashu, Nakashu, and river flows (Hayase, Hirase, Kashiwa). It can be pointed out that surveys on the environment have become very short.

In other words, the objects that can be operated in river management are not individual organisms but physical conditions such as river channel forms and flow conditions.
JP 2000-89664 A Japanese Patent No. 3546349 JP 2003-156330 A

  One of the factors that make it difficult for current water country survey data to be used in river management is that the survey target is heavily focused on living organisms, and the survey of the environment of the field is short.

  In other words, the objects that can be manipulated in river management are not individual organisms, but physical conditions such as river channel forms and flow conditions, etc. The number of torrents, sand gravel deposits, river changes, etc.) has not been acquired, so how to maintain the biological environment while ensuring a safe river channel against flooding There was a problem that it was not managed.

  In addition, the conventional water country survey is conducted only at some survey points set in the target section even at a high cost, so only local information can be obtained. There was a problem that could not be grasped.

  Therefore, there is a water environment information provision method that can easily provide visual information by quantifying and grasping the situation of the entire river and the base data easily with aviation laser data without overwhelming it, and displaying it appropriately in color. It is desirable to obtain.

The present invention classifies the features of the target area based on a color orthophoto image obtained by photographing the target area with an aircraft, laser data from the target area, DEM data based on the laser data, and DSM data. A feature environment status providing method for displaying a land cover classification map as a result of classification on a screen,
A first threshold of color intensity for an arbitrary band of red (R), green (G), and blue (B) determined by the luminance distribution of the image data group of the color orthophoto image, and one image comparing the color intensity of the arbitrary band of the color orthophoto image data, the the sun when the first threshold value on more than, classifying the image data in the first time less than the threshold value shade And a process of
When the image data is classified as Hinata, the correlation between the color intensities of the three bands of red (R), green (G), and blue (B) determined by the color intensity distribution of all image data in the target area Is applied to the red (R), green (G), and blue (B) color intensities of the image data for the sun, and the image data for the sun is applied to water, bare land, and vegetation. If the image data is classified as a shadow, and the step of classifying the image data,
A second threshold value of the reflection intensity that becomes the boundary between the water area and the bare land, and a third threshold value of the reflection intensity that becomes the boundary between the bare area and the vegetation, determined by the reflection intensity distribution of the laser data of the target area; Comparing the surface layer reflection intensity included in the laser data corresponding to the image data classified as the shade, and when the surface layer reflection intensity is smaller than the second threshold value, the surface layer reflection is reflected in the water area. Classifying the bare ground when the intensity is equal to or higher than the second threshold and equal to or lower than the third threshold, and classifying as the vegetation when the surface reflection intensity is higher than the third threshold;
Reading the surface elevation value of the DSM data and the ground elevation value of the DEM data, and setting the difference as the vegetation height;
Comparing the vegetation height with a fourth threshold, classifying the tree when the vegetation height is greater than or equal to the fourth threshold, and classifying the grassland when the vegetation height is less than the fourth threshold;
The gist is to repeatedly perform all the above steps for all the image data of the color orthophoto image, and to perform land cover classification for the image data of the target area.

(The invention's effect)
As described above , according to the present invention, it is possible to grasp the river flow situation, the state of the Nakasu, the size of the gravel deposited in the Nakasu, etc. without taking a great expense, and appropriately. By displaying this information in color and in easy-to-understand manner, it can be used for river management. The same effect can be expected with respect to the seaside and wetlands.

  FIG. 1 is a hardware configuration diagram of a terrestrial environment status providing system using laser data and color images by an aircraft according to the present invention.

  As shown in FIG. 1, a terrestrial environment situation providing system VPS1 using laser data from an aircraft includes a central information processing device (CPU unit) 1 including an appropriate combination of a processor, a microcomputer, a logic, a register, and the like, and a CPU unit. 1. Information input unit 2 such as a keyboard, mouse, interactive soft switch, external communication channel, etc. for inputting necessary control information and operation information, and display and printer for displaying and transmitting information from the CPU unit 1 in a broad sense The information output unit 3 including an external communication channel, the first storage unit 4 such as a ROM (ROM) in which information such as an operating system and application program read by the CPU unit 1 is stored, and the CPU unit 1 as needed. Second memory such as a RAM (RAM) for storing information to be processed and information from the CPU unit 1 Equipped with a 5 and the like. The first and second storage units 4 and 5 may be appropriately integrated and subdivided.

  The first storage unit 4 includes a water / land environment providing unit 10 that operates on a predetermined application, laser data Ri when a predetermined range of rivers, seasides, lakes, and wetlands are acquired by an aircraft, and color digital data Orthophoto image Pi (hereinafter referred to as orthoimage Pi), analysis area information (river channel area, wetland area, seaside area, etc.) that defines the analysis range of this image, laser data Ri, and orthoimage Pi Land cover classification information is stored. The laser data Ri and the ortho image Pi described above are stored in advance in the second storage unit 5 and stored in the first storage unit 4. Analysis area information, land cover classification information, and the like are generated in the second storage unit 5 and stored in the first storage unit 4.

  As shown in FIG. 2, the water land environment providing unit (program) 10 includes a sandbar relative height distribution calculation program 11 (hereinafter referred to as a state high classification map creating unit 11), a water surface gradient classification, and an image. River environment situation diagram creation consisting of Seto classification program 14 (hereinafter referred to as Seto classification unit 14) combined with analysis, riverbed particle size classification program 12 (hereinafter referred to as riverbed material information creation unit 12) by image analysis, etc. Unit 20, laser data editing unit 200 (program), reflection intensity map creation unit 21 (program), land cover classification map creation unit 22 (program), river channel analysis area creation unit 23 (program), and synthesis unit 24, etc., and canceling the slope of the river shown in FIG. 3 from the ortho image Pi and the laser data Ri, for example, the sand bar specific height image A or the rapids of the river shown in FIG. Etc. is SeraFukashi image B or the Kawado as shown in FIG. 5 sandbank glance, to obtain sand stone, the status of gravel it generates the bed material image C and the like at a glance the information output unit 3. In other words, by using the ortho image Pi and the laser data Ri, it is possible to obtain numerical data indicating the ratio of river channel waterfall, Nakashu, sand, and riverbed material. Therefore, it can be applied to the management and analysis of the river channel environment. The synthesizing unit 24 converts the ortho image Pi into a monochrome image (black and white), synthesizes the monochrome image and the image of the river channel area (may be either one of the images), and displays it on the display unit.

