JP2002063580A - Inter-image expansion image matching method using indefinite shape window - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an inter-image expansion imaging matching method capable of generating a three-dimensional image of the same region in a relatively short period by using the types of images different from each other and updating the map of the same region. SOLUTION: A computer turns any combination set among the combination between a satellite image (a photograph or a line scanner image) and an aerial image (a photograph or a line scanner image), a combination of aerial images with each other and a combination of scanner images with each other among the aerial photographic images, satellite photographic images and line scanner images of the same region into digital formats, stores the set in a hard disk 2, etc., and performs tocating work for the satellite image and the aerial image. A protean window is defined to an expansion stereo model Di that determines a function for projecting terrestrial three-dimensional coordinates onto two-dimensional coordinates on each image with respect to the respective images, and elevation is given to the ground object of luminance images I1 and I2 be three- dimensional data by giving the height of the indefinite shape window.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION Converting two-dimensional numerical map data having no elevation into three-dimensional data by using two or more kinds of images, such as aerial photographs and satellite images, taken of the same area (adding elevation values) ) Or a method for confirming the loss of a feature described in existing digital map data.

[0002]

2. Description of the Related Art In conventional photogrammetry, a measurement object is photographed from two different positions (or more) using the same kind of camera, a corresponding point is obtained on each image, and based on the principle of triangulation. An operation of determining a position in the three-dimensional space is performed.

[0003] Methods for automatically performing this work include:
Attention is paid to one point in one image constituting a stereo image or a region or a line structure (edge) partitioned on a regular grid, and a corresponding point, region, or line structure is represented in the other image. In general, a stereo matching method of searching and matching with is used.

[0004]

However, in this method, even if there is an area whose shape is known in advance, for example, since the photographing altitude and photographing date are different, the shape can be used as a search condition. Can not.

[0005] For this reason, matching using this shape as a constraint condition cannot be performed, so that the elevation cannot be determined efficiently, and as a result, three-dimensional data of the feature cannot be obtained.

[0006] This is because, when an elevation is given to a feature, for example, a building, etc. described in the past digital map data to make it into three-dimensional data, points or lines measured by a stereo image are re-examined. Unless the data is edited, three-dimensional numerical map data cannot be generated.

Recently, high-resolution commercial satellites have been put into practical use, and attempts have been made to acquire digital image data by mounting a line scanner on an aircraft. The situation where individual data is obtained is beginning to spread.

That is, unlike conventional photogrammetry, images taken from different positions in the same area without intending to take stereo images in advance have been individually acquired.

[0009] In the information society toward the 21st century,
With the spread of geographic information systems (GIS), it is necessary to update and maintain digital map data, which is the information base, in a timely manner.

However, even if an attempt is made to obtain the latest digital map in a timely manner, it is difficult to update the data in a timely manner by taking stereo aerial photographs once every several years as in the prior art, and the cost cannot be reduced. .

Therefore, it becomes useful to perform measurement equivalent to conventional photogrammetry or update a map by using various kinds of images which are imaged and individually acquired in a timely manner in the same area. However, there is no such system in reality.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and easily converts digital map data having no elevation into three-dimensional data using two or more kinds of various images photographing the same area. Accordingly, it is an object of the present invention to obtain an inter-image extended image matching method capable of updating an existing map.

[0013]

SUMMARY OF THE INVENTION The present invention relates to an imaging matching method using a combination of two types of images having different resolutions in the same area. Generating an extended stereo model, and a step of setting a polygonal point sequence of a known feature that is present in common to both images of the extended stereo model as a reference irregular-shaped window; (Z) to change the height (z), and projecting the extended stereo model on both images according to the height each time the height of the irregular window is changed Determining a position; and comparing the luminance values of the regions delimited by the irregular window at the projection positions of both images of the extended stereo model; and comparing the irregular window when the comparison result is the closest. Outputting the height (z) as the height of the feature in both images of the extended stereo model; and determining the three-dimensional coordinates of the feature using the output height (z). The gist is that

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS <First Embodiment> First Embodiment
Then, as shown in Fig. 1, aircraft (including helicopters)
And any combination of aerial image, satellite image, and line scanner image of the same area acquired by satellite (combination of satellite image (photo or line scanner image) and aerial image (photo or line scanner) or different altitude) Including a combination of aerial photographs of each other or a combination of scanner images of different altitudes) is digitally formatted by a computer and stored in a hard disk or the like, and the two resolutions are matched (the photographing altitude,
Based on the scale), the orientation work) is performed on these images.

