JP2016099316A - Feature change discrimination method, feature change discrimination device, and feature change discrimination program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a change discrimination device capable of accurately determining the change of a building between the previous fiscal year and the present fiscal year.SOLUTION: The discrimination device includes: a memory 380 for storing DSM image data in the previous fiscal year; a memory 370 for storing precision orthophoto image data in the previous fiscal year; a memory 280 for storing DSM image data in the present fiscal year; and a memory 270 for storing precision orthophoto image data in the present fiscal year. A computer performs: a step (500) of generating DSM/precision orthophoto composite image data in the previous fiscal year from the precision orthophoto image data in the previous fiscal year and the DSM image data in the previous fiscal year; a step (400) of generating DSM/precision orthophoto composite image data in the present fiscal year from the precision orthophoto image data in the present fiscal year and the DSM image data in the present fiscal year; and steps (900 and 1000) of displaying the DSM/precision orthophoto composite image data in the previous fiscal year and the DSM/precision orthophoto composite image data in the present fiscal year in a region allowing comparison between them on the screen.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a change determination method capable of accurately determining a change in a feature from DSM image data based on aerial photograph data of the previous year and aerial photograph data of the current year.

  Changes in fixed assets such as houses are identified in the precise orthophoto image (resolution: for example, 5.0 cm / pixel) taken in the previous year shown in FIG. 16A and in the current fiscal year shown in FIG. It is common for an operator to discriminate the difference between a photographed precise orthophoto image and the same screen. The above-mentioned transfer discrimination refers to changes such as extension and renovation of houses, extension and renovation of lands, and disappearance of houses.

  The precision orthophoto image is subjected to stereo matching processing on a stereo photograph taken by an aircraft, for example, in the same area of a city, to generate 3D coordinates in pixel units, and to perform 3D modeling.

  The above three-dimensional coordinates are obtained by taking a stereo photograph (image data) of the same region, measuring the photo coordinates of the reference point, etc., performing internal orientation, performing relative orientation, and further performing external orientation to each pixel. The three-dimensional coordinates are obtained.

  A three-dimensional model is generated by allocating three-dimensional coordinates (x, y, z), color information of a photograph, and the like to a mesh based on these three-dimensional coordinates.

  Then, the transformation is performed by rearranging the three-dimensional coordinates of the three-dimensional model in a pixel unit so as to be an orthographic projection with respect to the ground, and re-projecting as seen from above in the vertical direction at all points. I have a photo image.

  On the other hand, a house change interpretation support device disclosed in Patent Document 1 is disclosed as a means for determining a change in a building. This house change discrimination support device displays DSM data in two new and old periods or orthophoto images in two new and old periods, and displays a new anaglyph image to show it three-dimensionally.

JP 2012-146050 A

  However, in the conventional movement change discrimination, the precision orthophoto image of FIG. 16A of the previous year and the precision orthophoto image of FIG. The building is brown, and the building surrounded by the same red orthophoto image of this year's precision orthophoto image appears to be changing because the roof is white, but it looks like FIG. 16 (C) and FIG. 16 (D). As shown in the figure, it did not change in the 3D model. This is an influence due to a difference in shooting time and a difference in weather at the time of shooting.

  In other words, the building change cannot be accurately determined by comparing the previous orthophoto image with the current orthophoto image.

  On the other hand, since Patent Document 1 displays old and new DSMs on the same screen and discriminates changes in buildings by the operator, the burden on the operator is large.

  The present invention has been made to solve the above problems, and an object of the present invention is to provide a change discriminating apparatus capable of accurately discriminating changes in buildings between the previous year and the current year.

  In order to achieve the above object, the feature change determination method according to the present invention is characterized in that the previous year 3D model generated based on the aerial photograph data of the previous year taken by the aircraft over the area and the reference point is used. A first memory storing fiscal year DSM image data, a second memory storing orthophoto image data obtained by orthogonal transformation of the 3D model of the previous fiscal year as precise orthophoto image data of the previous fiscal year, and the region as the previous fiscal year A third memory storing the current year's aerial photograph data taken under the same conditions as when obtaining the current aerial photograph data and the current year's DSM image data of the current year's 3D model generated based on the reference point; The orthophoto image data obtained by orthogonally transforming the 3D model of the fiscal year was converted by a computer equipped with a fourth memory that stores the precise orthophoto image data of the fiscal year. A feature change determination method, wherein the computer reads out the previous year's precision orthophoto image data and the previous year's DSM image data, and generates the previous year's DSM / precise orthosynthesis image data. Reading out the current year's precision orthophoto image data and the current year's DSM image data, generating the current year's DSM / precision ortho-synthesized image data, and the previous year's DSM / precision ortho-synthesis image data and the current year's DSM The step of displaying the precise ortho-synthesized image data on the screen in an area where each can be compared.