  As shown in FIG. 33, the laser data editing unit 200 includes a laser data classification unit 201, an elevation value data extraction unit 202, a vegetation height calculation unit 204, and a reflection intensity data extraction unit 203. The input laser data Ri includes an X coordinate value, a Y coordinate value, a ground elevation value Zg, a surface elevation value Zt, a ground reflection intensity Ig, and a surface reflection intensity It. Yes.

  The laser data classification unit 201 classifies the laser data Ri into laser ground data Rg (X, Y, Zg, Ig) and laser surface layer data Rt (X, Y, Zt, It), and stores the first storage unit 4. To store.

  Subsequently, the altitude value data extraction unit 202 extracts altitude value data from the classified laser ground data Rg, and the ground mesh data ER (X, Y, Zg) as a digital altitude model (DEM), and the laser surface layer. Elevation value data is extracted from the data Rt, and surface mesh data SR (X, Y, Zt) is generated as a numerical ground model (DSM) and stored in the first storage unit 4.

  Next, the vegetation height calculation unit 204 reads the ground elevation value Zg and the surface elevation value Zt at the same XY coordinate point from the ground mesh data ER and the surface layer mesh data SR, calculates the difference between them, and mainly Find the height of the plant growing at the point (vegetation height h). And it stores in the 1st memory | storage part 4 as vegetation height data HD (X, Y, h) with a coordinate value.

  On the other hand, the reflection intensity data extraction unit 203 extracts the reflection intensity data from the classified laser ground data Rg, obtains the ground reflection intensity data HRg (X, Y, Ig), and the reflection intensity data from the laser surface layer data Rt. The surface layer reflection intensity data HRt (X, Y, It) is generated and stored in the first storage unit 4.

  As shown in FIG. 34, the laser ground data Rg and the laser surface data Rt are respectively obtained when the measurement point when the laser is irradiated is the ground and the surface of an object (plant etc.) protruding from the ground. The data of As can be seen from FIG. 34, the difference between the elevation values (Zg and Zt) at the same point is the vegetation height (h).

  The reflection intensity diagram creating unit 21 assigns a color corresponding to the reflection intensity of each mesh of the surface layer reflection intensity data HRt to obtain the reflection intensity distribution diagram HRi of FIG.

  That is, by having the reflection intensity map creating unit 21, the water / land environment providing unit 10 obtains the reflection intensity distribution map HRi shown in FIG. 7 in addition to the ortho image Pi (RGB in xy) shown in FIG. It will be. In the reflection intensity distribution diagram HRi of FIG. 7, red is assigned to a mesh (pixel) having a high reflection intensity, yellow when the reflection intensity is between medium and strong, yellow with green, blue with weak, and between weak and medium. Is assigned light blue. These color assignments are made by determination based on threshold values. Even if there are shadows on the river surface and sandbar surface due to trees, the reflection intensity is the reflection intensity of the laser, so there is almost no influence of the shadow.

  The land cover classification map creation unit 22 analyzes the color information of the mesh of the ortho image Pi in a land cover map memory (not shown) defined in the orthophoto coordinate system, and classifies it according to this color. Thus, as shown in FIG. 8, a land cover classification map E classified into trees, water (river surface), bare land, grassland, and other features is obtained. At this time, since the shadow portion cannot be distinguished, the reflection intensity distribution diagram HRi of the laser data corresponding to the shadow portion is read and classified into water, bare land, grassland, or trees.

  Moreover, the land cover classification map creation part 22 can also obtain a vegetation height distribution map as shown in FIG. The vegetation height distribution map reads the vegetation height data HD into the vegetation height memory in the coordinate system corresponding to the mesh position in the dark green area of the land cover classification map E, and the color corresponding to the vegetation height (h) value of each mesh Is assigned to the vegetation height memory mesh to obtain the vegetation height distribution map of FIG.

  In the present embodiment, red is assigned to a tree having the highest tree height, yellow is assigned to a tree that is higher than the middle, yellow is assigned to the middle, and blue is assigned to the lower. However, when displaying on the display unit, the ortho image Pi is converted into monochrome, and the vegetation height distribution map information of the vegetation height memory is synthesized to obtain the image of FIG. As shown in FIG. 9, the tree can be clearly seen even if there is a shadow.

  Here, the land cover classification map creation unit 22 will be described in detail with reference to the flowchart of FIG.

  In the present embodiment, as shown in FIG. 11, laser light is emitted downward while flying horizontally over the target area from an airplane, and is calculated from the time required for reciprocation, the position, attitude, and launch angle of the aircraft ( The computer obtains x, y, and z of the ground surface. This laser can be fired at a high frequency of, for example, 33,000 times to 1,000,000 times per second, so that, for example, an elevation point (Rx, Ry, Rz) at a density of about one point from 50 cm to 1 m. Acquisition is possible. In addition, the same range as the above laser is photographed with a digital camera, and a high-resolution color (blue / green / red or green / red / near infrared) image of, for example, 20 cm above the ground is acquired, and is combined with the laser data Ri. An image Pi can be obtained.

  As shown in FIG. 35, for example, the data accuracy is 20 cm × 20 cm indicated by A in the case of the unit mesh section of the ortho image Pi, while 1 m indicated by B in the case of the unit mesh section of the laser data Ri. × 1 m. Thus, the unit mesh of the ortho image Pi is smaller than the unit mesh of the laser data Ri due to the difference in the data collection method. For this reason, information on the precise ortho image Pi with high accuracy is mainly used. However, in the shaded portion or the like, the covering may not be determined by the ortho image Pi. Therefore, by performing various determination processes by treating the information of the unit mesh section B of the laser data Ri including the unit mesh section A of the ortho image Pi as a value common to each of the unit mesh sections A of the ortho image Pi, The accuracy of the obtained topographic information is raised.

<Embodiment 1>
The land cover classification map creation unit 22 analyzes the high-precision ortho image Pi acquired by the digital camera stored in the first storage unit 4 by a decision tree method, and the land cover state (trees) in and around the river channel , Water, bare land, etc.). At this time, since the shadow portion may not be classified by normal image classification, accuracy is ensured by using information on laser reflection intensity (Intensity) acquired simultaneously with photographing. In the image classification, the vegetation height data obtained by laser is used to improve the grassland and trees with large errors.

  The ortho image Pi is obtained by sequentially recording the intensity of light of three bands of red, blue, and green or three bands of green, red, and near-infrared in order of x and y in a sequence of meshes (pixels) at regular intervals. It is.