Then, a function for projecting three-dimensional coordinates on the ground to two-dimensional coordinates on each image is determined for each image. In this way, the combination of images in which two images are related by the function of projecting the three-dimensional coordinates of the ground onto each image coordinate is converted into an extended stereo model Di (luminance image I
1, luminance image I2).

By using such an extended stereo model Di, for example, it is possible to give an altitude to a two-dimensionally described feature on an existing digital map without having an altitude value, and to convert the feature into three-dimensional data.

That is, a feature Bu (known) having an arbitrary shape whose altitude is to be obtained is defined as a reference amorphous window Pl (polygon row) of a closed area, and has an arbitrary altitude (along the height). Determine.

Then, while changing the height of the irregular window Pl, the positions on the luminance images I1 and I2 are obtained for the luminance image I1 and the luminance image I2 by using the above-described function, and the irregular window Pl at this position is obtained. Luminance images I1 and I2 when projected
Are compared, and the height (elevation value H) of the irregular P1 when it is determined to be the most matching (approximate) by the evaluation function is determined as the height of the feature in the luminance image. FIG. 1 shows an example in which the altitude value is the highest when the altitude is H3.

In FIG. 1, in order to facilitate understanding, the heights of the irregular windows Pl are changed so that the luminance images I1 and I1 are changed.
2, the shape of the irregularly shaped window Pl when projected on FIG.

<Embodiment 2> In addition, in a map update, a new building or the like which is not included in the map may be added as three-dimensional data. In this case, in the conventional method, the operator must make three-dimensional data while observing in stereo, but the operator cannot perform stereo observation with a combination of different types of images, and there is no such system.

However, when the extended stereo model Di is used, since the luminance image I1 and the luminance image I2 are related by a function in the extended stereo model Di, as shown in FIG. Image I
1 is displayed on the screen, and the feature B
A closed area B1 corresponding to the contour of u is given. Then, an irregular window Plh is generated at a position where the projection coincides with the closed area, and the altitude H (z) is changed.

For example, when the altitude is changed in the order of 30 m, 80 m, and 150 m, the projection positions of the luminance images I 1 and I 2 of the irregular-shaped window Plh in the luminance images I 1 and I 2 when the altitude changes to 30 m, 80 m and 150 m are obtained. While
The luminance values in the luminance images I1 and I2 when the irregular window Plh is projected at this position are compared, and the height of the irregular P1h (elevation value H ) Is determined as the height of the feature in the luminance image. Therefore, the three-dimensional coordinates of the feature Bu can be estimated.

In FIG. 2, in order to facilitate understanding, the height of the irregular window Plh is changed to change the luminance image I1,
The shape of the irregular window Plh when projected on I2 is shown in a different manner.

That is, in the first and second embodiments, as shown in FIG. 3, the luminance image I of the extended stereo model in which the images of the aforementioned combination are related by a function.
In I and I2, the projection positions in the luminance images I1 and I2 are obtained by changing the height of the irregular windows Plh obtained from the common feature Bu, and are separated by the irregular windows at the projection positions. The height of the feature Bu is estimated by matching (approximately) the evaluation functions of the luminance values.

<Embodiment 3> When a digital terrain model can be used, the specific height of a feature is also obtained by subtracting the elevation at the feature position from the feature elevation z, as shown in FIG. . As a result, an accurate building height can be obtained.

Further, it is assumed that attention is paid to, for example, a building described in a past digital map. If this specific height is determined to be a value close to 0, it can be estimated that the building has been lost. Even for a building to which three-dimensional coordinates have already been given, the loss or change of the feature can be estimated by comparing the image at the position where the feature is projected between the newly captured image and the old image. .

Next, specific examples of the first, second, and third embodiments will be described below. FIG. 5 is a configuration diagram of the method for estimating the elevation and specific height of a feature according to the present invention. FIG. 6 is a schematic configuration diagram of the altitude search unit of FIG.

