JPH023446B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a method for measuring the slope of a two-dimensional object surface using a deformed grating image obtained by projecting a grating onto the surface of a three-dimensional object and deforming it according to the unevenness of the object surface as basic input data; This invention relates to a method of measuring the distortion of an object using a deformed lattice image, which has been deformed in accordance with the deformation of the object by printing a lattice on the surface of the object, as basic input data.

When measuring the slope (differentiation of shape) of a three-dimensional object using moire topography, generally a moire fringe contour image is used as input data, three-dimensional shape data is obtained from this, and then it is numerically differentiated. That's what I was looking for.

This moire fringe contour image is obtained by superimposing a deformed lattice image deformed according to the surface shape of a three-dimensional object and a reference lattice image.

Numerical differentiation of 3D shape data is troublesome, and furthermore, extracting 3D shape data from a moiré contour image requires a great deal of effort.

Naturally, various automatic analysis and processing methods for these moiré contour images have been attempted. However,
It has been found that it is extremely difficult to completely automate the analysis of moiré contour images. In other words,
This is because the direction of the unevenness of the object cannot be identified from only the moire contour image, so it is necessary for a human being to determine this and convey that information to the processing device.

Against this background, it has been proposed to use deformed grid images instead of moiré contour images as basic input information for three-dimensional shape analysis of object surfaces. For example, there is a method as shown in Japanese Patent Application Laid-Open No. 116263/1983.

Furthermore, when measuring the distortion of a two-dimensional object using the moire method, the situation similar to the above-mentioned moire topography can be said. In other words, in the method of measuring two-dimensional object distortion using the Moiré method, the reference grid of the object before deformation is printed, it is recorded on a photograph, etc., the object is deformed, and then the so-called deformed grid distorted by the deformation is recorded. The reference grid image recorded before deformation and the deformed grid image recorded after deformation are again recorded on a photograph or the like and superimposed to form a moiré equidisplacement line pattern. Then, the distortion pattern is obtained by differentiating this moiré constant displacement line pattern.

Even in the automatic analysis of this moiré constant displacement line pattern, manual analysis is extremely troublesome, and automation is required. However, for automation, it is necessary for humans to determine the sign of deformation (compression or tension) and inform the automatic analysis device, and fully automatic analysis devices have not yet been put into practical use.

In this case as well, a method may be considered in which the subject of analysis processing is replaced with a displacement pattern such as moiré, and the deformed grid itself is processed as basic input data. These operations for determining the slope of the surface of a three-dimensional object or the distortion of a two-dimensional object all involve differential operations that are sensitive to noise, and in order to obtain reliable data, a large number of analysis points and spatial averaging are required. required a procedure.

An object of the present invention is to perform a fully automatic analysis of the deformed lattice image as described above, and to directly determine the slope of the surface of a three-dimensional object or the distortion of a two-dimensional object. This objective is achieved by detecting the deformed grating image with a group of photodetecting elements arranged in an annular manner and measuring the peak position and interval of the detected signal.

The principle of the present invention will be explained in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic diagram showing an example of a typical apparatus for obtaining a deformed lattice image by projection using the moiré topography method.

In FIG. 1, S is a light source which projects an image of a projection reference grating G through a condenser lens onto the surface of an object to be measured by a projection lens _L1 .

When the projected lattice image is observed from a direction different from the projection optical axis, a deformed lattice image that conforms to the object surface shape is obtained.

The imaging lens _L2 forms the deformed lattice image on a suitable imaging surface, such as a screen, focusing glass, or photosensitive material.

This deformed lattice image is transmitted to a TV camera T for monitoring and a detection unit 1 by mirrors M ₁ , M ₂ and a half mirror B.
is brought to a predetermined imaging plane. Note that the detection unit 1 is moved to a predetermined position by a drive mechanism 2. For example, a second
A deformed lattice image as shown in the figure is formed.

According to the present invention, the slope (reciprocal of the pitch) at an arbitrary location and the direction in which the slope is maximum are determined from the spacing (pitch) of the grid lines of this deformed grid image, and the magnitude of the slope thus determined is calculated. The shape of the surface of a three-dimensional object or the distortion of a two-dimensional object can be automatically measured based on the gradient having a direction.