  Further, the laser data Ri is given as (X, Y, Zg, Ig, Zt, It).

  First, the ortho image Pi in the storage unit is read, and a land cover classification map memory defined by the coordinate system of the ortho image Pi is generated.

  Then, the mesh number of the ortho image Pi is set, and the image data (Px, Py, R, G, B) of the mesh with this mesh number is read (S200).

  Next, the color intensity of the three bands of red (R), green (G), and blue (B) of the read image data is compared with the first threshold value T1 (S201). This threshold value T1 is a threshold value determined from the statistic (brightness distribution) of the entire image. In addition, which color is used for the determination may be appropriately determined from the statistic (brightness distribution) of the entire image.

  If it is determined in step S201 that the color intensity is smaller than the threshold value T1, it is determined to be a shade. In step S201, when it is determined that the value of the color intensity is larger than the threshold value T1, it is determined to be sunny.

  When the read mesh (pixel) is classified into a shaded area, the color intensity cannot determine what the point is, so the corresponding (Px, Py) mesh from the surface reflection intensity data HRt. (X, Y) mesh section data of the laser data including is read (S202).

  Then, the surface reflection intensity (It) of the section is compared with the second and third thresholds T21 and T22 determined from the statistics of the laser reflection intensity data (S203). If it is smaller than the threshold T21, the mesh (pixel) is set as category C = 1 (water area) and stored in the land cover classification map E (S204). If the value is between the threshold values T21 and T22, the category (pixel) is set to category C = 2 (bare ground) and stored in the land cover classification map E (S205). On the other hand, when larger than threshold value T22, it shifts to processing (S209-S212) for judging vegetation height (h) mentioned below.

  In step S201, when it is determined that the value of the color intensity is larger than the threshold value T1, and the read mesh (pixel) is classified into the sunny area, three bands of red (R), green (G), and blue (B) The correlation between the color intensities is determined (S206).

  For example, if the color intensity of blue (B) is greater than the values of other red (R) and green (G), set the mesh (pixel) as category C = 1 (water area), and the land cover classification map E is stored (S208). Also, if the color intensity of red (R) is greater than the values of other blue (B) and green (G), the category C = 2 (bare) is set for the mesh (pixel), and the land cover classification Save in FIG. E (S207).

  Here, meshes (pixels) determined that the color intensity of green (G) is greater than the values of other blue (B) and red (R) are determined as vegetation, and are classified further finely. The data of the (X, Y) mesh section of the laser data including the corresponding (Px, Py) mesh is read from the data HD (S209).

  Then, the vegetation height (h) of the section is compared with the fourth threshold value T3 of the height for classifying the plant (S210). For example, the threshold T3 is set to 50 when classifying as a tree if there is a height of 50 cm or more.

  If it is determined in step S210 that the vegetation height (h) is larger than the threshold value T3, category C = 3 (tree) is set in the mesh (pixel) and stored in the land cover classification map E (S211). . On the other hand, when it is determined that the vegetation height (h) is smaller than the threshold value T3, category C = 4 (grassland) is set in the mesh (pixel) and stored in the land cover classification map E (S212).

  Needless to say, the conditional expression of the classification based on the correlation of the color intensities of the three bands differs depending on the color of the photographed terrain. Therefore, the conditional expression is appropriately determined from the statistics (color intensity distribution) of the entire image.

  The above processing is repeated for all the mesh ortho images Pi by counting up and repeating the mesh numbers, and the entire land cover classification map E is created. By setting the value of category C in this way, for example, when classified as a water area (C = 1), blue is classified when classified as a tree (C = 3), and grassland (C = 4 ) And yellow when it is classified as bare (C = 2), and so on. This coloring can be performed by using a lookup table. The land cover classification map information generated in this way is stored in the first storage unit 4 and output to the output unit 3 to obtain the land cover classification map E shown in FIG. Since this land cover classification map E has no shadow portion, when the river channel is shaded by a mountain or the like, it is possible to determine the situation of the shadow location that is difficult to see with the orthophoto. In addition, about the above flow, you may change suitably the decision branch of a decision tree, the color to be used, etc. according to the color ortho image to be used.

  Furthermore, as shown in FIG. 12, it is possible to enclose a predetermined area, enlarge the predetermined area, extract an image of the predetermined area from the orthophoto image Pi, and simultaneously display it in the vicinity of the enlargement frame. In this way, the environment of the shadow area can be grasped more.

   For this reason, it can be seen at a glance whether there are high trees or low trees on the dike in the Nakasu of the vegetation height distribution map. For example, if there are trees in Nakasu, it will be covered more and more in the future, and as a result, the entire Nakasu is expected to be fixed. In other words, in the event of a flood, it can be used as a flood control measure because it can be expected to be a factor that stops river flow.

Next, the river channel analysis area creation unit 23 will be described.
The river channel analysis area creation unit 23 stores the trajectory traced by the operator in the land cover classification map E (FIG. 8) on the screen as the river channel area Fa (also referred to as a quay area: xy coordinates) (FIGS. 12 and 8). .

  Then, the image information of the land cover classification map E in the river channel area Fa is read (extracted from the land cover classification memory), and the uncovered land bare land area Fb (sand state in Nakashu, bare land on both banks) is extracted (river channel area) (The yellow area of the land cover classification map) is stored (FIG. 13b). Also, the image information of the land cover classification map E in the river channel area Fa (FIG. 13a) is read (extracted from the land cover classification memory), and the water surface area Fc is extracted (light blue area of the land cover classification map of the river channel area). And memorize (not shown).

(Calculation of specific height distribution of sandbar relative to water surface)
Next, the calculation of the specific height distribution of the sand bar relative to the water surface will be described.

  In calculating the specific height distribution of the sand bar with respect to the water surface, the water surface elevation is calculated at arbitrary intervals along the flow path from the laser data Ri, and is adjusted to a certain reference surface. That is, the river channel elevation data is corrected so that the water surface gradient is virtually zero, the specific distribution of the sand bar with respect to the reference surface is calculated, and the stepwise display is performed as necessary. In a typical stage chart without this correction, the slight relative height of the sand bar relative to the water surface is difficult to discern due to the slope of the river, but with this method, the relative height of the sand bar from the water surface at all points on the flow path is difficult. Since it is expressed, it is possible to know the flooding frequency at the time of flooding that defines the river environment (see FIG. 3).

