JPH07113535B2 - Surface inclination measuring device - Google Patents

Description

Description: (1) Technical Field of the Invention According to the present invention, two slit lights in different directions are simultaneously projected onto a three-dimensional object plane, and a projected image is taken by a television camera to be taken on an image pickup plane. The present invention relates to a surface inclination measuring device for deriving a surface vector from an image and measuring the inclination of the surface of a three-dimensional object.

(2) Background of Technology Although industrial robots have been remarkably developed in recent years, it is necessary to three-dimensionally recognize an object in a work area when moving the robot or causing the robot to perform work. In three-dimensional object recognition, obtaining the distance to the object and the tilt information of the surface of the object is a particularly important issue.

(3) Conventional Technology and Problems Conventionally, a method for detecting the surface vector of a three-dimensional object is performed based on the principle shown in FIG. That is, the change in the brightness of the object plane 2 when the light source 1 is rotated around the Z axis is captured and analyzed by the camera 3. In this case, it is assumed that the brightness I _R has a sine wave-like change with a direct current component represented by the following relational expression, I _R (β) = a + bcosβ + csinβ (1), and the surface vector is derived from its phase and amplitude. Proposed as a method. This method has a problem that it takes a very long time because the light source 1 has to be rotated around the Z-axis and changes in I _R accompanying it have to be measured step by step, and when the object surface 2 is not a perfect diffusing surface. Has a problem that the sine wave of I _R is distorted and the measurement error increases. On the other hand, a method of calculating a surface vector based on the distance between the light source, the object surface, and the camera has also been proposed, but in this case, the distance when three points are not on a straight line must be measured, and therefore data processing is performed. It has a problem that it takes time.

(4) Object of the invention In view of the above problems, the object of the present invention is to project two slit light beams in different directions at the same time onto a three-dimensional object plane, and to obtain the x and y components of the projected image on the photographing plane. It is an object of the present invention to provide a surface inclination measuring device that measures the inclination of the surface of an object arbitrarily placed in a three-dimensional space by deriving a surface vector by obtaining it.

(5) Structure of the Invention According to the present invention, the object is to provide a projector for projecting two slit light beams having different directions onto a three-dimensional object, and a projector for projecting the slit light beam onto the three-dimensional object. A television camera that captures a projected image of slit light, an image processing unit that processes the captured image of the television camera as binarized data, and coordinate calibration that calibrates the position coordinates of the projector and the television camera. Provided is a surface inclination measuring device comprising a calculation section, a surface vector calculation section for calculating a surface vector of a projection image on the photographed image plane, and a control section for controlling directions of the projector and the television camera. It is achieved by

(6) Embodiment of the Invention FIG. 2 is a block diagram showing an embodiment of the surface inclination measuring device according to the present invention. In FIG. 2, 4 is a projector (projector) which has two mutually orthogonal slits and projects slit light onto a three-dimensional object from the slits, and 5 shows a projection image by the projected slit light. Reference numeral 6 denotes a TV camera for photographing, 6 is an image processing unit for processing a photographed image of the TV camera as binary data, and 7 is a coordinate calibration calculation for calibrating position coordinates of the projector 4 and the TV camera 5. 8 is a plane vector calculation unit for deriving a plane vector from the x and y components on the photographic image plane, and 9 is a control unit for controlling the directions of the projector 4 and the television camera 5. FIG. 3 is a perspective view showing a spatial arrangement relationship between the projector 4 and the television camera 5, FIG. 4 shows a coordinate relationship at the time of calibration in the spatial arrangement shown in FIG. 3, and FIG.
The coordinates of the image on the image pickup surface of the television camera in the spatial arrangement of the figure are shown, and FIG. 6 is a diagram showing the coordinate relationship when a three-dimensional object has an arbitrary inclination by using the apparatus shown in FIG.