FIG. 3 shows an example of a group of photodetecting elements (for example, N=0 to N=35) arranged in an annular shape installed in the detection unit 1 of FIG. 1 according to the present invention. FIG. 4 shows an example of the density distribution of the deformed lattice images detected from the group. In FIG. 3, the local pitch of the deformed lattice images is indicated by P, and the intervals of the deformed lattice images are respectively indicated by P ₁ to P ₁₀ . The detection output of the photodetector appears strong at positions close to the deformed grid lines, and weakly appears at positions away from the grid lines. Therefore, the photodetector directly above the grid lines will detect the maximum density, and the area between the images will show the weakest density, and the peaks of the output signal will be spaced apart, as shown in Figure 4. It appears after P′ ₁ to P′ ₁₀ . If we explain the part P' ₃ and P' ₈ , which have the narrowest interval among the peak intervals P' ₁ to P' ₁₀ , the maximum slope in this part of the deformed lattice image is 1/P. The direction is shown by the broken line. In this way, the magnitude and direction of the maximum slope are determined for each adjacent grid line. To specifically explain the direction determination of the gradient in the above part, the minimum peak spacing P' ₃ and P' ₈ in FIG. 4 correspond to the position of the photodetector in FIG. and determine N=25 and 27. Therefore, a straight line is drawn between the photodetectors N=9 and N=26 where the peak interval is the minimum, and the direction perpendicular to this straight line is defined as the direction of the gradient. The direction of this gradient does not indicate an absolute upward or downward direction.

In order to know the magnitude and direction of the gradient over the entire surface of the object, it is sufficient to move the detection element group on the observation plane and detect it one by one.

Figure 5a shows the gradient and direction of a cross section of the object at each point on the observation plane detected by the above method in vector form, and Figures b to d show the cross section of the object from this vector data. It shows how to find the shape. Generally, if there is little change in shape in the direction perpendicular to the cross section to be measured, the direction of the vector obtained by the present invention will be approximately constant as shown in FIG. The length of each vector corresponds to the magnitude of the gradient of the object, but since this gradient is based on the grid projection direction, a bias component (dashed line in FIG. 3A) is superimposed on the gradient. In order to obtain the gradient distribution of the surface of the object in the cross-section to be measured, it is sufficient to extract the components of these vectors in the direction of the cross-section to be measured, and plot the vector length on the vertical axis as shown in FIG.
Since a bias component is also added to this gradient distribution curve, by subtracting this, a gradient distribution from which a certain gradient component has been removed is obtained as shown in FIG. This shows the tilt amount distribution in the direction orthogonal to the observation direction, and by integrating this, the cross-sectional shape shown in the figure d is obtained.

FIG. 6 is an example in which this cross-sectional shape is obtained over the entire surface of the object and displayed in a perspective view.

The method of the present invention for measuring distortion of a two-dimensional object will be explained with reference to FIG. A lattice is fixed to the object surface so that the lattice deforms as the two-dimensional object is distorted. FIG. 7 shows an example of a deformed grating obtained by deforming a reference grating printed on the surface of an object. When this is detected by the aforementioned group of photodetectors arranged in an annular shape, a detection signal of the density distribution of the deformed grating as shown in FIG. 8 is obtained. By reading the minimum peak interval P of this signal, a numerical value proportional to the amount of distortion of the object can be obtained, and furthermore, the position of the minimum peak interval is the direction of distortion.

In order to determine the slope of a three-dimensional object or the distortion of a two-dimensional object, conventional methods require the three-dimensional shape distribution and deformation distribution of the object and numerically differentiate this data. Distortion of two-dimensional objects can be automatically detected. In the conventional method, because a differential operation is performed, it is susceptible to the influence of noise, and in order to obtain reliable data, it was necessary to determine the shape or amount of deformation for an extremely large number of data points, but these difficulties have been overcome by the present invention. Ta.

[Brief explanation of drawings]

Fig. 1 is a schematic diagram showing an example of a device for obtaining a deformed lattice image using the moire topography method, Fig. 2 is a plan view showing an example of a deformed lattice image, and Fig. 3 is a photodetector arranged in an annular shape. FIG. 4, a plan view showing an example of the element group, shows an example of the concentration distribution of a deformed lattice image detected by the photodetecting element group. Figure 5a is a plan view showing the magnitude and direction of the slope of one cross section of the object at each point on the observation plane detected by the photodetection element group, and Figures 5b to 5d are the vector data of figure a. FIG. 6 is an explanatory diagram for determining the cross-sectional shape of an object from the object. FIG. 6 shows an example in which the cross-sectional shape is determined over the entire surface of the object and is displayed in a perspective view. Fig. 7 is a plan view showing an example of a deformed grating obtained by deforming a reference grating printed on the surface of an object, and Fig. 8 is a plan view showing an example of a deformed grating obtained by deforming a reference grating printed on the surface of an object. 3 is a graph showing an example of a signal detected as a result. Symbols in the figure: S...light source, G...reference grating,
_L1 , _L2 ...lens, T...TV camera, P...interval, 1...detection section.

Claims (1)

[Claims] 1. A grating image projected onto a three-dimensional object and deformed according to the shape of the object surface is detected by a group of photodetecting elements arranged in an annular manner, and the maximum A method of measuring the slope of a three-dimensional object surface by determining the magnitude of the tilt and the direction of the maximum tilt. 2 A grating image that is fixed on a two-dimensional object and deforms as the object deforms is detected by a group of photodetecting elements arranged in an annular manner, and the magnitude of the maximum inclination and the maximum A method to measure the distortion of a two-dimensional object by determining the magnitude and direction of its inclination.