  FIG. 14 is a flowchart for explaining the outline of calculating the specific height distribution of the sand bar with respect to the water surface.

  As shown in FIG. 14, first, data is read from the land cover classification map E (S400). Then, a mesh (Px, Py) of category C = 1 is extracted to obtain a water area (S401).

  Next, the mesh section information of the laser data including the water area mesh is read from the ground mesh data (DEM) ER (S402). Thereby, the elevation value (Zg) of the water area mesh is obtained.

  An elevation value (Zg) is set as the coordinate value of the water area mesh, and a water surface gradient division map M is created (S403).

Subsequently, an inequality triangular network (TIN) is generated from the water area mesh, and the elevation value (Zs) of the reference water surface is calculated (S404). An image of a specific calculation method is shown in the water section S. Three arbitrary points on the edge whose actual elevation is known (in the water section S, only two points are displayed because of the cross section) are extracted, and a triangle having the three points as vertices is created. The altitude at each point on the triangular surface is the reference elevation value (Zs) of the water surface. FIG. 32 shows a state where a TIN is generated on an actual image.

  Then, the difference between the altitude value (Zs) of the reference water surface and the altitude value (Zg) of the water area mesh is calculated (S405), and a sandbar specific height distribution map A is created (S406). This tells you how high the sandbar is from the surface of the water.

  FIG. 15 is a flowchart for explaining the calculation of the specific height distribution of the sand bar relative to the water surface according to another embodiment.

  The state classification classification creating unit 11 of the river environment situation diagram creating unit 20 corresponds to an area Fd (bare land, grassland, tree area; hereinafter referred to as a specific height distribution providing area Fd) other than the river surface area Fc of the river channel area Fa. Read Z of laser data having information. Then, the color value corresponding to the Z value is assigned to the mesh of the specific height distribution providing area memory having the same coordinate system as that of the ortho image Pi. Red is assigned when the threshold value a is equal to or greater than the threshold value a, yellow is assigned between the threshold value a and the middle threshold value b, green is assigned when the threshold value is lower than the threshold value b and the threshold value c is greater than the threshold value, and blue is assigned.

  That is, by displaying the image information of the specific height distribution providing area memory, the stage chroma H before gradient cancellation shown in FIG. 16 is obtained. In this figure, as the river goes upstream, the river becomes more red and the downstream becomes more blue. As shown in FIG. 16, although the height of the state area is not known, it will become clear if it is as follows. That is, as shown in FIG. 32, from the DEM (reference plane) of the river surface of the ground mesh data ER, TIN is generated for the region other than the river surface, and X, Y, and Z of these TINs are extracted (the river surface Reference DEM).

  Then, the elevation of the bare vegetation on the riverbank in the state shown in FIG. 3 is extracted by obtaining the difference between each DEM Z value (laser data) in the river channel area and each DEM Z value based on the river surface.

  In FIG. 3, colors corresponding to the Z values of states and the like are assigned so that the height of the state can be visually recognized.

  Further, as shown in the flowchart of FIG. 15, as another embodiment, it may be obtained as follows.

  In FIG. 15, the operator draws “Miosuji” in the river of the stage chart before the slope cancellation, reads the “water area” laser data Mpi in the land cover classification map E of this “Miosuji”, and at the inputted interval Mti, The Z value is read (S51).

  In addition, the Z value of the laser data at the downstream end of “Miosuji” on the screen is read and used as a reference value BZi (S52).

  Then, a correction function βi for obtaining the Z value of each interval Mti as the reference value Bzi is obtained (S53).

  Next, the mesh number Kmpi of the stage chroma H (specific height distribution providing area memory) before the gradient cancellation is set (S54).

  Next, a water surface gradient zero diagram memory for obtaining a water surface gradient zero diagram is defined in the same coordinate system as the stepped chroma diagram H before the gradient cancellation (S55).

  Then, the Z value of the stage chroma H (priority distribution providing memory) before the gradient cancellation of the mesh number Kmpi set in step S54 is read, and this Z value is corrected to the reference value Bzi based on the correction function βi (FIG. 17). Data Rbi is obtained (S56).

  Next, the data Rbi is stored in the river channel area memory (S57). That is, data with the gradient as the reference value is created in the mesh number Kmpi of the stage chroma H before the gradient cancellation.

  Next, it is determined whether or not the Z value of the last mesh Kmpi in the stage chart H before the gradient cancellation has been corrected to the reference value (S58). If the correction for the last mesh Kmpi has not been completed, the mesh Kmpi is determined. Is updated to the next mesh, and the process returns to step S54 (S59).

  If it is determined in step S58 that the Z values of all the meshes Kmpi in the stage chart H before the gradient cancellation have been corrected to the reference value, the region data kpi other than the river surface area from the water surface gradient map (memory for the river channel area diagram). Is extracted (S60).

  That is, only bare land, grassland, and tree areas in the river channel area are extracted.

  Next, the z value of the extracted data kpi and its mesh number Kmpi are read and assigned to the river area memory of this mesh number (defined in the water surface gradient zero diagram) (S61).

  Next, the Z value of each mesh Kmpi is read, a color corresponding to the Z value is allocated (memory for the river channel area) (S62), and is synthesized into an ortho and output to the screen (S63). Red is assigned when the Z value is greater than or equal to the threshold value Ab, yellow is assigned when the threshold value is greater than or equal to abc and less than or equal to Ab, yellow and green are assigned when the value is less than abc and greater than or equal to acd, and blue is assigned. That is, the specific height distribution diagram A shown in FIG. 3 can be obtained. This coloring can be performed by using a lookup table.

(Se / Kashiwa)
Next, the Se / Kashiwa classification unit 14 will be described. FIG. 18 is a flowchart for explaining the outline of Seto image generation.

  The Seze classification part 14 classifies the water surface gradient from the elevation value (Z value in the river surface area Fa) of the water surface of the laser data Ri acquired by the aviation laser. Furthermore, the wave surface state of the water surface is evaluated based on the texture of the orthophoto image Pi, and by superimposing these, the steep water surface waved part of the rapids and the smooth water surface of the ridge and the smooth water surface are accurately distributed. Express well (see Figure 4).

  Regarding the gradient of the water surface, it may be difficult to obtain the water surface elevation directly depending on the laser measuring instrument. In that case, the elevation value of the shoreline (the frame of the water surface area) may be substituted.