For example, the digital image input unit 1 stores two images in the form of digital images on a hard disk 2 of a computer as an image processing apparatus. At this time, in the case of an image supplied as a film or photographic paper, such as an aerial photograph, the image is read by a scanner and converted into a digital image by an A / D converter. The image luminance at the pixel coordinates (x _1pix , y _1pix ) and (x _2pix , y _2pix ) of the digital image of the two converted pictures is represented by I
₁ (x _1pix , y _1pix ) and I
₂ (x _2pix , y _2pix ).

The hard disk 2 includes an altitude H above the ground when the extended stereo model is photographed, a photographing position interval between the images (base line length B), and a ground resolution average D corresponding to one pixel of the two images. , Known polygons (irregular shaped windows) and the like are stored.

Next, the orientation element calculating section 3 calculates a coordinate conversion function FX and an inverse coordinate conversion function for calculating the image coordinates of each image on which three-dimensional coordinates on the ground are projected for the two luminance images I1 and I2. GX (collectively referred to as orientation information) is determined. This is what is called orientation work in aerial photogrammetry. The orientation method and parameters to be determined differ depending on the image sensor (for example, an area sensor and a linear array sensor), but finally, the pixel coordinates of the left and right images on which the point (x, y, z) in the three-dimensional space is projected. (X _1pix , y _1pix ) and (x _2pix , y
_2pix ) can be determined.

That is, the following four coordinate conversion functions FX1
(X, y, z), FY1 (x, y, z), FX2 (x,
y, z) and FY2 (x, y, z) can be determined analytically or procedurally.

X _1pix = FX1 (x, y, z) Formula (1) y _1pix = FY1 (x, y, z) Formula (2) x _2pix = FX2 (x, y, z) Formula (3) y _2pix = FY2 (x, y, z) Equation (4) Also, if Equations (1) and (2) are solved analytically or procedurally with z = constant, the point on each image is at an altitude z. The x coordinate and the y coordinate on the ground at the time of the assumption can be obtained.

X = GX1 ( _x1pix , _y1pix , z) Equation (5) y = GY1 ( _x1pix , _y1pix , z) Equation (6) Similarly, solving equations (3) and (4) gives x = GX2 ( _x2pix , _y2pix , z) Formula (7) y = GY2 ( _x2pix , _y2pix , z) Formula (8).

For the case of an area sensor image such as an aerial photograph or an area CCD sensor, the function FX1 (x, y,
z), FY1 (x, y, z), FX2 (x, y, z),
The details of FY2 (x, y, z) are described in FIG.

It should be noted that the orientation method is to use an existing method. For example, an area CCD sensor is described in detail in reference [1], and a linear center projection sensor is described in detail in reference [2].

[1] Japan Society of Photogrammetry: Analytical Photogrammetry revised edition, 1989. "2" Osamu Uchida: Research on automation of three-dimensional measurement using stereo satellite images, 1989.

In FIG. 7, for example, the coordinate conversion function of the area sensor image is such that the coordinates of the projection center of the center projection image are P ₀ = ^t [X ₀ Y ₀ Z ₀ ], and the rotation angle of the camera with respect to the ground coordinates is T = ^{Let t} [ω φ κ] and the sum of these (external orientation elements) be E = ^t [tT tP ₀ ] (t on the left shoulder indicates a transposed matrix or transposed vector).
At this time, ground coordinates P = t ^[x y z] is the camera coordinate system (relative three-dimensional coordinate system to the projection center) with p _r is expressed as follows.

P _r = ^t [x _{ry r} z _r ] = R (T) (R−P ₀ ) Equation (22) where R (T) is a rotation matrix defined by the following equation: .

[0039]

(Equation 1) Also, the photographic coordinates at which P is projected (image of the projection center (principal point)
Two-dimensional coordinates with the origin as) p (P, E) = ^t [xy]
Is represented by the following equation.