  In order to achieve the above object, the feature change discrimination device according to the present invention is characterized in that the previous year's 3D model generated based on the aerial photograph data of the previous year taken by an aircraft over the area and the reference point is used. A first memory storing fiscal year DSM image data, a second memory storing orthophoto image data obtained by orthogonal transformation of the 3D model of the previous fiscal year as precise orthophoto image data of the previous fiscal year, and the region as the previous fiscal year A third memory storing the current year's aerial photograph data taken under the same conditions as when obtaining the current aerial photograph data and the current year's DSM image data of the current year's 3D model generated based on the reference point; A fourth memory storing orthophoto image data obtained by orthogonally transforming the 3D model of the fiscal year as precise orthophoto image data of the current fiscal year, and the previous year's precision orthophoto The image data and the previous year's DSM image data are read out, and the previous year's DSM / precise ortho-synthesized image data is generated from the previous year's DSM / precise ortho-synthesized image data. Read DSM image data and generate DSM / precision ortho-synthesized image data for this year's DSM / precision ortho-synthesized image data, and DSM / precision ortho-synthesized image data for the previous year and DSM / precision ortho-synthesized image for the previous year. And a display control unit for displaying data in an area that can be compared with each other on the screen.

  In order to achieve the above object, the feature change determination program according to the present invention is characterized in that the previous year's 3D model generated based on the aerial photograph data and reference points of the previous year taken over the area by an aircraft. A first memory storing fiscal year DSM image data, a second memory storing orthophoto image data obtained by orthogonal transformation of the 3D model of the previous fiscal year as precise orthophoto image data of the previous fiscal year, and the region as the previous fiscal year A third memory storing the current year's aerial photograph data taken under the same conditions as when obtaining the current aerial photograph data and the current year's DSM image data of the current year's 3D model generated based on the reference point; A computer having a fourth memory storing orthophoto image data obtained by orthogonally converting the 3D model of the year as orthophoto image data of the year A feature change determination program to be executed, the step of reading out the previous year's precision orthophoto image data and the previous year's DSM image data and generating them from the previous year's DSM / precision orthocomposite image data; Read out the current year's precise orthophoto image data and the current year's DSM image data, generate the current year's DSM / precision ortho-synthesized image data, and the previous year's DSM / precision ortho-synthesized image data and the current year's DSM / precision The computer executes the step of displaying the ortho-synthesized image data on the screen in an area where each can be compared.

  According to the present invention, it is possible to accurately discriminate changes in buildings between the previous year and the current year.

It is a block diagram explaining the concept of the feature change determination apparatus of this Embodiment. (A) is a figure showing an example of the current year's DSM / ortho composite image data, and (b) is a figure showing an example of the previous year's DSM / ortho composite image data. It is a schematic block diagram of the feature change discrimination | determination apparatus of this Embodiment. It is a block diagram explaining the structure of this year DSM and a precise ortho image creation part with which the feature change discrimination | determination apparatus of this Embodiment is provided. It is a block diagram explaining the structure of the previous year DSM and a precise ortho image creation part with which the feature change discrimination | determination apparatus of this Embodiment is provided. It is the schematic diagram which showed 3D modeling typically. It is the figure which showed an example of the produced | generated 3D model. (A) shows the building in the current year's DSM image in the ZX plane, and (b) shows the building in the current year's DSM image in the XY plane. It is an example of the table | surface which linked | related and memorize | stored the three-dimensional coordinate (x, y, z) currently allocated to the mesh of the DSM image of this year, and the gray scale value according to an altitude value (z). It is the figure which showed an example of the ortho image change detection table or the DSM change detection table. (A) is an example of displaying the change area superimposed on the previous year's DSM / precision ortho composite image data, and (b) is the change area superimposed on the current year's DSM / precision ortho composite image data. It is an example of a display screen. It is an example of a display screen in which the previous year DSM / ortho composite image and the current year's DSM / ortho composite image are superimposed and displayed in different colors. It is the flowchart which showed the process sequence by the elevation change extraction part of the feature change discrimination | determination apparatus of this Embodiment. It is an example of the table | surface which matched the mesh number, the altitude value (zi) of the previous year DSM image, the altitude value (zi) of this year's DSM image, and its determination result. (A) is an example of the previous year's precision orthophoto image (FIG. 15 (a)), (b) is an example of the current year's precise orthophoto image, and (c) is the previous year's DSM / ortho composite. It is an example of an image, (d) is an example of this year's DSM and ortho synthetic image. It is the figure which showed the conventional precision orthophoto image.

