KR100395773B1 - Apparatus for measuring coordinate based on optical triangulation using the images - Google Patents

Abstract

The present invention relates to an apparatus for detecting three-dimensional coordinates of the measuring points of the measurement object by imaging the measurement object using the source camera and the target camera, the source camera and the target camera of the present invention are respectively installed on one fixed axis Each object to be measured is imaged, and the processor calculates three-dimensional coordinates of the measurement points captured by the source camera and the target camera

(Where b denotes the center distance between the cameras in the X _T axis direction, (u _Ti , v _Ti ,) denotes the coordinate of the i th object point in the target image, and (u _To , v _To ,) denotes the target image. The image center coordinate θ i is detected as the rotation angle of the source camera with respect to the object point i or the rotation angle of the target camera with respect to the object point i) and displayed on the display unit.

That is, in the present invention, the three-dimensional position of the measurement points is determined by using the image coordinates of the measurement points of the measurement object captured by the source camera and the image coordinates of the measurement points of the measurement object captured by the target camera. There is an effect that it is possible to collectively detect the three-dimensional position with respect to the plurality of measurement points.

Description

Optical triangulation three-dimensional measuring device using two photos {APPARATUS FOR MEASURING COORDINATE BASED ON OPTICAL TRIANGULATION USING THE IMAGES}

The present invention relates to an apparatus for measuring three-dimensional coordinates of an object, and more particularly, optical triangulation three-dimensional measurement using two photographs capable of simultaneously detecting three-dimensional coordinates of several measurement points in a measurement object using at least two photographs. Relates to a device.

As a method of measuring three-dimensional coordinates of an object, a triangulation method using a geometric relationship between a light source forming a measuring point on an object and a sensor that recognizes the measuring point is widely used. Such methods are largely structured light and stereo. It is divided into stereo vision method.

The structured light method is a method of illuminating a measurement object using a light source having a geometrical pattern, and obtaining an image of an illuminated object to measure coordinates of a measurement point. Among the disadvantages of using the laser projector 11 and the camera 12 as shown in FIG. 1, the projection angle after forming the measuring point 13 on the object by using the laser beam of the laser projector 11, the projection angle (θ) ) And the coordinates of the measurement point 13 using the position (x ', y') of the image point 15 formed on the image plane of the camera 12 and the distance b between the laser 11 and the camera 12. It is a method of measuring (x, y, z).

Using the geometry of FIG. 1, the three-dimensional coordinates of the measuring point 13 are calculated from equation (1).

Here, (x, y, z) ^T is three-dimensional coordinates of the measurement point 13, (x ', y') ^T means image coordinates, f means the focal length.

However, the above-described structured light method requires a lot of measurement time because of the sequential measurement in units of points, and has a disadvantage in that it adversely affects the accuracy of the measured value.

2 shows a conventional stereo vision method. In FIG. 2, a point P on an object existing in space is projected on both cameras to form a pair of points called conjugate pairs (p _l , p _r ) on the image plane. The plane (ΔC _l C _r P), which is made up of the optical center of the camera (C _l , C _r ) and the point of the object in space projected by the camera, is called the epipolar plane, and the intersection of the epipolar plane and the image plane Is defined as the epipolar line. Two points on the image plane have the same y coordinate, and the object point on the left image is placed on the epipolar line of the right image. The points that appear in the left image may or may not exist in the right image, but if they do, they will always appear on the epipolar line. Motion errors generated during camera movement or numerical errors inserted in the calculation process can cause epipolar lines to shift, resulting in vertical disparities corresponding to vertical mismatches, but many stereo vision algorithms assume zero zero.

If two cameras are firmly connected and the optical axes are parallel and separated by a distance (b), then the line connecting the centers of two camera lenses separated by the distance (b) only in the x-axis direction is called a baseline. do. In FIG. 2, the image planes are modeled in the same plane, and if the base line is perpendicular to the optical axis and in the direction of the x axis, the image planes lie parallel to the x axis.

It is assumed that the coordinates (X, Y, Z) of the object in space are measured with the lens center of the left camera as the origin. Each of the image coordinates of the left image and the right image _{(x 'l, y' l} ), (x 'r, y' r) Let the corresponding point (P) on the store the left image (p _l) and the object From the relationship, the relationship as shown in equation (2) is established.

Similarly, Equations 3 and 4 are established from the correspondence between the shop p _r of the right image and the point P on the object.

Here, f is a focal length, which is the distance from the lens center of each camera to the image plane.

The coordinates (x ' _l , y' _l ) and (x ' _r , y' _r ) of the left and right image planes can be measured, and the focal length f is determined by camera calibration, so At 2, 3, and 4, we don't know anything about camera coordinates of objects in space, so we can find X, Y, and Z.

In Equations 2 and 3, the following Equation 5 can be obtained.