[0040]

(Equation 2) Here, c is the focal length of the aerial photograph. Equation (24) is
If the external orientation elements of the camera and the three-dimensional coordinates of the ground point are known, it means that the corresponding photograph coordinates can be known.
The photographic coordinates p are physically the actual coordinates on the film surface or on the CCD element. However, if the distortion and non-linearity of the scanner when reading the film and the arrangement error of the CCD element can be ignored, the image coordinates pimg and Can be expressed by a two-dimensional projective transformation Hin. As described above, if E and Hin can be known, it is possible to convert the three-dimensional coordinates of the ground point into image coordinates, and the four coordinate conversion functions FX1
(X, y, z), FY1 (x, y, z), FX2 (x,
y, z) and FY2 (x, y, z) are obtained.

It is assumed that the photograph coordinates p = (x, y).
Then this is p in camera coordinates_rp = (X, y, -c)
It is. As a result, the point in p = (x, y)
Altitude Z_pIf there is, this point in the ground coordinate system
3D coordinates P_p= (X, Y, Z_p) Is given by
You.

[0042]

(Equation 3) After inverse projective transformation Hin ^-1 to the image coordinates PIMG, by applying the equation (25), it is possible to convert from the image coordinates to three-dimensional coordinates in the original constraint of Z = Z _p. That is,
Coordinate transformation functions GX1 (x, y, z), GY1 (x, y,
z), GX2 (x, y, z) and GY2 (x, y, z) can be obtained.

On the other hand, the digital map input section 4 inputs an existing digital map. This digital map is assumed to hold features (for example, building data) in a vector format (coordinate point sequence data). If the digital map is image data, it is converted into a vector format by raster / vector conversion.

Next, the processing target feature data extraction unit 5 extracts the feature data to be subjected to the elevation from the numerical map data in accordance with the instruction from the operator. In general, differences in the type of object (building, road, etc.) are coded in a numerical map, so that only the figure of the target code needs to be extracted at the time of extraction.

[0045] Then, convert the extracted feature Bu polygon point sequences Pl _i (closed polygon, typically the first and last columns point where the same point (p1, p2, ..., represented by pn, p1)) to I do. At this time, if the target feature Bu is building data, there is no problem since many of the objects are usually a sequence of polygon points. If the feature is not a polygon sequence,
The data is edited by CAD software or the like and converted into a polygonal point sequence in advance.

The on-image feature measuring unit 6 determines that a feature such as a building exists on the image of the extended stereo model Di, but that the target feature shape is not described in the digital map data. , Extended stereo image Di (luminance images I1, I
2) is displayed on the screen, and then the shape of the feature is input by operating the mouse. In this case, feature shape is a polygon point sequences Pl _i, inputted as closed graphic listing the image coordinates shaped (irregular window). For example, in FIG. 2, B1 is selected from the luminance image I1, and an irregular window of B1 is obtained.

Next, the altitude search unit 7 searches for altitude
Each feature polygon point sequence Pl obtained by the image feature measurement unit 6
_The following processing is repeated for _i (i = 1,..., n _i ).

The altitude search range and search interval setting unit 7a in the altitude search unit 7 provide an altitude search range (Hmin, Hma
x) and the search interval dH are set.

For Hmin, the altitude of the ground surface of the area or a value smaller than the altitude is given, and for Hmax, the value of the maximum altitude of the feature estimated from the number of floors or the like of the map or the like is given.

Topographic model (Numerical map 5 issued by the Geospatial Information Authority of Japan)
When a 0 m mesh (altitude) can be used, the altitude of the ground surface at each feature position can be estimated.

When it cannot be used, the same Hmin and Hmax may be appropriately set for all the features. Also, dz may be given the resolution in the height direction measured from the extended stereo model.

For example, if the photographing position interval (base line length) between the images is B, the average of the ground height at the time of photographing the extended stereo model is H, and the ground resolution average corresponding to one pixel of the two images is D, dH is given by the following equation.

[0053]

(Equation 4) At this time, the search altitude group H _k is defined as follows.

H _k = Hmin + dH × k (k = 0, 1, 2,... N _k ) Equation (10) where _nk is the maximum _{k such} that H _k ≦ Hmax.

The irregular window defining section 7b defines the inside of each polygon point sequence of the feature data as an irregular window for stereo matching. Specifically, as shown in FIG. 8, equidistant mesh feature polygon point in the column (dx, dy) is generated, which is referred to as irregular window Pl _i.