  FIG. 1 is a configuration diagram illustrating the concept of the feature change determination device of the present embodiment. In the present embodiment, the feature is described as a city building.

  As shown in FIG. 1, the feature change discriminating apparatus according to the present embodiment includes a current year DSM / precision ortho image creation unit 200, a previous year DSM / precision ortho image creation unit 300, and a current year DSM / precision ortho image synthesis unit. Unit 400, previous year DSM / precision ortho composition unit 500, ortho image change extraction unit 600, altitude change extraction unit 700, etc., and the previous year precise ortho photo image data OGb and current year precise ortho photo image data OGa. Are automatically extracted, and altitude change (Z value) in mesh units is automatically extracted from the previous year DSM image data Gb and the current year DSM image data Ga.

  The current fiscal year is, for example, the fiscal year 2013, and the previous fiscal year may be several years ago.

  The above-mentioned current year DSM / precision ortho image creation unit 200 reads the current year's aerial photograph data Ea and the current year's GCP information Ja (reference point) and performs the current year's 3D modeling to obtain the current year's 3D model Fa of the city. Then, based on the current year 3D model Fa, the current year precise orthophoto image data OGa and the current year DSM image data Ga are generated.

  The previous year's DSM / precision ortho image creation unit 300 reads the previous year's aerial photograph data Eb and the previous year's GCP information Jb and performs 3D modeling of the previous year to generate the city's previous year 3D model Fb. Based on the 3D model Fb, previous year precise orthophoto image data OGb and previous year DSM image data Gb are generated.

  Furthermore, this year's DSM / precision orthocompositing unit 400 superimposes this year's DSM image data Ga on this year's precise orthophoto image data OGa by raster synthesis of LVM (Logical Volume Manager) based on this year's GCP information Ja. This year's DSM / ortho composite image data AGa shown in FIG. 2A is generated and displayed on a screen (not shown).

  The previous year DSM / precise ortho image composition unit 500 superimposes the previous year's DSM image data Gb on the previous year's ortho photo image data OGb by LVM (Logical Volume Manager) raster composition based on the previous year's GCP information Jb. The previous year DSM / ortho combined image data BGb shown in FIG. 2B is generated and displayed next to the image of the current year DSM / ortho combined image data AGa.

  The ortho image change extraction unit 600 obtains a color difference between the current year's accurate orthophoto image data OGa and the previous year's accurate orthophoto image data OGb. If this color difference is equal to or greater than the threshold value αp, there is a change. The data OGa is stored in association with the mesh number.

  The elevation change extraction unit 700 obtains a difference between the mesh unit elevation (Z value) of the current year DSM image data Ga and the previous year DSM image data Gb, and changes if the elevation difference is equal to or higher than the elevation threshold value αq. The data is stored in association with the mesh number of the current year's DSM image data Ga.

<Specific example>
FIG. 3 is a schematic configuration diagram of the feature change determination device 100 according to the present embodiment. As shown in FIG. 3, the computer main unit 10 includes the above-mentioned current year DSM / precision ortho image creation unit 200, the previous year DSM / precision ortho image creation unit 300, the current year DSM / precision ortho image creation unit 400, In addition to the previous year DSM / precision orthocompositing unit 500, the ortho image change extracting unit 600, and the elevation change extracting unit 700, it includes a change area determining unit 800, a year-by-year display combining unit 850, a mode determining unit 900, and the like. Has a program. These programs are stored in a ROM (not shown) by a CPU (not shown), read into a program execution RAM (not shown), and processed by the above-described units. Is displayed on the screen 15 by the display control unit 1000. Each process is executed using the memory 270, the memory 280, the memory 370, the memory 380, the memory 410, the memory 510, the memory 610, the memory 710, and the like.

(Description of each part)
FIG. 4 is a configuration diagram illustrating the configuration of the current year DSM / precision ortho image creation unit 200 included in the feature change determination device of the present embodiment.

  As shown in FIG. 4, this year's DSM / precision ortho image creation unit 200 includes programs such as this year's 3D modeling processing unit 230, current year's precise ortho generation processing unit 250, and this year's DSM processing unit 260. have. These execute processing using the memory 210, the memory 220, the memory 270, and the memory 280.

  The memory 210 stores digital aerial photograph data Ea of the current year in which a predetermined area is photographed by an aircraft. The aerial photograph data Ea has a resolution of 25 cm or less (2 cm, 5 cm,...).

  In addition, the memory 220 stores GCP (Ground Control Point) at a plurality of points in a predetermined area.

