JP2010224735A - Three-dimensional space data generating method and three-dimensional space data generating device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for generating three-dimensional space data for at least one image by utilizing space data without needing an image captured from multi-point of view. <P>SOLUTION: A three-dimensional space data generating system has a two-dimensional space data DB 101, a three-dimensional space data DB 102, a host system 104, and a terminal 103. They are configured to be connectable through a LAN, for instance. The two-dimensional space data DB 101 and the three-dimensional space data DB 102 are connected to the host system 104. A user utilizes an input part 107, a display part 108, or the like of a terminal 103 to transfer camera image data, edit data, or the like to the host system 104 and to store the three-dimensional space data. The host system 104 has a retrieval processing part 105 and a three-dimensional space data generating part 106. Space coordinates of the two-dimensional space data are associated with the camera image, and the height is measured to generate three-dimensional space data. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

The present invention relates to three-dimensional spatial data creation technology, and more particularly to a technology for creating three-dimensional spatial data from two-dimensional spatial data. The present invention relates to a technique for creating three-dimensional spatial data using a photographed image and two-dimensional spatial data in which the object exists.
Gyro sensor

  Currently, the digitization of maps is progressing. This makes it possible to have a lot of information on the map, and the usefulness of the map is expected to increase. In addition, a service is provided in which a 3D model or a 360 ° viewpoint image is placed on a map, or a 3D model is created and placed on a map. These techniques include techniques for detecting the height of an object or the like based on images taken from multiple viewpoints and creating a three-dimensional model, or creating a three-dimensional model using CAD. Three-dimensional spatial data can be created by the method.

  As conventional three-dimensional spatial data creation methods, for example, techniques described in Patent Documents 1 to 3 below are known.

JP 2003-317081 A JP 2002-8012 A Japanese Patent Laid-Open No. 2002-189412

  However, currently, when creating 3D spatial data, for example, a Cessna machine that maintains a certain altitude uses a number of photographs that have taken overlapping positions and have taken overlapping images. The altitude is calculated and a method of creating three-dimensional spatial data is used. In addition to this, when creating three-dimensional space data, there is a problem that it takes time since an image obtained by photographing one object from multiple viewpoints is required.

  An object of the present invention is to provide a technique that does not require images taken from multiple viewpoints and that can use spatial data to create three-dimensional spatial data if there is at least one image.

According to one aspect of the present invention, there is provided a three-dimensional spatial data creation method for creating three-dimensional spatial data from two-dimensional spatial data by computer processing based on a photographed image of an object associated with a photographing direction and a photographing position. A step of retrieving the two-dimensional space data of the object from a two-dimensional space database storing two-dimensional space data based on the photographing direction and the photographing position; and the retrieved two-dimensional space data and the photographed image And associating the vertex of the two-dimensional space data of the object with the vertex of the captured image;
Obtaining a height of the object using the photographing direction and the photographing position under a condition that the center of the photographed image is the center of the photographing direction. A data creation method is provided. When creating three-dimensional spatial data, not only an image taken with a camera equipped with a gyro sensor or GPS but also two-dimensional spatial data corresponding to the image is used. Thus, the captured camera image is associated with spatial coordinates from the two-dimensional spatial data, and the height is measured from the spatial coordinates.

  The step of obtaining the height of the object sets a center of the captured image (center of the capturing direction), sets an intersection of a perpendicular drawn from the center and an outline of the object, and the intersection and the The distance to the center is the sum of the first distance between the photographing position and the intersection, the second distance obtained from the photographing angle obtained from the photographing position and the photographing direction, and the height of the photographing position, and Obtaining a height of the object from a first height which is a distance between the intersection and the center and a remaining second height of the object; And calculating and adding the height of 2 from the ratio of the first height to the pixel value of the captured image. Furthermore, it is preferable to include a step of obtaining the coordinates of the three-dimensional space data at a position not visible in the captured image by interpolation based on the two-dimensional space data.

  The present invention may be a program for causing a computer to execute the above-described three-dimensional spatial data creation method, or a computer-readable recording medium for recording the program.

  According to another aspect of the present invention, there is provided a three-dimensional spatial data creation device that creates three-dimensional spatial data from two-dimensional spatial data by computer processing based on a photographed image of an object associated with a photographing direction and a photographing position. A search processing unit for searching the two-dimensional space data of the object from a two-dimensional space database storing two-dimensional space data based on the shooting direction and the shooting position; and the searched two-dimensional space data; In contrast to the captured image, the vertex of the two-dimensional space data of the object is associated with the vertex of the captured image, and the center of the captured image is the center of the capturing direction. And a three-dimensional spatial data creation unit that obtains the height of the object using the photographing direction and the photographing position. It is.