  In addition, when evaluating the ripples on the water surface from an aerial image, the wave condition is not known simply by evaluating the brightness of the image because it is affected by water quality, bottom sediment, and reflection of sunlight, but in this embodiment, The image of the water part that has been converted into a single band by averaging three bands (RGB) is separated into high frequency / low frequency components.

  By calculating the distribution of the dispersion value using the moving window, it is possible to extract only the ripples reflected in the high-frequency components of the image and to perform relative classification (if low-frequency / high-frequency separation must be obtained). , Picking up changes in luminance information due to other than waves). This processing flow is shown in FIG.

  The Seto classification unit 14 reads the mesh of the ortho image Pi in the river surface area Fc (S70), and sequentially obtains the average of these three bands (RGB) to obtain a monochrome image (S71) and (S72). Then, the local average is sequentially obtained by applying the window wi to the monochrome image, and these low frequency component images are obtained (S74).

  Next, the difference between the monochrome image obtained in step S71 and the low frequency component image obtained in step S74 is obtained (S75) to obtain a high frequency component image (S76). That is, the wave component is extracted.

  Next, by obtaining the local standard deviation of the high-frequency component image, a raw image (also referred to as a roughness image) representing the water surface ripple condition is obtained (S78). Then, an average value (μ) and a standard deviation value (σ) of the whole drawing image are calculated (S79). Next, with reference to these values, the value of each pixel in the drawing is equally divided within a certain range (for example, 0 to μ + 2σ) to obtain a wave condition division diagram in which the wave condition is relatively divided into 16 gradations ( S80), (S81).

  In order to obtain the Hirose image, first, the wave condition is discriminated. FIG. 19 is a flowchart for explaining the swell state determination.

  The river surface area Fc and the ortho image Pi are compared, and the ortho image Pci of the river surface area Fc is read (S71). Next, average processing of three bands (RGB) is performed for each mesh position (pixel) Mpi of the ortho image pci (S72) to obtain a monochrome image Pm in the river surface area.

  Next, a window wi of a predetermined size is applied to the monochrome image Pm in the river surface area (S73), and the monochrome luminance value of the mesh in this window is averaged (low wave frequency component extraction) (S74). The results are sequentially stored in the memory (S75). In the present embodiment, this is called a river surface low frequency component image EWi.

  Next, it is determined whether or not the averaging process has been performed on the monochrome image of the last-numbered window wi (S76). When it is determined that the averaging process is not performed on the monochrome image of the last window, the position (i, j) of the window wi is updated to the next position (S77), and the process returns to step S73.

  If it is determined in step S76 that the averaging process has been performed on the monochrome image in the last window, the mesh position Mpi is set (S78).

  Then, by subtracting the luminance value of the low frequency component image EWi of the mesh Mpi from the luminance value of the orthophoto image Pcim of the river surface of the mesh Mpi, a high frequency component image of the mesh is obtained (S79), and this is stored in the memory. Store sequentially (S80). In other words, an image of the rippling part of the river surface was obtained.

  Next, as shown in the flowchart of FIG. 19, it is determined whether or not the last wave component of the mesh Mpi has been extracted (S81). If it is determined in step S81 that the last wave component of the mesh Mpi has not been extracted, the process returns to step S78.

  If it is determined in step S81 that the wave component of the last mesh Mpi has been extracted, the wave component image Qi stored in step S80 is multiplied by the window wi (S83), and the standard of each pixel in the window wi is determined. A deviation (roughness map image) is obtained (S84) and stored in memory (roughness image RWi) (S85).

  Next, it is determined whether or not a drawing image of the window at the last position has been obtained (S86). When the drawing image of the last window is not obtained, the window position is updated to the next position, and the process returns to step S83 (S87).

  If it is determined in step S86 that the base image of the window at the last position has been obtained, the roughness image RWi in the memory is multiplied by the window wi (S88), and the standard deviation is obtained by averaging (S89). Conversion is performed (S90). The value of each pixel of the raw map is equally divided within a certain range (for example, 0 to μ + 2σ) to obtain a wave condition division diagram in which the wave condition is relatively divided into 16 gradations.

  Next, it is determined whether or not the wave condition QRi of the last window wi has been obtained (S91). If it is determined in step S91 that it is not the last, the window position is updated to the next position, and the process returns to step S89.

  Next, the color separation processing of the Se and Hanga classification chart QRi will be described with reference to FIGS. 21, 22, and 23. FIG.

  As shown in FIG. 21, the mesh position Mpi of the orthoimage Pci of the river surface is set (S100). Next, the luminance eh of the ripple component image Qi of the mesh Mpi is read (S101), and this eh is compared with the ripple threshold Ni (S102).

  In step S102, if eh is larger than the threshold value na, which indicates that the ripple is high, the presence of the ripple is identified in the ripple distribution map memory (same as the coordinate system of the river surface area, which is generated at the time of instructing the calculation of the stream / soil). After defining (S103), the process proceeds to step S107 described later.

  If it is determined in step S102 that eh is less than or equal to na, it is determined whether or not nb <eh <na (S104).

  If it is determined in step S104 that eh is nb <eh <na, a medium wave identification code is defined in the area of the wave distribution map memory at the position corresponding to the mesh (S105).

  If it is determined in step S104 that eh is not nb <eh <na, an identification code for no ripple is defined in the area of the ripple distribution memory at a position corresponding to the mesh (S106).

  In step S107, it is determined whether or not the last mesh Mpi has been identified (S107). If it is determined that identification has not been performed, the mesh Mpi is updated and the process returns to step 100 (S108).

  Next, as shown in FIG. 22, the mesh number Mpi is set for the elevation data (DEM) of the river surface generated based on the laser data (S110).

  Next, the laser data Zi of the mesh Mpi is read from the elevation data (DEM) of the river surface (S111).

  Next, based on the flow direction, the z value of the laser data at the next mesh position is read from the elevation data (DEM) of the river surface (S112), the difference between the two is obtained, and this is used as the water surface gradient KGi (S113). .

  Then, the water surface gradient KGi is compared with the threshold value Li (S114). When it is determined that KGi is equal to or greater than the threshold value Li, a steep slope identification code is assigned to EPi in the water surface gradient map memory (corresponding to the river surface area: created at the time of Seto processing) (S115), and processing is performed in step S117 described later. Move.

  If it is determined in step S114 that KGi is equal to or smaller than Li, a mesh that is not steep is assigned to the water surface gradient map Epi (S118).

  Then, it is determined whether or not the last mesh MPi has been assigned. If no assignment has been made, the mesh position is updated, and the process returns to step S110.