  This year's 3D modeling processing unit 230 reads the overlapping two aerial photograph data in the memory 210 and the GCP in the memory 220, and copes with pixels that photograph the same point in the two color photograph data in the same region. The camera's shooting position and orientation (external orientation) are estimated based on the points and reference points, and the three-dimensional coordinates of each pixel (refer to FIG. 6) are obtained using the parallax calculation formula using the parallax. Further, noise removal is performed, and the color information of the corresponding points is given to the mesh of the three-dimensional model based on these three-dimensional coordinates, and the current year's 3D model (see FIG. 7) is generated in the memory 240. That is, the parallax is calculated with respect to the same point of the object shown in a plurality of photographic images, and a three-dimensional model of the object is reconstructed. Specific software uses SfM (Structure From Motion) or the like. This city's three-dimensional model is assigned photo color information, and is referred to as a three-dimensional model this year.

  The current year's precise ortho generation processing unit 250 converts the current year's three-dimensional model in the memory 240 into an orthographic image and generates an orthophoto image. This orthophoto image is referred to as a precision orthophoto image OGa this year.

  The current-year DSM generation processing unit 260 reads the current-year three-dimensional model in the memory 240, generates a DSM (Digital Surface Model), and stores the DSM (Digital Surface Model) in the memory 280. In the present embodiment, this is referred to as this year's DSM image Ga.

  That is, the current year DSM image Ga of the present embodiment obtains the DSM shown in FIG. FIG. 8A shows a building in the current year DSM image Ga in the Z-X plane, and FIG. 8B shows a building in the current year DSM image Ga in the XY plane. As shown in FIGS. 8A and 8B, since SfM is used to construct the three-dimensional model, the edge portion is firmly reproduced.

  FIG. 9 shows the three-dimensional coordinates (x, y, z) assigned to the mesh (m1, m2,...) Of the current year DSM image Ga and the gray scale value corresponding to the altitude value (z). It is shown in a table. For example, the gray scale value is represented by a brightness of 0 to 255, and the brightness is set higher as the elevation value (z) is higher, and the brightness is set lower as the elevation value (z) is lower.

  FIG. 5 is a schematic configuration diagram of the previous year's DSM / precision ortho image creation unit 300. The previous year DSM / precision ortho image creation unit 300 shown in FIG. 5 has programs such as the previous year 3D modeling processing unit 330, the previous year precision ortho generation processing unit 350, and the previous year DSM processing unit 360. ing. These use the memory 310, the memory 320, the memory 340, the memory 370, and the memory 380 to execute the same processing as the precise ortho / DSM creation unit 200 this year. Since the processing of each part is the same as that of the precision ortho / DSM creation part 200 this year, the description is omitted.

  Returning to FIG. 3, the current year's DSM / precise ortho-synthesis unit 400 includes the current year's DSM image Ga of the memory 270 generated by the current year's DSM / precise ortho image creation unit 200, and the current year's precise orthophoto image OGa of the memory 280. The current DSM image Ga is superimposed on the current precise orthophoto image OGa and synthesized, and this is temporarily stored in the memory 410 and displayed on the screen 15 as the current DSM / ortho composite image AGa (FIG. 2). reference).

  The previous year's DSM / ortho combining unit 500 reads the previous year's DSM image Gb in the memory 370 generated by the previous year's DSM / precise ortho image creation unit 300 and the previous year's accurate orthophoto image OGb in the memory 380 and reads the previous year's precision The previous year DSM image Gb is superimposed on the orthophoto image OGb and synthesized, and this is temporarily stored in the memory 510 and displayed on the screen 15 as the previous year DSM / ortho composite image BGb (see FIG. 2).

  The ortho image change extraction unit 600 includes the current year precise orthophoto image OGa in the memory 280 generated by the current year DSM / precision ortho image creation unit 200 and the previous year DSM / precision ortho image creation unit 300 in the memory 380. Compare the previous year's precision orthophoto image OGb and extract the change. This compares the color information for each mesh unit, and determines that there is a change if this color difference is greater than or equal to a predetermined value (color threshold). Then, a table (ortho image change detection table) as shown in FIG. 10 in which “1” with change or “0” without change is written for each mesh unit mi is stored in the memory 610. Here, the color information can be applied in any color space such as RGB, CMYK, HSV.

  The elevation change extraction unit 700 includes the current year DSM image Ga generated by the current year DSM / precision ortho image creation unit 200 and stored in the memory 270, and the previous year DSM / precision ortho image creation unit 300 generates the memory 370. Are compared with the previous year DSM image OGb stored in, and the change is extracted. Specifically, this calculates the difference in elevation value (z) for each mesh unit, and determines that there is an elevation change when this difference is greater than or equal to a predetermined value (elevation threshold). Then, a table (DSM change detection table) as shown in FIG. 10 in which “1” indicating the change in altitude or “0” indicating no change is written for each mesh unit mi is stored in the memory 710.