The difference between x coordinates in image coordinates x ' _r -x' _{l In} other words, the difference between two points corresponding to a conjugate pair when two images are combined is called disparity. Equation 6

On the other hand, as described above, the conventional structured light method mainly uses a laser as a light source to form a measurement point on an object, and at this time, the three-dimensional measurement point based on the image coordinate of the image point formed on the image plane and the positional relationship between the light source and the camera. Find the coordinates. However, this method must include the process of scanning the object because the measurement in points is made, it takes a lot of measurement time and adversely affects the measurement results. In addition, since the result depends on the reflectance of the object surface, even if the same shape, there is a problem that it is difficult to obtain the same result depending on the material or the state of the surface.

In addition, the stereo vision method is a method of measuring three-dimensional coordinates using a pair of images obtained from two parallel cameras. As the length of the baseline increases, the range of overlapping images decreases. The disadvantage is that it is difficult to secure the measurement area.

SUMMARY OF THE INVENTION The present invention has been made to solve this problem, and an object of the present invention is to provide an optical triangular three-dimensional measuring apparatus using two photographs for calculating three-dimensional coordinates of a measuring point using a CCD camera, which is a virtual light source, instead of an actual light source. .

In order to achieve the above object, the present invention provides a device for detecting the three-dimensional coordinate position of the measurement points located on the measurement object, a source camera for imaging the measurement object; A target camera for photographing the measurement object at a predetermined distance from a source camera; A display unit capable of displaying three-dimensional coordinate values; Three-dimensional coordinates of the measurement points captured by the source camera and the target camera

(Where b denotes the center distance between the cameras in the X _T axis direction, (u _Ti , v _Ti ,) denotes the coordinate of the i th object point in the target image, and (u _To , v _To ,) denotes the target image. The image center coordinate θ i is provided as a processor for detecting and displaying the rotation angle of the source camera with respect to the object point i or the rotation angle of the target camera with respect to the object point i).

1 is a view showing a method of detecting the position of a measuring point using a conventional structured light method;

2 is a diagram illustrating a method of detecting a position of a measurement point using a conventional stereo vision method.

3 is a schematic diagram of an optical triangulation three-dimensional measuring apparatus using two photographs according to the present invention;

4 is a view showing a geometric relationship between a source camera and a target camera and a coordinate system used according to the present invention;

A 5 is the starting by the center projection in consideration with a laser beam axis in Figure 1 the projection center (O _S) of the source camera showing the geometric relationship of the light beam entering the projection center of the target figure,

FIG. 6 is a view showing a geometric relationship when the projection center of the lens is separated from the camera center of the source and the target by a predetermined distance b _{z in the} Z _T direction without being placed on the X _T axis line;

7 is a block diagram of an optical triangulation three-dimensional measuring apparatus using two photographs according to the present invention;

8 is a schematic diagram showing another embodiment of an optical triangulation three-dimensional measuring apparatus using two photographs according to the present invention.

<Description of the symbols for the main parts of the drawings>

32: source camera 33: target camera

40: processor 50: display unit

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention;

3 shows a schematic diagram of a system for practicing the present invention. As shown, the present invention includes two cameras 32 and 33 fixed to one parallel fixed shaft 31, one of which is rotated about the lens axis. It is configured to be possible. In this configuration, the camera 32 captures an image to be imaged 34 to be imaged in a fixed state, and the camera 33 rotates around the lens axis to capture the image in a state in which the image angle with respect to the image object 34 is set. Do it.

According to the present invention having the above-described configuration, it is possible to obtain three-dimensional coordinates of a measurement point by assuming that each pixel photographed by the cameras 32 and 33 is a single point light source, and to obtain image coordinates of the image point corresponding thereto.

4 shows a geometric relationship between a source camera providing a virtual light source (camera 33 in this embodiment) and a target camera providing an image point (camera 32 in this embodiment). The coordinate system used is shown. As shown, the source camera 33 is positioned with a certain angle θ ₁ about the target camera 32.

In FIG. 4, the measurement point P forms an image point on the image plane passing through the respective camera projection centers O _T and O _S. Considering the projection center O _S of the source camera 33 as the laser optical axis of FIG. 1, the geometric relationship of light rays starting from the projection center into the projection center of the target may be modeled as shown in FIG. 5.

Referring to FIG. 5, it can be seen that one image of the source camera 33 serves as a multiple light source as a plurality of laser light sources simultaneously project rays. Therefore, by applying the equation 1 by obtaining the projection angle (θ _i ) of the light ray for each light ray it is possible to obtain the three-dimensional coordinates of the measurement points.

The rotation angle θ ₁ of the source camera 33 with respect to the target camera 32 is known exactly as a configuration condition of the measurement system as in the structured light method, and the projection angle of each ray has an image plane at the projection center. It can be obtained as shown in Equation 7 through the distance and the image coordinates.

Here, θ _i represents a projection angle with respect to the object point i, us _i represents a u component among the image coordinates of the i-th point formed on the source, and us _o represents a u component among the image center coordinates of the source. Θ ₁ can also be known from the system configuration conditions as the rotation angle of the source camera 33 with respect to the Y _T axis.