  Next, as shown in FIG. 23, the mesh ei of the wave component image Qi (wave component information memory) generated based on the ortho image, and the water surface gradient map Si (water surface gradient memory) generated based on the laser data The mesh epi is set (S121).

  Next, the identification code assigned to the meshes epi and ei is read (S122). Then, it is determined whether or not the mesh ei has a ripple and the epi has a steep slope (S123).

  If it is determined in step S123 that ei is rippled and identification is assigned to epi as abrupt, it is determined as Hayase (S124).

  Next, Hayase's identification is assigned to the area of the memory for the map and map at the position corresponding to ei or eip (generated in response to the command to generate the map image) (S125), and the process proceeds to step S126. Transfer.

  If it is determined in step S123 that ei has a wave and epi does not have an abrupt identification, it is determined whether an identification that is waved and epi is not abrupt is assigned (S128).

  If it is determined in step S128 that there is a wave and it is not abrupt, Hirase's identification is assigned to the memory for the map (S129).

  If it is determined in step S128 that there is no undulation and it is not abrupt, the identification of the cocoon is assigned to the memory for drawing the map (S130).

  In step S126, it is determined whether or not a determination result has been assigned to the last kei in the memory for drawing images (S126). If it is not the last time, kei is updated and the process returns to step S121 (S127).

  If it is determined in step S126 that an identification result has been assigned to the last kei, this Hirose chart table is read and displayed on the screen (S131). That is to say, the Seto classification diagram of FIG. 24 is obtained by this processing.

  FIG. 24 shows a case where a photo is assigned to a specific area. In other words, a photo (or a moving image) is allocated to a specific area of the set / 淵 image and stored in the memory, and when the specific area is selected, the photo (or moving image) is displayed together with the set / set classification chart from the memory. As a result, as shown in FIG. 25, it is possible to understand the situation of the rivers such as the rapids, the hills, and the rivers, which is effective in grasping the river environment.

  In addition, by applying this wave condition classification technique to the coast as well as the river, it is possible to grasp the water depth distribution in the coastal area.

  In combination with laser measurement at low tide, it can be applied to grasp coastal erosion, which has become a problem in various places in recent years.

<Embodiment 2>
(River bed material)
FIG. 26 is a flowchart for explaining the display of the classification image of the riverbed material. The riverbed material information creation unit takes advantage of the fact that the particle size distribution of the riverbed material is reflected in the texture of the image, and uses the average of the three bands of the color image to produce a monochrome image (hereinafter referred to as a single band image). The particle size of the riverbed material is relatively divided by calculating the dispersion value using the moving window.

  Furthermore, by selecting representative points for each classified hierarchy and confirming the particle size composition on site, the particle size distribution of the riverbed material over a wide area will be clarified.

  Similar to the festival and Seto processing, the image is separated into high and low frequency components, and only the former is targeted, resulting in changes in texture due to factors other than fine particle size (such as differences in wet and dry conditions and shadows of trees). Has reduced the impact.

  In addition, the present invention can be applied not only to rivers but also to coastal areas, and the particle size distribution of the beach may be grasped and compared with the river that is the supply source.

  That is, the riverbed material information creation unit 41 reads the ortho image Pi of the bare area, performs an average of the three bands (S131, S132), and obtains a single band image (monochrome image) (S133).

  Next, a local average (filter size 1 m) of the single band image is obtained, and a low frequency component image is created (S135).

  Then, a difference between the single band image and the low frequency component image is obtained (S136), and a high frequency component image is obtained (S137).

  Next, a local standard deviation (filter size is, for example, 1 m) of the high-frequency component image is obtained (S138), and a riverbed material map image is created (S139). Next, a raw image is used to linearly convert a certain range (for example, [0, μ + σ]) to [0, 15] (S140, 141), and the riverbed material (particle size) image (μ) shown in FIG. And [sigma] create a roughness image average and standard deviation (S142). The color values are assigned by the same processing as that shown in FIGS. 21 and 22 described in the above-mentioned steps for setting the color distribution for obtaining the riverbed material image.

  When applying normal filters, shadows and boundaries (edges) between different land covers are emphasized, but this effect is reduced by removing low-frequency components.

  By such processing, as shown in FIG. 27, the color is displayed according to the particle size of the material of the sand bar, and when the photograph is linked to this point, the photograph of this place is displayed as shown in FIG. It will be.

  The sandbar wet-wet map creation unit assumes that the riverbed material is uniform, and uses the fact that the wetter sandbar has lower lightness, averages the three bands, and clarifies the water status of the sandbar by the brightness of the single band.

   For this reason, as shown in FIG. 29, the sandbar wet / dry diagram creation unit 42 reads the ortho image of bare ground (S151), averages RGB (three bands) of this image (S152), and obtains a single band image. (S153). Next, the average and standard deviation of the sand bar are calculated (S154), linearly deformed (S155), and the sand bar moisture image shown in FIG. 30 is obtained (S155). The value of each pixel of the raw map is equally divided within a certain range (for example, 0 to μ + 2σ) to obtain a wave condition division diagram in which the wave condition is relatively divided into 16 gradations.

  The image shown in FIG. 30 is obtained by performing such processing. In the image shown in FIG. 30, the orthophoto is converted into black and white except for the sandbar area, and the composition unit displays it with the color image of the sandbar.

<Embodiment 3>
The above results are comprehensively managed and analyzed by GIS, and the items shown in Fig. 31 are aggregated for the following indicators for each unit distance along the flow path. Clarify sections where deterioration is progressing.

・ Ratio of riverbanks and forest land in the river channel ・ Representative grain size of riverbed materials ・ Forestation index (forest value area x tree height)
-Water surface area-Forestry potential distribution (overlay of sandbar specific height and particle size distribution)
-Distribution of habitat potential of various organisms The present invention naturally includes various embodiments not described herein. The technical scope of the present invention is determined only by the invention specifying matters according to the scope of claims reasonable from the above description.