  According to the present invention, there is an advantage that three-dimensional space data can be easily created using a camera image. In addition, a plurality of camera images are not required, and a portion that cannot be recognized by the camera images can be interpolated using two-dimensional spatial data.

It is a figure which shows the system structural example of the three-dimensional spatial data creation apparatus by one embodiment of this invention. It is a flowchart figure which shows the flow of the three-dimensional space data creation process by this Embodiment. It is a figure which shows an example of the camera image by this Embodiment. It is a figure which shows the two-dimensional space data corresponding to the camera image shown in FIG. It is a figure which shows the example by which the searched two-dimensional space data and a camera image are displayed as a pair. It is a figure which shows the mode of the height measurement of a camera image. It is a figure which shows the relationship between a camera image and two-dimensional space data. It is a figure which shows the figure for calculating | requiring height D3 to the center of a camera image. It is a figure which shows the figure for calculating | requiring the height of a building. It is a figure which shows a mode that 3D space data is calculated | required by interpolation from 2D space data in consideration of an impossible viewpoint. It is the completion figure which gave three-dimensional space data to the building.

  Hereinafter, a three-dimensional spatial data creation technique according to an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, a building is described as an example of the creation target of the three-dimensional spatial data, but the target is not limited to the building.

  FIG. 1 is a diagram showing an example of the system structure of a three-dimensional spatial data creation device according to an embodiment of the present invention. As shown in FIG. 1, the three-dimensional spatial data creation system according to the present embodiment includes a two-dimensional spatial data database 101, a three-dimensional spatial data database 102, a host system 104, and a terminal 103. These are configured to be connectable by, for example, a LAN. The terminal 103 is configured by a PC that visually edits spatial data and camera images, for example, using a GUI or the like. A two-dimensional spatial data database 101 and a three-dimensional spatial data database 102 are connected to the host system 104. In the present embodiment, the two-dimensional spatial data refers to two-dimensional spatial data (planar data) when the building is viewed from directly above because the building is an object. If the object is different, the direction of viewing the two-dimensional space data may be different.

  Using the input unit 107 and display unit 108 of the terminal 103, the user can transfer camera image data, editing data, and the like to the host system 104. The three-dimensional space database 102 stores the three-dimensional space data created by the three-dimensional space data creation device according to the present embodiment.

  The host system 104 includes a search processing unit 105 and a three-dimensional spatial data creation unit 106. Generally, it has a CPU (control unit) and a memory (storage unit) and is stored in, for example, a non-volatile memory (HDD or the like), and performs the processing shown in FIG. 2 on the computer (CPU). By executing it, the following functions will be demonstrated. Details of these functions will be described later.

  FIG. 2 is a flowchart showing a processing flow of the three-dimensional spatial data creation method according to the present embodiment. For example, FIG. 2 is a diagram showing an example of processing by the system shown in FIG. The processing shown in FIG. 2 is also shown in the description after FIG.

  FIG. 3 is a diagram illustrating an example of a camera image according to the present embodiment, and the object is an example but a building. A camera image 301 shown in FIG. 3 is a camera image of a building taken from the position (x, y, z) of the camera 314. The position of the camera is determined by a three-dimensional space (x, y, z).

  The vertex A ′ 302 to the vertex F ″ 313 in the camera image 315 are each vertex of the building. By defining spatial coordinates for each of the vertices 302 to 313, three-dimensional spatial data can be created based on the camera image 301.

  FIG. 4 is a diagram showing the two-dimensional space data of the object corresponding to the captured camera image 315 shown in FIG. 3, and in this case, the data corresponds to a plan view of the building as viewed from directly above. . As shown in FIG. 4, from the two-dimensional space data 401, two-dimensional coordinates (x, y) up to the vertices A402 to F407 of the building that is the object can be acquired.

  When the camera image 315 is input from the terminal 103 (step S1 in FIG. 2), the data of the camera image 315, the camera direction, and the position data are passed to the search processing unit 105 of the host system 104 (step S2 in FIG. 2). In the search processing unit 105, the two-dimensional space data 401 including the building shown in the camera image 301 is searched in the two-dimensional space data DB 101 using the GPS position data and camera direction information as search keys (FIG. 2). Step S3).

  FIG. 5 is a diagram showing a state in which both the two-dimensional space data 401 and the camera image 301 searched in step S3 of FIG. 2 are displayed on the display unit 108 of the terminal 103 as a pair (FIG. 2). Step S4).