  As illustrated in FIG. 10, the change region determination unit 800 reads the DSM change detection table in the memory 710 or the ortho image change detection table in the memory 610, and the range where “1” is adjacent and continuous is equal to or greater than the threshold value. Are determined as a DSM change region or an ortho image change region. For example, as shown in a region 721 in FIG. 10, when the number of meshes in a region where “1” is adjacent and continuous is greater than or equal to a predetermined number, the region 721 is determined as a continuous region, while in FIG. As shown in regions 722 and 723, when “1” is not continuous but is isolated or adjacent to each other but the number of meshes in the region is less than a predetermined number, this region 722 and 723 is not displayed. It is determined as a continuous area. In this way, by detecting the continuous region or the non-continuous region, it is possible to remove the erroneous detection location due to noise.

  Then, as shown in FIG. 11A, the change area determination unit 800 adds, for example, an elevation difference change presence / absence table or an ortho image change detection table to the previous year's DSM / precision ortho composite image data stored in the memory 510. A continuous area in which identifiers with changes are continuously displayed is displayed as a surrounding change area.

  Further, as shown in FIG. 11B, similarly, the change area determination unit 800 adds the elevation difference change presence / absence table or the ortho image change detection to the current year's DSM / precision ortho composite image data stored in the memory 410. A continuous area in which identifiers with a change in the table are continuous is displayed as a surrounding change area.

  The year-by-year display compositing unit 850 displays the previous year DSM / ortho composite image and the current year DSM / ortho composite image by overlapping them with different colors. For example, as shown in FIG. 12, a DSM image expressed in gray scale is converted from the previous year's DSM / ortho composite image so that it is expressed in lightness of red (R), and red becomes stronger as the altitude is lower. To be. On the other hand, of the current year's DSM / ortho composite image, the DSM image expressed in gray scale is converted so as to be expressed with the lightness of blue (B), and the blue becomes stronger as the altitude is lower.

  Based on the instructions input in a dialog box (not shown), the mode determination unit 900 is configured to display the current year DSM / precision ortho image creation unit 200, the previous year DSM / precision ortho image creation unit 300, and the current year DSM / precision ortho image synthesis unit. 400, the previous year DSM / precision orthocompositing unit 500, the ortho image change extracting unit 600, the altitude change extracting unit 700, the change region determining unit 800, the year-by-year display combining unit 850, and the like are activated.

  For example, in the case of displaying the previous year 3D model, the previous year 3D model created by the previous year precise ortho / DSM creation unit 300 is displayed on the screen 15 by the display control unit 1000 as the previous year 3D model display mode.

  In the case of displaying the current year's 3D model, the current year's 3D model created by the current year's precision ortho / DSM creating unit 200 is displayed on the screen 15 by the display control unit 1000 as the current year's 3D model display mode.

  In the case of displaying the previous year's DSM / ortho composite image, the previous year's DSM / ortho composite image 500b is activated as the previous year's DSM / ortho composite image display mode, and the previous year's DSM / ortho composite image BGb is displayed. Is displayed on the screen 15.

  In the case of displaying the current year's DSM / ortho composite image, this year's DSM / ortho composite image display unit 400 is activated as the current year's DSM / ortho composite image display mode, and the current year's DSM / ortho composite image AGa is displayed. Is displayed on the screen 15.

  The display control unit 1000 displays an image based on the mode determined by the mode determination unit 900 on the screen with the designated color information. Further, based on the determination of the mode determination unit 900, the image of the current year is displayed below or next to the image of the previous year.

  The elevation change extraction unit 700 of the feature change determination device configured as described above will be described in detail with reference to the flowchart of FIG.

  The altitude change extraction unit 700 includes the current year DSM image Ga generated by the current year DSM / precision ortho image creation unit 200 and stored in the memory 270, and the previous year DSM / precision ortho image creation unit 300 generated and stored in the memory 370. The assigned DSM image Gb of the previous year is allocated (S1).

  Next, the mesh number mi of the current year DSM image Ga is designated (S3). Then, the difference between the altitude value (z value) associated with the mesh number mi of the current year DSM image Ga and the altitude value (z value) associated with the same mesh number mi of the previous year DSM image Gb is obtained. (S5: See FIG. 9).

  Then, it is determined whether or not the difference is equal to or greater than a preset altitude threshold Th (S7).

  When it is determined that the altitude threshold Th or more, as shown in FIG. 14, the mesh number mi, the altitude value (zi) of the previous year DSM image, the altitude value (zi) of the current year DSM image, and the determination result (“ 1 "or" 0 ") is stored in correspondence (S9).