In such a configuration, the coordinate system O _p − of the projector 4 is
x _p y _p z _p x _p to the optical axis and z _p perpendicular to the, y
Determine the _p- axis. On the other hand, with respect to the coordinate system O _c −x _c y _c z _c of the camera 5, the horizontal axis of the imaging surface is selected as x _c and the vertical axis is selected as y _c , and z _c is determined to be orthogonal to these (see FIG. 4). And O _p −x _p y _p z _p
Is O _c −x _c y _c z _c , α around the z _c axis, β around the y _c axis,
It is supposed to rotate γ around the x _c axis. The first calibration of both coordinate systems is to obtain the Euler angles α, β, γ.

Here, the plane π _o is set to be orthogonal to the z _p axis. That is, π _o ⊥z _p (2) For the slit light of the projector 4, as shown in FIG. 4, a slit having a length s on the x _p axis and a long light on the y _p axis are provided as shown in FIG. Consider a slit of size t. That is, = sp (3) = tp (4) where dot · indicates that it is a vector quantity.
i, j, k are unit vectors. The two slit light beams that are orthogonal to each other are projected on the surface π _o , and the projected images formed on π _o at that time are respectively defined as and. According to the equation (2), = sp (5) = tp (6) where the O _p −x _p y _p z _p system is the Euler angles α, β with respect to the O _c −x _c y _c z _c system. , Γ, and the transformation matrix is (α, β,
If γ), the following equation is obtained.

Using this relational expression, the expressions (5) and (6) are: = s cosα cosβc + s sinα cosβc−s sinβ
c (9) = t (-sinα cosγ + cosα sinβ sinγ) c + t (cosα cosγ + sinα sinβ sinγ) c + t cosβ sinγc (10) On the other hand, the expression of, on the O _c −x _c y _c z _c system, = Xec + Yec + Zec ... (11) = Xfc + Yfc + Zfc ... (12) From the relationship between (9) and (10), Xe = s cosα cosβ (13) Ye = s sinα cosβ (14) Ze = -s sinβ (15) Xf = t (-sinαcosγ + cosα sinβ sinγ) (16) Yf = t (cosα cosγ + sinα sinβ sinγ) (17) Zf = t cosβsinγ (18) where Xe and Ye are on the respective imaging planes. Since x _c and y _c components of Xf and Yf are x _c and y _c components of, they can be measured (see FIG. 5).

From (13) and (14), Ye / Xe ＝ tanα, ∴α ＝ tan ^-1 Ye / Xe… (19) Also, from Eqs. (13) to (15), Ze ² / A ² ＝ s ² sin ² β / s ² cos ² β From (16) to (18) Therefore, α, β, γ can be obtained by using (19) to (21). However, for decoding β and γ, the O _p −x _p y _p z _p system,
It can be determined by the positional relationship of O _c −x _c y _c z _c .

Since the calibration is completed, the projector and the TV camera are rotated by Euler angles α _p , β _p , γ _p , α _c , β _c , γ _c from the reference plane to the object plane π ₁ having an arbitrary inclination. Then, the derivation from and when the slit light is projected is performed with reference to FIG.

First, the coordinate systems after rotation of the projector and the camera are respectively O _p ′ −x _p ′ y _p ′ z _p ′ and O _c ′ −x _c ′ y _c ′ z _c ′.
O _p ′ −x _p ′ y _p ′ z _p ′ is for O _p −x _p y _p z _p Also, O _c ′ −x _c ′ y _c ′ z _c ′ is O _c −x _c y _c z _c And this formula is Can be rewritten as Here, the dash "'" indicates the transposed matrix of. Therefore, O _p ′ −x _p ′ y _p ′ z _p ′ is O _c ′ −
For x _c ′ y _c ′ z _c ′ here, Suppose that

On the other hand, the captured image on the object plane is of the O _p ′ −x _p ′ y _p ′ z _p ′ system.
Since it is in the x _p ′ z _p ′ plane, However Substituting equation (28) into equation (27), Xeh ₂₁ + Yeh ₂₂ + Zeh ₂₃ = 0 (29) Ze that cannot be measured However, h ₂₃ ≠ 0. Also, the captured image on the object plane
Since it is in the y _p ′ z _p ′ plane of the O _p ′ −x _p ′ y _p ′ z _p ′ system, However, When is substituted into the above formula, Xfh ₁₁ + Yfh ₁₂ + Zfh ₁₃ = 0 (33) However, h ₁₃ ≠ 0.