[0056] This process can either polygon point sequence Pl _i (coordinates terrestrial coordinates) digital map obtained from, or processed by any measurement result on the image (coordinates image coordinates) obtained from different.

When the geographical feature data of the polygon point sequence obtained from the digital map originates from the digital map, as shown in FIG.
An equally spaced mesh point group MPi is generated around each polygon point sequence of the feature data Bu. The mesh sets the origin (ox, oy) at arbitrary ground coordinates, generates it at intervals of dx and dy in the x and y directions, and generates M (o
x, oy, dx, dy).

Note that ox and oy may be arbitrarily set. Dx and dy are one of digital aerial photographs.
Set the ground resolution from pixels to several pixels.

[0059] defined where i-th feature F _i present in the interior region Plin _i surrounded by the polygon point sequence Pl _i representing the intersection points MP _i of mesh as follows.

MPi = Plin _i ∩M (ox, oy, dx, dy) Equation (11) Using the irregular window Pli obtained in this way, for example, in FIG. 1, the height of the irregular window Pl is changed. The brightness image I1,
The projection position of the luminance image I2 is obtained, and the luminance values of both images at this projection position are compared.

Then, the altitude H at which the evaluation function of the luminance value is most matched (approximate) is determined as the altitude z of the features B1 and B2.

[0062] When the polygon point sequence obtained from the on-image measurement results are irregular window definition section 7c (also referred to as a second amorphous window definition section), when Pl _i is the measurement results on the image Creates a mesh in accordance with the altitude H _k determined in the evaluation function calculation loop unit 7d.

[0063] Now, assume that Pl _i was measured on the image 1, the i-th image coordinates _{(px j,} py j) When,
Ground coordinates _(X j when this point was in altitude _{H k,}
Y _j ) is calculated by using Expressions (5) and (6) as follows: X _j = GX1 (px _j , py _j , H _k ) Expression (12) Y _j = GY1 (px _j , py _j , H _k ) It is represented by equation (13). After returning all points on the polygon point sequence to ground coordinates in this way, they are converted in the same way as for digital maps,
A mesh intersection group MP _ik is obtained. Pl _i get similarly mesh point group _{MP ik} by even expression when it is measured on the image (12) and (13).

By using such an irregular shaped window, for example, as shown in FIG. 2, a brightness image I1 obtained by an aircraft is displayed on a screen, and a feature B1 whose coordinates are known by mouse operation or the like on this image. Is given (closed window Plh: polygon). Then, the altitude H (z) of the irregular shaped window Plh is changed in the feature Bu.

The above-described evaluation function calculation loop unit 7d calculates the evaluation value E (H _k ) for all H _k (k = 0, 1, 2,..., N _k ).

[0066]

(Equation 5) Here, Ekj is an evaluation function value when it is assumed that the j-th point (xj, yj) of the mesh intersection group MPi is at the height of the altitude Hk. The evaluation value E (Hk) evaluates the degree of coincidence of the pixel values of the images on the extended stereo model, and decreases as the coincidence increases.

That is, the j-th point (xj, yj, Hk) of the group of mesh intersections MPi at the height of the altitude Hk is placed on the two images I1 and I2 by the equations (1) to (4), respectively. When projected, the pixel values V1kj and V at that position
2kj the _{V1 kj (x j, y j} ) = I 1 (FX1 (x j, y j, H k), FY1 (x j, y j, H k)) V2 kj (x j, y j) = I _{2 (FX2 (x j, y} j, H k), FY2 (x j, y j, H k)) formula (15) _{wherein, V1 kj (x j, y} j), V2 kj (x j, y
_j ) is a pixel value when projected, and (x _j , y _j , H
_k ) indicates a polygon point group of the irregular window when it is at H.

If expressed as in the above equation (15), the sum of the squares Edif of the pixel value differences and the correlation coefficient of the pixel values are (-1).
, Etc. can be used.

For example, Edif _k is represented by the following equation.

[0070]

(Equation 6) Further, Ecor _k is expressed by the following equation.