  Next, it is determined whether the mesh number mi is the last (S11). If it is not the last, the mesh number mi is updated, and the process returns to step S3 (S13).

  By providing the above-described units, in the present embodiment, as shown in FIG. 15, the previous year DSM / ortho combined image BGa (FIG. 15 (c)) and the current year DSM / ortho combined image AGa (FIG. 15 (d)). )) Are displayed next to each other on the screen 15 at the same time. Since the previous year's DSM / ortho composite image BGa (FIG. 15 (c)) and the current year's DSM / ortho composite image AGa (FIG. 15 (d)) overlap with each other, the DSM image data and the precise orthophoto image data overlap. The user appears to have emphasized the portion changed by the gray scale of the DSM image data on the precision orthophoto image data, and it becomes easy to find the changed portion. As a result, it is possible to accurately determine the change in the building between the previous year and the current year.

  Here, the previous year's DSM / ortho composite image BGa (FIG. 15 (c)) and the current year's DSM / ortho composite image AGa (FIG. 15 (d)) are simultaneously displayed on the screen 15. Alternatively, the display may be switched for several seconds. This switching display makes it easier for the user to find the changed portion because the changed portion is easily noticeable.

  The previous year's DSM / ortho composite image BGa (FIG. 15 (c)) and the current year's DSM / ortho composite image AGa (FIG. 15 (d)) are based on the elevation difference change presence / absence table or the ortho image change detection table. The change area K is surrounded. Therefore, the user can concentrate on the enclosed change area K and compare the image in FIG. 15C with the image in FIG. Can be distinguished well. At this time, the previous year's DSM / ortho composite image BGa (FIG. 15 (c)) and the current year's DSM / ortho composite image AGa (FIG. 15 (d)) are switched and displayed at the same screen position several seconds at a time. Also good. By displaying in this way, the change area K is fixed, and only the DSM / ortho composite image appears to be switched between the previous year and the current year. Thereby, the changed part seems to be emphasized more, it becomes easy to find this changed part, and the change of the building of the previous year and this year can be discriminated accurately.

  Furthermore, as shown in FIG. 15, the previous year's precision orthophoto image (FIG. 15 (a)) and the previous year's DSM / ortho composite image BGa (FIG. 15 (c)) are displayed adjacent to each other at the same time. The current year precise orthophoto image (FIG. 15B) and the current year DSM / ortho composite image AGa (FIG. 15D) may be displayed adjacent to each other at the same time. In this case, the user compares the previous year's DSM / ortho composite image BGa (FIG. 15 (c)) with the current year's DSM / ortho composite image AGa (FIG. 15) while comparing the precision orthophoto image and DSM / ortho composite image of the same year. (D)) can be compared with each other, and it becomes easier to find the changed part.

  The previous year's precision orthophoto image (Fig. 15 (a)) and the current year's precise orthophoto image (Fig. 15 (b)) are generated by orthogonal transformation of a 3D model constructed using SfM. The previous year's DSM / ortho composite image BGa (FIG. 15 (c)) and the current year's DSM / ortho composite image AGa (FIG. 15 (d)) are respectively generated based on the precise orthophoto image of the current year. Therefore, the edge portion is firmly reproduced, and the visibility of the user is high.

  The embodiment described above exemplifies the configuration and processing procedure of an apparatus for embodying the technical idea of the present invention, and limits the arrangement and combination of components, the order of processing, and the like. It is not a thing.

  The technical idea of the present invention can be variously modified within the technical scope described in the claims. It should be noted that the drawings are schematic and the configuration of the apparatus is different from the actual one.

DESCRIPTION OF SYMBOLS 100 ... Feature change discrimination | determination apparatus 200 ... This year's DSM and precision ortho image creation part 210,220,240,270,280,310,320,340,370,380,410,510,610,710 ... Memory 230 ... Now 3D modeling processing unit for fiscal year 250 ... Precision ortho generation processing unit for current fiscal year 260 ... DSM creation processing unit for current fiscal year 300 ... DSM and precise ortho image creation unit for previous fiscal year 330 ... 3D modeling processing unit for previous fiscal year 350 ... for previous fiscal year Precision ortho generation processing unit 360 ... Previous year DSM creation processing unit 400 ... Current year DSM / precision ortho synthesis unit 500 ... Previous year DSM / precision ortho synthesis unit 600 ... Ortho image change extraction unit 700 ... Elevation change extraction unit 800 ... Change Area determination unit 850 ... Year-specific display composition unit 900 ... Mode determination unit 1000 ... Display control unit

Claims (15)