Finally, when the surface vector of π ₁ is derived, the surface vector is given by the outer product of and. Therefore, Therefore, the direction cosine for the O _c ′ −x _c ′ y _c ′ z _c ′ system of π ₁ is 7 and 8 show the operational flow of the device according to the present invention.
In FIGS. 7 and 8, at the time of calibration, the coordinate systems of the projector and the camera installed at the reference point are calibrated. Here, for simplification in analysis, two orthogonal slit lights are projected onto the reference plane π ₀ . The image processing unit 6 performs threshold processing to detect and obtain x and y components. Next, the coordinate calibration calculation unit 7
In equation (19) to (21), the Euler angle α,
Calculate β and γ. This calibration does not need to be performed if it is performed first unless the positional relationship is changed.

Next, the case of deriving the inclination of an arbitrary surface will be described.
Control the orientation of the projector and camera to point the target at the target point. The control unit 9 sets the postures of the projector and the camera to x, y,
It is provided to have a function that allows the posture to be freely changed around the z and axis. The control unit 9 can read the rotation angle from the reference point. In the coordinate calibration calculation unit 7, α, β, γ, α _p , β _p , γ _p , α _c , β _c , γ _c
From _o . Next, in the image processing unit 6, the value on the image pickup surface is detected by threshold processing to calculate the lengths of the x and y components. Further, in the surface vector calculation unit 8, Ze and Zf are derived using the equations (30) and (34), and the equation (3
6), u, v, w are obtained from (37) and the direction cosine of π ₁ is derived from Eq. (38).

(7) Effects of the Invention With the surface inclination measuring device according to the present invention, the inclination of the object surface can be measured in a relatively short time, for example, on the order of several seconds.

[Brief description of drawings]

FIG. 1 is a diagram for explaining a conventional method, FIG. 2 is a block diagram of a surface inclination measuring device as an embodiment of the present invention, and FIG. 3 is a projector and a television camera in the device of FIG. FIG. 4 is a perspective view showing the spatial arrangement of FIG. 4, FIG. 4 is a diagram for explaining the coordinate relationship at the time of calibration in the spatial arrangement of FIG. 3, and FIG. 5 is the image pickup surface of the television camera in the spatial arrangement of FIG. FIG. 6 is a diagram for explaining the coordinate relation of the images, FIG. 6 is a diagram for explaining the coordinate relation in the spatial arrangement when the surface of the three-dimensional object in the apparatus of FIG. 2 has an arbitrary inclination, and FIG. 2 is a flow chart showing a calculation flow at the time of calibration in the apparatus of FIG. 2, and FIG. 8 is a flow chart showing a calculation flow of deriving the inclination of the object plane in the apparatus of FIG. 1 ... Light source, 2 ... 3D object plane, 3, 5 ... TV camera, 4 ... Projector, 6 ... Image processing unit, 7 ... Coordinate calibration calculation unit, 8 ... Plane vector calculation unit, 9 ...... Control unit.

Claims (1)

[Claims] 1. A projector for projecting two slit lights in different directions onto a three-dimensional object, a television camera for photographing a projected image of the slit light projected on the three-dimensional object, An image processing unit that processes an image captured by a television camera as binary data, a coordinate calibration calculation unit that calibrates the position coordinates of the projector and the television camera, and a plane of the projected image on the captured image plane. A surface inclination measuring device comprising a surface vector calculation unit for calculating a vector and a control unit for controlling the directions of the projector and the television camera.