[0071]

(Equation 7)Then, the feature elevation determination unit 7e calculates the i-th feature F_iMark
High H_FiWith the evaluation function E (H_kHeight H that minimizes
_kminAnd That is, H_Fi= H_kmin  E (H_kmin) = Min (E (H_k)) (K = 0, 1, 2,... N_k) Equation (18) Next, the feature height calculation unit 7f determines whether or not there is a terrain model.
Is a two-dimensional polygon point sequence Pl_iThe height of the ground surface
Altitude H_FiBy subtracting from_{R i}age
Output. In addition, for later use in reliability analysis,
High price H_FiEvaluation function value E at_iAlso as attribute information of features
Output.

Then, the three-dimensional feature data output unit 7e
Adding altitude values _{H Fi} to z coordinates of the two-dimensional polygon point sequence Pl _i, further outputs the ground surface ratio height _{H Ri} as the three-dimensional feature data Pl3D _i was added as attribute data.

Next, the three-dimensional map output unit 8 outputs all P
Together with 13D _i , it is output as a three-dimensional map with elevation.

Next, an explanation will be supplemented for the evaluation of the reliability of the altitude estimation.

In the following case, the following example can be considered as a method for estimating the reliability of the altitude estimation.

When the evaluation function value _Ei is large. That is, a certain threshold value as T2, as the method of determining the value of _{E i} ≧ T2 Equation (19) T2, than the average value of the evaluation function value _{E i,} is it reasonable to set the 1~2σ large value is there.

When the obtained height H _Fi is at both ends of the search range. _{That, H Fi = Hmin or H Fi} = Hmax formula (20) in this case, there is a possibility that the search range is not set correctly.

In the case of estimating the loss or change of a feature, it can be estimated that the possibility of loss or change of the feature is high in the following cases.

When the ground surface specific height is small. That is, assuming that a threshold value close to 0 is T1, | H _Ri | ≦ T1 Expression (21) In addition, as a cause of low reliability, there is a case where a change occurs in the height, size, or the like of a feature. .

For example, as shown in FIG. 9, even when the elevation of a feature is Horg at the time of photographing in the year A, if the image is photographed a few years later (aging), the elevation changes from Horg to Hn.
ew in some cases.

In some cases, the size (planar shape) of the building may change over time (photographed in year C).

In the case of estimating a change of a feature having a known altitude,
The movement of a feature whose altitude is already known is estimated as follows.

The elevation of the feature is set to Horg, and the search range of the elevation is set to Horg−dH ≦ H _k ≦ Horg + dH.

If there is an aging, the estimated altitude is H
It is significantly different from org, or the reliability of the altitude estimation is low from the evaluation function value.

It is also possible to construct images of different old and new photographing timings in an extended stereo model, estimate the altitude, and determine that there is an aging when the reliability of the altitude estimation becomes low.

Further, for example, as shown in FIG. 10, it is possible to cope with a case where a feature has a complicated shape or an actual height is not constant. Experiments have shown that when compared to the results of actual aerial photogrammetry and laser scanner measurements on buildings, they show values close to the mode elevation of the buildings.

When the feature height is close to 0, it can be automatically extracted as a candidate for a lost portion of the feature to improve the efficiency of correction.

<Embodiment 4> Next, updating of a map will be described with reference to the flowchart of FIG. In this description, it is assumed that a three-dimensional map (old), a two-dimensional map (old), and an extended stereo model for new shooting are stored in the disk in advance.

In this state, the two-dimensional map (old) is read, and the extended stereo model is read. As described above, the height is given to the irregular shaped window of the known feature, and the height is changed. The projection positions on both images of the extended stereo model are obtained by using the above method, the luminance values of the areas delimited by the irregular windows at the projection positions are calculated, and the luminance values are compared. Elevation processing is performed by extended stereo matching using the height of the window as the height of a feature in the extended stereo model (S1)
This three-dimensional map data is stored as an intermediate result (S
2).

Then, a change extraction process is performed by extended stereo matching for obtaining a change from the three-dimensional map (old), the three-dimensional map data of the intermediate result, and the new extended stereo model (S3), and this is extracted on the photograph ( It is stored as a three-dimensional map of the intermediate result before being assigned to the extended stereo model (S4).

Next, editing processing on the photograph is performed (S
5). This is preferably done using a digital photogrammetry workstation.