A first memory storing previous year's aerial photograph data taken by the aircraft over the region and previous year's 3D model previous year's DSM image data generated based on the reference point;
A second memory that stores orthophoto image data obtained by orthogonal transformation of the 3D model of the previous year as precise orthophoto image data of the previous year;
A third data storing the current year's 3D model current year's DSM image data generated based on the current year's aerial image data and the reference point taken in the same conditions as when the previous year's aerial image data was obtained. Memory and
A feature change determination method performed by a computer having a fourth memory storing orthophoto image data obtained by orthogonally converting the 3D model of the current year as orthophoto image data of the current year,
The computer is
Reading the previous year's precision orthophoto image data and the previous year's DSM image data, and generating these data for the previous year's DSM / precise orthosynthesis image data;
Reading out the current year's precision orthophoto image data and the current year's DSM image data, and generating the current year's DSM / precision orthocomposite image data;
A feature change discrimination method comprising: displaying the previous year's DSM / precision ortho-synthesized image data and the current year's DSM / precision ortho-synthesized image data in a region where each can be compared on the screen. The computer reads the aerial photograph data of the previous year, and estimates the photographing position and orientation of the camera based on corresponding points and reference points of pixels photographing the same point in two color photograph data in the same region. Using the parallax, the three-dimensional coordinates of each pixel are obtained by the triangulation calculation formula, and further noise removal is performed, and the color information of the corresponding points is given to the mesh of the three-dimensional model based on these three-dimensional coordinates. And generating the previous year 3D model,
Read the aerial photograph data of the current year, find the camera's shooting position and orientation based on the corresponding points and reference points of the pixels shooting the same point in the two color photograph data of the same region, Using the parallax, the three-dimensional coordinates of each pixel are obtained by a triangulation calculation formula, noise is further removed, and the color information of the corresponding points is given to the mesh of the three-dimensional model based on these three-dimensional coordinates, 2. The feature change discrimination method according to claim 1, wherein the step of generating a 3D model for the current year is performed. An altitude difference change presence / absence table in which the previous year DSM image data and the current year DSM image data are compared in units of meshes, and an identifier that is changed when the altitude difference (z) is equal to or greater than an altitude threshold is associated with the mesh. Storing in a fifth memory;
An area in which the identifier indicating that there is a change in the elevation difference change presence / absence table is continuous in the previous year's DSM / precision ortho composite image data or the current year's DSM / precision ortho composite image data displayed on the screen is enclosed The feature change discrimination method according to claim 1 or 2, wherein the step of displaying is performed. An ortho image difference change presence / absence table that compares the previous year's precision orthophoto image data and the current year's precise orthophoto image data in units of meshes, and associates an identifier that is changed when the color difference is equal to or greater than the color difference threshold with the mesh. Storing in a sixth memory;
A region in which the orthoimage change presence / absence change table is continuously present in the previous year's DSM / precision ortho composite image data or the current year's DSM / precision ortho composite image data displayed on the screen is enclosed. The feature change discrimination method according to claim 1 or 2, wherein the step of displaying is performed.   5. The feature change determination method according to claim 1, wherein the region is a city. A first memory storing previous year's aerial photograph data taken by the aircraft over the region and previous year's 3D model previous year's DSM image data generated based on the reference point;
A second memory that stores orthophoto image data obtained by orthogonal transformation of the 3D model of the previous year as precise orthophoto image data of the previous year;
A third data storing the current year's 3D model current year's DSM image data generated based on the current year's aerial image data and the reference point taken in the same conditions as when the previous year's aerial image data was obtained. Memory and
A fourth memory storing orthophoto image data obtained by orthogonally transforming the 3D model of the current year as current orthophoto image data;
The previous year's precision orthophoto image data and the previous year's DSM image data are read out, and the previous year's DSM / precise orthosynthesis image data is generated from the previous year's DSM / precise orthosynthesis image data,
The current year's precision orthophoto image data and the current year's DSM image data are read out, and this year's DSM / precision orthocomposite image data is generated,
A display control unit for displaying the previous year's DSM / precision ortho-composited image data and the current year's DSM / precision ortho-composited image data in an area where each can be compared on the screen;
A feature change discrimination device characterized by comprising: Read the previous year's aerial photo data, and estimate the camera's shooting position and orientation based on the corresponding points and reference points of the pixels shooting the same point in the two color photo data of the same region, Using the parallax, the three-dimensional coordinates of each pixel are obtained by a triangulation calculation formula, noise is further removed, and the color information of the corresponding points is given to the mesh of the three-dimensional model based on these three-dimensional coordinates, A previous year 3D modeling section for generating a previous year 3D model;