Using the geometry of FIG. 5, three-dimensional coordinates of an object may be calculated as shown in Equation 8.

Here, b denotes the center distance between cameras in the X _T axis direction, (u _Ti , v _Ti ,) denotes the coordinate of the i-th object point in the target image, and (u _To , v _To ,) denotes the image center coordinates. Indicates.

Substantially, the projection center of the lens constituting the camera is as shown in Equation 8 so that the camera centers of the source and the target do not lie on the X _T axis line but are separated by a predetermined distance b _{z in the} Z _T direction. In this case, the geometry of the ray is configured as shown in FIG. 6, and the coordinate calculation equation at this time can be obtained using Equations 9 and 10.

7 is a block diagram of an apparatus for obtaining three-dimensional coordinates of measurement points using the above-described method.

As shown, the present invention includes a source camera 33 and a target camera 32, and a processor 40 for obtaining coordinates of the measurement points using the measurement points captured by the cameras 32 and 33. Here, in order for the processor 40 to perform Equation 8 or Equations 9 and 10, the rotation angle of the source camera 33 ( Information about the measurement points picked up by the source camera 33 and the target camera 32, and the distance between the source camera 33 and the target camera 32 (b or (b). _x , b _z ) information will be needed, and this information will have to be provided to the processor 40 by the operator.

The processor 40 detects three-dimensional coordinates of the measurement points using Equation 8 (or Equations 9 and 10) by using the information provided by the operator and the information about the measured measurement points, and detects the detected 3 Dimensional coordinates are displayed through a separate display unit 50.

Meanwhile, although two cameras 32 and 33 are used in the embodiment of FIG. 3, as shown in FIG. 8, one camera 36 is attached to the single-axis feed stage 35, and the camera 36 is attached. Can be used simultaneously as a source camera and a target camera. In other words, the camera 36 is positioned at an arbitrary position (source position) to capture the measurement object 34. Then, by moving the camera 36 to the target position through the single-axis feed stage 35 and rotating it around the lentz axis, the image of the measurement object is captured and the source imaging screen and the target screen necessary for detecting the three-dimensional coordinates of the measurement point of the measurement object. Can be obtained.

Meanwhile, in the embodiment of FIG. 3, the relative position b between the source camera 33 and the target camera 32 is set, but in the embodiment of FIG. 8, the state position b may be adjusted as necessary. It can be seen that. That is, in the apparatus of FIG. 8, the rotation angle and the relative distance of the camera 36 may be adjusted as necessary.

As described above, in the present invention, by measuring the three-dimensional position of the measurement point by using the measurement point of the measurement object imaged by the source camera and the measurement object imaged by the target camera, the three-dimensional image of the plurality of measurement points formed on the measurement object. Position can be collectively detected, and in particular, the target or source camera of the present invention can change the imaging angle, so that the relative positions of the two cameras are parallel as the imaging angles of the two cameras are parallel to each other, as in the conventional stereo vision. There is an effect that there is no restriction that must be located below a predetermined distance.

Claims (5)

An apparatus for detecting three-dimensional coordinates of the measuring points located on the measurement object, A source camera which picks up the measurement object; A target camera configured to capture the measurement object at a predetermined distance from the source camera; A display unit capable of displaying three coordinate values; Three-dimensional coordinates of the measurement points captured by the source camera and the target camera (Where b denotes the center distance between the cameras in the X _T axis direction, (u _Ti , v _Ti ,) denotes the coordinate of the i th object point in the target image, and (u _To , v _To ,) denotes the target image. Image center coordinate, θ i is the rotation angle of the source camera with respect to the object point (i) or the rotation angle of the target camera with respect to the object point (i)) Optical triangulation three-dimensional measuring apparatus comprising a processor for detecting by the display to display on the display unit. The method of claim 1, The projection center of the source camera or the target camera having a predetermined rotation angle θ i with respect to the object point i is from the projection center of the target camera or source camera having no rotation angle θ i with respect to the object point i. The processor is spaced apart from the predetermined distance in the Z-axis direction, and the processor calculates three-dimensional coordinates of the measurement points captured by the source camera and the target camera. Wherein b _x is a separation distance in the x _T- axis direction, and b _z is a separation distance in the z _T- axis direction. The method according to claim 1 or 2, The rotation angle θ i is And detect, fs is the optical triangulation three-dimensional measuring device, characterized in that represents the u component in the focal length, us _i is the u component from the image coordinates of Rimed i-th point source us _o is the image coordinates of the center of the source of the source image . The method of claim 3, wherein And the source camera and the target camera are installed on one fixed axis, respectively. The method of claim 3, The source camera and the target camera are composed of one camera, and the camera is installed to be movable on the single axis transfer stage, and is driven by the source camera to capture the measurement object and then transferred through the single axis transfer stage. An optical trigonometry three-dimensional measuring apparatus for driving the target camera and imaging the measurement object.