It is a hardware block diagram of the terrestrial environment condition provision system using the laser data and color image by an aircraft. It is a schematic block diagram explaining this invention. It is explanatory drawing of the sand bar specific height image which canceled the gradient of the river. It is explanatory drawing of the Seto image which understands Hayase of a river, a kite, etc. at a glance. It is explanatory drawing of a riverbed material image. It is explanatory drawing of an ortho image. It is explanatory drawing of a reflection intensity distribution figure. It is explanatory drawing of a land cover classification | category map. It is explanatory drawing of a vegetation height distribution map. It is a flowchart explaining a land cover classification map creation part in detail. It is explanatory drawing explaining collection of the laser data by an aircraft, and acquisition of an ortho image. It is explanatory drawing explaining the attached image in a land cover classification | category figure. It is explanatory drawing explaining a bank-outside area mask (river area) and a bank-outside bare field (river surface area). It is a flowchart explaining the outline of specific height distribution calculation of the sand bar with respect to the water surface. It is a flowchart explaining the specific height distribution calculation of the sand bar with respect to the water surface of other embodiment. It is explanatory drawing explaining the stage chroma before a gradient cancellation. It is explanatory drawing explaining cancellation of a water surface gradient. It is a flowchart explaining the outline of a Seto image generation. It is a flowchart explaining a ripple condition discrimination. It is a flowchart explaining a ripple condition discrimination. It is a flowchart explaining the color classification process of a stream and a haze image. It is a flowchart explaining the color classification process of a stream and a haze image. It is a flowchart explaining the color classification process of a stream and a haze image. It is explanatory drawing of a Seto image. It is explanatory drawing explaining the photographic image (a moving image is included) attached to the Seto image. It is a flowchart explaining the display of the classification image of a riverbed material. It is explanatory drawing explaining a riverbed material classification | category figure. It is explanatory drawing explaining the photograph image (a moving image is included) attached to the riverbed material classification | category figure. It is explanatory drawing explaining the process of the sandbar dry / wet map preparation part. It is explanatory drawing explaining a sandbar dry / wet classification chart. 10 is an explanatory diagram for explaining an analysis screen according to Embodiment 3. FIG. It is explanatory drawing explaining TIN provided with respect to a water surface. It is a block diagram explaining the detailed structure of a laser data editing part. It is explanatory drawing explaining ground laser data (DEM) and surface layer laser data (DSM). It is a comparison figure of the mesh division of an ortho image and laser data.

Explanation of symbols

1 Central information processing device (CPU section)
2 Information input unit 3 Information output unit 4 First storage unit 5 Second storage unit 10 Water and land environment provision unit 11 State high classification map generation unit 12 River bed material information generation unit 14 Seto classification unit 20 River environment situation diagram Creation unit 21 Reflection intensity map creation unit 22 Land cover classification map creation unit 23 River channel analysis area creation unit 24 Composition unit 200 Laser data editing unit 201 Laser data classification unit 202 Elevation value data extraction unit 203 Reflection intensity data extraction unit 204 Vegetation height calculation Part

Claims (14)