In this editing process, the three-dimensional data of the intermediate feature obtained in step S4 is overlaid on a new extended stereo model and displayed on a monitor, and the three-dimensional data and the newly photographed extended stereo model are displayed. The operator manually edits by comparing with. Here, a general stereo image newly shot may be used as the extended stereo model, but a stereo model composed of an image obtained when an old two-dimensional map is created and a newly shot image is used. Is also good.

Then, a process of updating the old three-dimensional data to the three-dimensional data obtained by this editing process is performed (S
6).

That is, by this series of processing, digital map data having no elevation can be easily converted into three-dimensional data, and an existing map can be updated.

[0095]

As described above, according to the present invention, a two-dimensional map can be efficiently converted into a three-dimensional map. In particular, when it is not necessary to take a new aerial photograph, a city three-dimensional model used for applications such as computer graphics and city three-dimensional simulation can be created at low cost from an existing map.

Further, by making the numerical map three-dimensional, a new photographed map can be superimposed on the stereo aerial photograph in the secular change correction work of the two-dimensional map.
Modification becomes easy.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of an extended imaging method using an irregular window according to a first embodiment.

FIG. 2 is a schematic configuration diagram of an extended imaging method using an irregular window according to a second embodiment.

FIG. 3 is an explanatory diagram illustrating a change in the height direction of an irregular shaped window.

FIG. 4 is an explanatory diagram illustrating a specific height according to a third embodiment.

FIG. 5 is a specific configuration diagram of each embodiment.

FIG. 6 is a configuration diagram of an elevation search unit.

FIG. 7 is an explanatory diagram illustrating geometric characteristics of an area sensor.

FIG. 8 is an explanatory diagram of meshing of feature data.

FIG. 9 is an explanatory diagram illustrating a change in feature data at different shooting times.

FIG. 10 is an explanatory diagram illustrating a method of obtaining an elevation of a building having a complicated shape.

FIG. 11 is a flowchart illustrating updating of a map according to the third embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Digital image input part 2 Hard disk 3 Orientation element calculation part 4 Numerical map input part 5 Processing object feature data extraction part 6 Image feature measurement part 7 Elevation search part

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G09B 29/00 G09B 29/00 Z (72) Inventor Osamu Uchida 4-2-18 Shinjuku, Shinjuku-ku, Shinjuku-ku, Tokyo Shinjuku Kofu Building Asia Air Survey Co., Ltd. (72) Mitsuteru Sakamoto Inventor 4-2-18 Shinjuku, Shinjuku-ku, Tokyo Shinjuku Kofu Building Asia Air Survey Co., Ltd. (72) Ryosuke Shibasaki 1-5-Sarie, Koto-ku, Tokyo 3-408 F term (reference) 2C032 HA02 5B050 BA02 BA07 BA17 DA02 DA04 DA05 DA10 EA07 EA18 EA19 EA27 EA28 FA13 5B057 AA13 AA14 CA08 CA12 CA13 CA16 CA18 CB08 CB13 CB18 CD01 CD14 CE11 5L096 AA41 FA18 FA09 GA04

Claims (3)

[Claims] 1. An imaging matching method using a combination of two types of images having different resolutions in the same area, wherein an orientation process is performed on the two types of images to generate an extended stereo model related by a function. And a step of using a polygon point sequence of a known feature which is present in both images of the extended stereo model in common as a reference irregular window; and giving a height (z) to each of the irregular windows, Changing a height (z); obtaining a projection position on both images of the extended stereo model according to the height each time the height of the irregular window is changed; Comparing the luminance values of the areas delimited by the irregular window at the projection positions of the two images of the stereo model; and calculating the height (z) of the irregular window when the comparison result is the closest to the extended window. An irregular window having a step of outputting as a feature height of both images of the teleo model, and a step of determining three-dimensional coordinates of the feature using the outputted height (z). Extended Image Matching Method between Images 2. The method according to claim 1, further comprising the step of: providing geographical data of the surface of a region, and subtracting the height from the geographical data to obtain a specific height of the feature every time the height of the feature is determined. 2. The extended imaging matching method between images using an irregular window according to claim 1. 3. The method according to claim 1, further comprising the step of notifying that the feature has disappeared when an image based on the irregular shaped window does not exist in a new image. Extended Image-to-Image Imaging Matching Method Using Undefined Windows