Read the aerial photograph data of the current year, find the camera's shooting position and orientation based on the corresponding points and reference points of the pixels shooting the same point in the two color photograph data of the same region, Using the parallax, the three-dimensional coordinates of each pixel are obtained by a triangulation calculation formula, noise is further removed, and the color information of the corresponding points is given to the mesh of the three-dimensional model based on these three-dimensional coordinates, The feature change discrimination device according to claim 6, wherein a 3D modeling unit for generating a 3D model for the current year is performed. An altitude difference change presence / absence table in which the previous year DSM image data and the current year DSM image data are compared in units of meshes, and an identifier that is changed when the altitude difference (z) is equal to or greater than an altitude threshold is associated with the mesh. An elevation change extractor stored in a fifth memory;
An area in which the identifier indicating that there is a change in the elevation difference change presence / absence table is continuous in the previous year's DSM / precision ortho composite image data or the current year's DSM / precision ortho composite image data displayed on the screen is enclosed The feature change determination device according to claim 6 or 7, wherein the change region determination unit to display is performed. An ortho image difference change presence / absence table that compares the previous year's precision orthophoto image data and the current year's precise orthophoto image data in units of meshes, and associates an identifier that is changed when the color difference is equal to or greater than the color difference threshold with the mesh. An ortho-image change extracting unit that stores the image in a sixth memory;
A region in which the orthoimage change presence / absence change table is continuously present in the previous year's DSM / precision ortho composite image data or the current year's DSM / precision ortho composite image data displayed on the screen is enclosed. The feature change determination device according to claim 6 or 7, wherein the change region determination unit to display is performed.   The feature change determination device according to claim 6, wherein the area is a city. A first memory storing previous year's aerial photograph data taken by the aircraft over the region and previous year's 3D model previous year's DSM image data generated based on the reference point;
A second memory that stores orthophoto image data obtained by orthogonal transformation of the 3D model of the previous year as precise orthophoto image data of the previous year;
A third data storing the current year's 3D model current year's DSM image data generated based on the current year's aerial image data and the reference point taken in the same conditions as when the previous year's aerial image data was obtained. Memory and
A feature change determination program executed by a computer having a fourth memory storing orthophoto image data obtained by orthogonally converting the 3D model of the current year as orthophoto image data for the current year,
Reading the previous year's precision orthophoto image data and the previous year's DSM image data, and generating these data for the previous year's DSM / precise orthosynthesis image data;
Reading out the current year's precision orthophoto image data and the current year's DSM image data, and generating the current year's DSM / precision orthocomposite image data;
The computer program executes the step of displaying the previous year's DSM / precision ortho-synthesized image data and the current year's DSM / precision ortho-synthesized image data in a region where each can be compared on the screen. Read the previous year's aerial photo data, and estimate the camera's shooting position and orientation based on the corresponding points and reference points of the pixels shooting the same point in the two color photo data of the same region, Using the parallax, the three-dimensional coordinates of each pixel are obtained by a triangulation calculation formula, noise is further removed, and the color information of the corresponding points is given to the mesh of the three-dimensional model based on these three-dimensional coordinates, Generating a 3D model for the previous year;
Read the aerial photograph data of the current year, find the camera's shooting position and orientation based on the corresponding points and reference points of the pixels shooting the same point in the two color photograph data of the same region, Using the parallax, the three-dimensional coordinates of each pixel are obtained by a triangulation calculation formula, noise is further removed, and the color information of the corresponding points is given to the mesh of the three-dimensional model based on these three-dimensional coordinates, 12. The feature change determination program according to claim 11, wherein the computer executes a step of generating a 3D model for the current year. An altitude difference change presence / absence table in which the previous year DSM image data and the current year DSM image data are compared in units of meshes, and an identifier that is changed when the altitude difference (z) is equal to or greater than an altitude threshold is associated with the mesh. Storing in a fifth memory;
An area in which the identifier indicating that there is a change in the elevation difference change presence / absence table is continuous in the previous year's DSM / precision ortho composite image data or the current year's DSM / precision ortho composite image data displayed on the screen is enclosed The feature change determination program according to claim 11 or 12, wherein the computer executes the step of displaying. An ortho image difference change presence / absence table that compares the previous year's precision orthophoto image data and the current year's precise orthophoto image data in units of meshes, and associates an identifier that is changed when the color difference is equal to or greater than the color difference threshold with the mesh. Storing in a sixth memory;
A region in which the orthoimage change presence / absence change table is continuously present in the previous year's DSM / precision ortho composite image data or the current year's DSM / precision ortho composite image data displayed on the screen is enclosed. The feature change determination program according to claim 11 or 12, wherein the computer executes the step of displaying. 15. The feature change determination program according to claim 11, wherein the area is a city.