The features of the target area are classified by a color orthophoto image obtained by photographing the target area with an aircraft, laser data from the target area, DEM data and DSM data based on the laser data, and the result of the classification A feature environmental status providing method for displaying a land cover classification map on a screen,
A first threshold of color intensity for an arbitrary band of red (R), green (G), and blue (B) determined by the luminance distribution of the image data group of the color orthophoto image, and one image comparing the color intensity of the arbitrary band of the color orthophoto image data, the the sun when the first threshold value on more than, classifying the image data in the first time less than the threshold value shade And a process of
When the image data is classified as Hinata, the correlation between the color intensities of the three bands of red (R), green (G), and blue (B) determined by the color intensity distribution of all image data in the target area Is applied to the red (R), green (G), and blue (B) color intensities of the image data for the sun, and the image data for the sun is applied to water, bare land, and vegetation. If the image data is classified as a shadow, and the step of classifying the image data,
A second threshold value of the reflection intensity that becomes the boundary between the water area and the bare land, and a third threshold value of the reflection intensity that becomes the boundary between the bare area and the vegetation, determined by the reflection intensity distribution of the laser data of the target area; Comparing the surface layer reflection intensity included in the laser data corresponding to the image data classified as the shade, and when the surface layer reflection intensity is smaller than the second threshold value, the surface layer reflection is reflected in the water area. Classifying the bare ground when the intensity is equal to or higher than the second threshold and equal to or lower than the third threshold, and classifying as the vegetation when the surface reflection intensity is higher than the third threshold;
Reading the surface elevation value of the DSM data and the ground elevation value of the DEM data, and setting the difference as the vegetation height;
Comparing the vegetation height with a fourth threshold, classifying the tree when the vegetation height is greater than or equal to the fourth threshold, and classifying the grassland when the vegetation height is less than the fourth threshold;
A feature environment situation providing method comprising: repeatedly performing all the above steps for all image data of the color orthophoto image, and performing a land cover classification on the image data of the target area. In the designated river channel area, extracting the ground elevation value corresponding to the image data from the DEM data corresponding to the image data classified into the water area, respectively, and creating a water surface gradient section;
In order to calculate the first reference water surface, which is the virtual elevation when the first sand bar surrounded by the water area existing in the river channel area is the water surface, the ground elevation at one point at the edge of one water area The ground elevation values at two points on the edge of the other water area adjacent to the value are extracted, and a triangle having the three points as vertices is repeatedly generated to generate an unequal triangular network, and each triangular plane is defined as the first plane. The process of making the reference water surface of
Calculating the elevation value of the first reference water surface from the ground elevation value of the three points;
In order to obtain the second reference water surface, which is a virtual elevation when the second sandbar part surrounded by a water area on one side and the other side surrounded by a dike is the water surface,
Extracting two arbitrary points from one point at the edge of this water area and a point group generated on a two-dimensional dike line created separately, it is not possible to repeatedly create a triangle with these three points as vertices. to generate equal triangulation, a step of each triangle plane and the second reference surface of the water,
2. The feature environment situation providing method according to claim 1, further comprising the step of calculating an elevation value of the second reference water surface from the ground elevation values at three points on the second reference water surface. Calculating the difference between the elevation value of the first reference water surface in the river channel area and the ground elevation value of the first sand bar portion;
Calculating the difference between the elevation value of the second reference water surface and the ground elevation value of the second sandbar part;
3. The feature environment situation providing method according to claim 2 , further comprising the step of creating a sandbar specific height distribution map based on a difference between these ground elevation values. Obtaining a water surface gradient for each arbitrary section from the water surface gradient section diagram;
Extracting color orthophoto image data corresponding to the section to obtain a monochrome image by averaging three bands;
Extracting a low frequency component by averaging a monochrome luminance value over a window of a predetermined range with respect to the monochrome image;
Subtracting the low frequency component from the monochrome image to extract a high frequency component;
3. The feature environment state providing method according to claim 2 , further comprising the step of obtaining a water surface undulation state based on the water surface gradient and a local standard deviation of the high frequency component. Extracting color orthophoto image data corresponding to the first and second sandbar portions to obtain a monochrome image by averaging three bands;
Extracting a low frequency component by averaging a monochrome luminance value over a window of a predetermined range with respect to the monochrome image;
Subtracting the low frequency component from the monochrome image to extract a high frequency component;
3. The feature environment status providing method according to claim 2 , further comprising the step of determining the particle size of the river bed material of the first and second sandbar portions by the local standard deviation value of the high frequency component. A step of creating a color sand bar specific height distribution map in the river channel area by assigning an arbitrary color to the image data according to a difference value of a ground elevation value of each image data in the sand bar specific height distribution map is further performed. The feature environment situation providing method according to claim 3, wherein: The method further comprises the step of assigning an arbitrary color to the mesh according to the result of classification of the mesh of the color orthophoto image, creating a color land cover classification map of the target area and displaying it on a screen. Item feature environment status providing method according to any one of Items 1 to 6 . The features of the target area are classified by a color orthophoto image obtained by photographing the target area with an aircraft, laser data from the target area, DEM data and DSM data based on the laser data, and the result of the classification A feature environment situation providing program for displaying a land cover classification map on a screen,
On the computer,
A first threshold of color intensity for an arbitrary band of red (R), green (G), and blue (B) determined by the luminance distribution of the image data group of the color orthophoto image, and one image comparing the color intensity of the arbitrary band of the color orthophoto image data, the the sun when the first threshold value on more than, classifying the image data in the first time less than the threshold value shade Means to
When the image data is classified as Hinata, the correlation between the color intensities of the three bands of red (R), green (G), and blue (B) determined by the color intensity distribution of all image data in the target area Is applied to the red (R), green (G), and blue (B) color intensities of the image data for the sun, and the image data for the sun is applied to water, bare land, and vegetation. Means to classify,
When the image data is classified as a shade, the shade image data is
A second threshold value of the reflection intensity that becomes the boundary between the water area and the bare land, and a third threshold value of the reflection intensity that becomes the boundary between the bare area and the vegetation, determined by the reflection intensity distribution of the laser data of the target area; Comparing the surface layer reflection intensity included in the laser data corresponding to the image data classified as the shade, and when the surface layer reflection intensity is smaller than the second threshold value, the surface layer reflection is reflected in the water area. Means for classifying the bare fabric when the intensity is greater than or equal to the second threshold and less than or equal to the third threshold, and classifying the vegetation when the surface reflection intensity is greater than the third threshold;
Means for reading the surface elevation value of the DSM data and the ground elevation value of the DEM data, and setting the difference as the vegetation height;
Comparing the vegetation height with a fourth threshold, and classifying the vegetation height as a tree when the vegetation height is greater than or equal to the fourth threshold, and classifying as vegetation when the vegetation height is less than the fourth threshold. Means for reading out the surface elevation value and the ground elevation value of the DEM data, and setting the difference as the vegetation height,
Means for repeatedly performing all the above steps for all image data of the color orthophoto image, and classifying land cover for the image data of the target area;
Feature environment status providing program to execute the function as. In the computer,
Means for extracting a ground elevation value corresponding to the image data from the DEM data corresponding to the image data classified into the water area in a designated river channel area, and creating a water surface gradient section;
In order to calculate the first reference water surface, which is the virtual elevation when the first sand bar surrounded by the water area existing in the river channel area is the water surface, the ground elevation at one point at the edge of one water area The ground elevation values at two points on the edge of the other water area adjacent to the value are extracted, and a triangle having the three points as vertices is repeatedly generated to generate an unequal triangular network, and each triangular plane is defined as the first plane. Means to make the reference water surface of
Means for calculating an elevation value of the first reference water surface from the ground elevation values at the three points;
In order to obtain the second reference water surface, which is a virtual elevation when the second sandbar part surrounded by a water area on one side and the other side surrounded by a dike is the water surface,
Extracting two arbitrary points from one point at the edge of this water area and a point group generated on a two-dimensional dike line created separately, it is not possible to repeatedly create a triangle with these three points as vertices. Means for generating an equitriangular network, wherein each triangular surface is the second reference water surface;
Means for calculating an elevation value of the second reference water surface from the ground elevation values at three points of the second reference water surface;
The terrestrial environment situation providing program according to claim 9, for further executing the function as. In the computer,
Means for calculating a difference between an elevation value of the first reference water surface in the river channel area and the ground elevation value of the first sand bar portion;
Means for calculating a difference between an elevation value of the second reference water surface and the ground elevation value of the second sandbar part;
A means for creating a sandbar specific height distribution map based on the difference between these ground elevation values,
The terrestrial environment situation providing program according to claim 9 for further executing the function as. In the computer,
Means for obtaining a water surface gradient for each arbitrary section from the water surface gradient section diagram;
Means for extracting image data of a color orthophoto image corresponding to the section to obtain a monochrome image by averaging three bands;
Means for extracting a low frequency component by averaging a monochrome luminance value over a window of a predetermined range with respect to the monochrome image;
Means for subtracting the low frequency component from the monochrome image to extract a high frequency component;
Means for determining the water surface undulation state from the water surface gradient and the local standard deviation of the high frequency component;
The terrestrial environment situation providing program according to claim 9 for further executing the function as. In the computer,
Means for extracting color orthophoto image data corresponding to the first and second sandbar portions to obtain a monochrome image by averaging three bands;
Means for extracting a low frequency component by averaging a monochrome luminance value over a window of a predetermined range with respect to the monochrome image;
Means for subtracting the low frequency component from the monochrome image to extract a high frequency component;
Means for determining the particle size of the river bed material of the first and second sandbar portions by the local standard deviation value of the high-frequency component;
The terrestrial environment situation providing method according to claim 9, for further executing the function as. In the computer,
Means for assigning an arbitrary color to the image data according to the value of the difference in ground elevation value of each image data in the sand bar specific height distribution map, and creating a color sand bar specific height distribution map in the river channel area;
The terrestrial environment situation providing program according to claim 10, for further executing the function as. On the computer,
Means for assigning an arbitrary color to the mesh in accordance with the result of classification of the mesh of the color orthophoto image, creating a color land cover classification map of the target area and displaying it on a screen;
The terrestrial environment situation providing program according to any one of claims 8 to 13 , wherein the program further causes the function to be executed.