  The user takes a correspondence between the camera image 301 and the two-dimensional space data 401 by, for example, a GUI operation, and defines the value of the coordinate space in the camera image 301. For example, when the vertex A ′ 302 of the camera image 301 is determined by a mouse operation (see a white arrow) or the like and the vertex A 402 to be matched in the two-dimensional space data 401 is clicked, the camera image 301 is automatically assigned the spatial coordinates. A value is set. Currently, there is a pointer at point A, and the coordinates of the ground are given to this feature point. Then, the same value is copied and pasted to the vertex A 'on the image coordinates. Thus, as shown in the lower part of FIG. 5, the latitude and longitude of each vertex (feature point and representative point) are given in the camera image. The latitude and longitude of this representative point are also displayed at the bottom of the screen. In this way, data can be input.

  Here, the vertices of the building that can be determined in the camera image 301 are A ′ 302, B ′ 303, C ′ 304, and E ′ 306. Other vertices cannot be recognized by the camera image 301. The spatial coordinates of the two-dimensional spatial data 401 can be defined for each other vertex that can be acquired from the camera image 301 (step S5 in FIG. 2).

  The camera image 301 and the two-dimensional space data 401 whose spatial coordinates are defined at the vertices are sent to the three-dimensional space data creation unit 106 of the host system 104, and the three-dimensional space data creation unit 106 creates the three-dimensional space data. (Step S6 in FIG. 2).

  FIG. 6A is a diagram illustrating a state of measuring the height of a camera image. Here, the height can be obtained under the condition that the center 601 of the camera image is the center of the shooting direction.

  First, the center (camera direction center) 601 of the camera image is set, and an intersection 602 between a perpendicular drawn from the center 601 of the camera image and a house frame on the camera image is set on the camera image. Next, the spatial coordinates of the set intersection point 602 are calculated from the camera lens model and the two-dimensional spatial data.

  FIG. 6B is a diagram illustrating a relationship between a camera image and two-dimensional space data. An intersection 602 on the camera image 604 coincides with an intersection 602 between a perpendicular drawn from the center 601 set in FIG. 6A and a house frame (outline) on the camera image. This intersection point 602 is set on a line segment E'D 'of the house frame. The vertexes E ′ and D ′ of the house frame on the camera image 604 coincide with the vertex D ′ 305 on the camera image and the vertex E ′ 306 on the camera image in FIG. 3.

  The vertexes E ′ and D ′ of the house frame on the camera image 604 pass through the focal point 603 of the camera lens, and correspond to the vertex D (x4, y4) 405 and the vertex E (x5, y5) 406 of the two-dimensional space coordinates. It is attached. The vertices D405 and E406 coincide with the vertices D405 and E406 in FIG. That is, the intersection 602 on the camera image can be associated with the point 602 ′ on the line segment DE through the focal point 603 of the camera lens. Since the point 602 ′ on the line segment DE becomes a point on the actual two-dimensional map data, the ratio from the vertex D405 to the point 602 ′ on the line segment DE and the ratio from the point 602 ′ on the line segment DE to the vertex E406 Thus, the spatial coordinates (x01, y01) of the point 602 ′ on the line segment DE can be calculated. That is, the spatial coordinates of the intersection 602 in FIG. A can be calculated to be (x01, y01).

  FIG. 6C is a diagram illustrating a figure for obtaining the height D3 to the center of the camera image. FIG. 6D is a diagram illustrating a figure for obtaining the height of a building. These figures show the process of calculating the actual building height from the above results.

First, distances D1 605, D2 606, heights Z 607, and D3 608 shown in FIG. 6C are obtained. The height Z 607 is z in the position information (x, y, z) of the camera. The distance D1 605 can be obtained from the coordinate value (x01, y01) of the point 602 and the camera position information (x, y, z) by the following equation.
D1 ² = (x−x01) ² + (y−y01) ²

The distance D2 606 can be calculated by the following equation because information on the angle θ609 can be obtained from the gyro direction of the camera.
D2 = Z / tanθ

Therefore, the desired height D3 608 is D3 = (D2 + D1) z / D2.
Can be calculated with

