JPH0436328B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] A method for measuring the lens optical axis of an imaging system, in which parallel light is incident on the imaging system while rotating the imaging system about a rotation axis parallel to the lens optical axis, and the optical axis is measured on the imaging surface. By measuring the optical axis position on the imaging plane based on the trajectory of the image, the position of the optical axis on the imaging plane can be easily measured.

[Industrial Application Field] The present invention relates to a method for measuring the optical axis of an imaging system lens, which measures the position of the optical axis of an imaging system lens on an imaging surface.

Generally, once a lens is assembled, there is no need to measure the optical axis to determine where the optical axis is. However, when performing three-dimensional three-dimensional recognition using a television camera, for example, it is difficult to perform accurate three-dimensional recognition unless you know which pixel on the imaging surface corresponds to the optical axis, which is the reference for position measurement. I can't.

[Prior Art] Conventionally, even when performing image recognition using a television camera as an imaging system, it is not necessary to know exactly where the optical axis of the lens is located on the imaging surface of the television camera. It was sufficient to align the optical axes of the lens and camera using the mechanical precision of the camera and lens.

[Problems to be Solved by the Invention] As described above, such a technique has not existed in the past, and there is a need for a technique that accurately and quickly determines which position on the imaging surface the optical axis corresponds to.

[Means for Solving the Problems] In order to provide the above-mentioned technique, the present invention first sets the optical axis of the imaging system lens and the rotation axis in parallel when measuring the optical axis of the imaging system lens, Then, this imaging system is rotated around the rotation axis, parallel light is incident on the lens, and the position of the optical axis on the imaging surface is measured based on the locus of the image formed on the imaging surface of this parallel light. .

[Function] As shown in FIG. 1, the optical axis A and rotation axis B of the imaging system consisting of the thin lens 1 with the focal length f and the imaging surface 3 are parallel, and the direction of incidence of the parallel light 2 is the optical axis. A and an arbitrary angle θ, and when the image is formed on the imaging surface 3,
The image 5 is generated from the intersection 4 of the optical axis A and the image plane 3 by γ(=f
are connected at points separated by tanθ).

Then, when this optical system is rotated around the rotation axis B, the imaging point of the parallel light 2 moves and its locus changes between the optical axis A and the imaging plane 3 as shown in FIG. Draw a circle with radius γ centered at intersection 4.
Therefore, by detecting the locus of this circle and finding the center of this circle, the position of the intersection point 4 on the imaging plane 3 can be measured. Note that O in the figure indicates the optical center of the lens.

Further, as in the above case, when the optical axis A and the rotation axis B are parallel, and in addition, when the incident direction of the parallel light is parallel to these, as shown in FIGS. 3 and 4, , an image 5 of parallel light is formed on the intersection 4 of the optical axis A and the imaging plane, and the optical system is rotated along the rotation axis B.
Even when rotating around , the image 5 is always focused on the intersection point 4. Therefore, if the position of the image 5 does not move on the image plane even when the optical system is rotated, the position of the intersection of the optical axis A and the image plane can be measured.

This is true not only for the thin lens shown in this example, but also for combination lenses such as camera lenses, if we consider the positions of the principal points and nodes of the lens system.
It will be exactly the same.

[Example] Hereinafter, an example of the method for measuring the optical axis of a lens according to the present invention will be described. First, as a first example, a case will be described in which the optical axis of the lens and the rotation axis are parallel to each other, which corresponds to FIGS. 1 and 2. In this embodiment, in order to make the rotation axis and the lens optical axis parallel, the good mechanical precision of the annular member on the end surface of the lens 11 is utilized. That is, as shown in FIG. 2, a lens to be measured 11 is attached to a television camera 10 and equipped with a fine movement device such as a two-axis goniometer, so that the direction of the optical axis A can be slightly changed from the vertical direction. Furthermore, this mounting base 12 is attached to a rotary table 13 attached to a base 15 so that the optical axis is directed in the vertical direction. Here, since the direction of the rotation axis B of the rotary table 12 is substantially perpendicular to the base 15, the optical axis A and the rotation axis B are substantially in the same direction.

Then, while rotating the turntable 12, the displacement of the annular member 16 fixed to the tip of the lens 11 from the base 15 is measured, and the goniometer of the mounting base 12 is adjusted to determine the displacement of this end 16 from the base 15. Let the change in be zero.

This displacement is measured by a non-contact type microsensor 18 provided on a stand 17 mounted on a base 15, as shown in FIG.