  The height D3 608 and the remaining height D4 610 of the building in FIG. 6D are calculated from the ratio of pixel values, for example. As a result, the height of the house frame can be calculated as D3 + D4. That is, if the gamera image is digital data, since the number of pixels is known, the ratio between D3 and D4 is also known. D4 can be determined based on D3 and this ratio. You may measure the length directly from the photo. FIG. 7 is a diagram for defining spatial coordinates of unviewable viewpoints 702 and 703 that cannot be actually recognized by the camera image 301. Even if the vertices cannot be obtained from the camera image, the two-dimensional space data is used to create vertices (for example, 702 and 703) that cannot be seen in the camera image by interpolation from the two-dimensional space data. be able to. When the object is a building, plane data as shown in FIG. 4 is obtained as the two-dimensional space data, for example. Using this two-dimensional spatial data, all the positions that cannot be seen in the three-dimensional camera image, for example, all the vertices at the positions where the vertices at the bottom of the house are actually invisible in the image, such as the fence, are all interpolated. The spatial coordinates of the vertices can be obtained.

  FIG. 8 is a diagram in which spatial coordinate values are given to all vertices from the above results. Since the coordinates defined by these vertices are not image coordinates but space coordinates (world geodetic system etc.), they can be arranged on the map as they are (step S6 in FIG. 2).

  The three-dimensional spatial data created in this way is stored in the three-dimensional spatial data database 102 (step S7 in FIG. 2).

  As described above, according to the three-dimensional spatial data creation technique according to the present embodiment, it is possible to create three-dimensional spatial data at low cost using a camera image. In addition, a plurality of camera images are not required, and a portion that cannot be recognized by the camera images can be interpolated using two-dimensional spatial data. For example, 3D spatial data can be created using 2D spatial data even when the entire house cannot be seen due to a fence or behind a house that cannot be seen from a camera image.

  In addition, the coordinates of the created 3D space data are not image coordinates, but directly give coordinates of the geodetic system, so that they can be projected directly onto the map, and a 3D map can be easily created.

  In the above-described embodiment, the configuration and the like illustrated in the accompanying drawings are not limited to these, and can be appropriately changed within a range in which the effect of the present invention is exhibited. In addition, various modifications can be made without departing from the scope of the object of the present invention.

  In addition, a program for realizing the functions described in the present embodiment is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system and executed to execute processing of each unit. May be performed.

  The present invention can be used as a three-dimensional spatial data creation device.

DESCRIPTION OF SYMBOLS 101 ... Two-dimensional spatial data database, 102 ... Three-dimensional spatial data database, 103 ... Terminal, 104 ... Host system, 107 ... Input part, 108 ... Display part.

Claims (5)

A three-dimensional spatial data creation method for creating three-dimensional spatial data from two-dimensional spatial data by computer processing based on a photographed image of an object associated with a photographing direction and a photographing position,
Retrieving two-dimensional spatial data of the object from a two-dimensional spatial database storing two-dimensional spatial data based on the photographing direction and photographing position;
Comparing the retrieved two-dimensional space data with the captured image and associating the vertex of the two-dimensional space data of the object with the vertex of the captured image;
3D space data generation, comprising: obtaining the height of the object using the shooting direction and the shooting position under the condition that the center of the shot image is the center of the shooting direction Method. The step of obtaining the height of the object comprises
The center of the photographed image (the center of the photographing direction) is set, the intersection of the perpendicular drawn from the center and the outline of the object is set, and the distance between the intersection and the center is determined from the photographing position and the center. A step of obtaining from the photographing angle and a sum of a first distance with an intersection, a photographing angle obtained from the photographing position and the photographing direction, and a second distance obtained from a height of the photographing position;
Obtaining a height of the object from a first height which is a distance between the intersection and the center and a remaining second height of the object, wherein the second height is determined as the first height; The method according to claim 1, further comprising a step of calculating and adding from a ratio between a height of the image and a pixel value of the photographed image.   The three-dimensional space according to claim 1, further comprising a step of interpolating the coordinates of the three-dimensional space data at a position not visible in the captured image based on the two-dimensional space data. Data creation method.   The program for making a computer perform the three-dimensional spatial data creation method of any one of Claim 1 to 3. A three-dimensional spatial data creation device that creates three-dimensional spatial data from two-dimensional spatial data by computer processing based on a photographed image of an object associated with a photographing direction and a photographing position,
A search processing unit for retrieving the two-dimensional space data of the object from a two-dimensional space database storing two-dimensional space data based on the photographing direction and the photographing position;
The searched two-dimensional space data and the photographed image are contrasted, the vertex of the two-dimensional space data of the object and the vertex of the photographed image are associated, and the center of the photographed image is defined as the photographing direction. And a three-dimensional spatial data creation unit that obtains the height of the object using the photographing direction and the photographing position under the condition of the center of the image.

Images