Here, the lens 11 is designed and manufactured on the assumption that it is rotationally symmetrical with respect to the optical axis A except in special cases, so its mechanical precision is high. The general eccentricity tolerance given to a single lens is 1' to 2' inclination and 0.01 mm to 0.02 mm in displacement. (Refer to ``Lecture 23'' Problems Associated with Lens Design (1)). Therefore, the accuracy of the lens tip is sufficiently good. For example, when a lens with a diameter of 100 ml is rotated and the displacement of the tip in the optical axis direction is 10 μm or less, the optical axis deflection of the lens is
It is less than 0.006°. It is quite possible to measure the displacement on the order of 1 μm with a general displacement detection mechanism, and if a sufficiently accurate fine movement device for the mount 12 is used, the displacement of the lens tip in the optical axis direction can be measured on the order of 10 μm. It has been confirmed through experiments that the following is easy to do. The above figure of 0.006° is accurate enough for practical use, so
This means that the lens optical axis and rotation axis can be set parallel.

Next, as shown in FIG. 3, the lens 11 is irradiated with parallel light from the collimator 19 at an arbitrary angle θ with respect to the optical axis A, and the rotating table 13 is rotated to rotate the lens 11 and camera 10. Rotate around axis B. At this time, the lens should be focused at infinity.

Then, an image trajectory as shown in Fig. 1 and 2 appears on the camera's imaging surface, and this circle is detected by an image sensor placed on the imaging surface and electrically processed to determine the center of this circle. Once determined, the position of the optical axis A on the imaging plane can be determined. Therefore, according to this embodiment, the optical axis position on the imaging plane can be easily measured by simply making the optical axis A and the rotation axis B parallel.

Next, a second embodiment of the present invention will be described.
In this embodiment, the method shown in FIGS. 3 and 4 is performed.

In this embodiment, first, the irradiation direction of the parallel light and the rotation axis B of the rotary table 12 are made parallel. This is achieved by performing autocollimation using a plane mirror, as shown in FIG. That is, a rotary table 20 having a fine movement mechanism, such as a two-axis goniometer, for minutely changing the direction of the rotation axis B is provided on the base 15, and a mirror surface 21a and a bottom surface are formed parallel to each other on this rotary table 20. flat mirror 2
1 and an autocollimator 23 installed on a stand 22 provided on the base 15 on this plane mirror 21.
The fine movement mechanism of the rotating table 20 is adjusted while looking through the eyepiece 23a so that the reflected light is always focused on the light source position of the autocollimator 23 even when the rotating table 20 is rotated. When this adjustment is completed, the direction of the irradiated light from the autocollimator 23 and the rotation axis of the rotary table 20 become parallel.

Next, the rotation axis B of this rotary table and the lens optical axis A are made parallel, but this is as explained in the previous embodiment based on FIG. 2, and detailed explanation will be omitted.

After the optical axis A and the rotation axis B become parallel, the lens 11 is irradiated with parallel light parallel to these axes from the previously used autocollimator 23, and the rotation table is rotated. Then, the image shown in Fig. 14 appears on the imaging surface.
An image as shown in FIG. 1 is formed and does not move due to the rotation of the lens 11, and this point is the desired position of the optical axis on the imaging surface.

Therefore, according to this embodiment, no special processing is required after irradiating parallel light and obtaining an image.
The position of the optical axis on the imaging plane can be determined without using special equipment.

[Effects of the Invention] As explained above, according to the present invention, the optical axis of the lens and the rotation axis are made parallel, the lens is rotated while parallel light is incident on the lens, and the locus of the image of the parallel light is Since the position of the optical axis on the imaging surface can be measured based on the above, it is possible to measure the position of the optical axis on the imaging surface extremely easily and accurately. This provides effects such as rapid measurement and accurate three-dimensional recognition.

[Brief explanation of drawings]

1, 2, 3, and 4 are diagrams for explaining the principle of the present invention, and FIGS. 2 to 5 show examples of carrying out the present invention. When the axis and the rotation axis are parallel, Fig. 3 shows that when parallel light is irradiated onto the lens at an arbitrary angle, Fig.
This figure is a perspective view showing the case where the parallel light irradiation direction and the rotation axis are parallel, and FIG. 5 is a perspective view showing the case where parallel light is irradiated parallel to the lens optical axis. A... Lens optical axis, B... Rotation axis, 2... Parallel light, 3
...imaging surface, 5...trajectory.

Claims (1)

[Claims] 1. The rotational axis of an integrated imaging system including a lens and imaging surface with almost no eccentricity is set parallel to the optical axis of the lens, and the imaging system is rotated around this rotational axis. At the same time, a parallel ray is incident on the lens from a direction at an angle θ with respect to the optical axis, and the center position of the circular image is determined based on the locus of the image formed on the imaging surface of the parallel ray. A method for measuring an optical axis of a lens, the method comprising measuring the optical axis position.