JPS6355642B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a method for automatically detecting and tracking the ridgeline of a three-dimensional object in a non-contact manner, and measuring the coordinates of the ridgeline in a non-contact manner. The present invention relates to a detection device and an edge coordinate measuring device.

As a standard for inspection and production at the production stage of certain structures, machine parts, molds (press molds, casting molds), models, etc., the three-dimensional coordinate values of the edges where multiple surfaces of those objects intersect, that is, the ridgelines, are It is often used as data representing the shape. For example, in the manufacturing stage of models and molds for the bodies of automobiles, etc., the shape of the surface is generally a free-form surface, and three-dimensional coordinate values of ridgelines are used as the manufacturing standard and inspection standard for the surface in most cases. Furthermore, especially in the case of external models and molds such as bodies, most of the ridge lines act as characteristic lines after completion and affect the appearance, so the ridge line coordinate values must be measured and inspected as particularly important data. The measurement of general surfaces and free-form surfaces of these three-dimensional objects has already been automated using three-dimensional coordinate measuring machines, but the measurement of the coordinate values of the ridgeline is done by visually aligning the measuring needle with the apex of the ridgeline. The number of man-hours required for edge line measurement is extremely large in processes where machining and surface measurement are automated, and the development of an automatic measurement method is desired due to the variation in measurement accuracy. There is.

BACKGROUND ART Conventionally, as a method for measuring the position of an edge line, a method using deflection scanning of a laser beam for the purpose of tracing a weld line or the like is known. This conventional method uses a mirror attached to a vibrating element such as a tuning fork as an optical deflector, and does not use a projecting lens. For this reason, the range for detecting the position of edges, etc. is limited by the amplitude of the vibrating element, and since the light is deflected in a fan-shaped manner, the sensitivity for detecting the deviation between the projection optical axis and the edge varies depending on the distance to the target object. The disadvantage was that it was difficult to make high-precision measurements and automatic tracking of steep ridge steps. In addition, this conventional method does not have a function to measure the distance to the target and automatically compensate, and therefore the distance at which the ridgeline position can be effectively tracked depends on the width of the detection slit and the angle between the receiving axis and the emitting axis. It can be decided. Limited to a very narrow range within a certain distance. In order to expand this range, this conventional method reduces the angle between the light emitting axis and the light receiving axis, but as a result, the accuracy of measuring the positional deviation of the ridge line decreases, and it is difficult to There were drawbacks that made it difficult to apply.

In addition, as mentioned above, this conventional method does not have the function to set the distance, so it requires offline artificial distance setting, and this measurement method alone cannot automatically track the ridgeline on any complex three-dimensional object. Moreover, it has been impossible to accurately measure the three-dimensional coordinate values of the ridgeline.

Another method is to measure the position and shape of a three-dimensional object in a non-contact manner by projecting a slit-shaped light beam onto the surface of the object, and then looking at the image of the light trajectory bent at the ridge line at a different angle. There is a method in which the ridgeline is determined by observing it with a television camera (storage-type imaging device) and performing triangulation.

In the case of such a method, it is necessary to take measures against "blurring" of the image due to changes in distance. That is, when an image is blurred and spans several pixels, some image processing such as narrowing processing is required to accurately determine the image position, which tends to slow down the processing time.

Furthermore, when such an accumulation type imaging device is used, afterimages become a problem. That is, even if high-speed measurement is attempted, the accuracy will not improve due to the afterimage of the previous image, and the speed will eventually have to be sacrificed to some extent. Furthermore, in order to measure the entire shape of the object to be measured, it is necessary to move either the object or the measuring device, but it is often difficult to move structures, press molds, and clay models for automobile exterior modeling. , it would be necessary to move the Ikioi measuring device. However, when using an image pickup tube,
The shape and dimensions of the measurement device are large and heavy, and durability, vibration resistance, and stability are issues, and in addition to the aforementioned slow speed, it is difficult to use it as a measurement probe for 3D measuring machines and other online measurements. The drawback was that it was difficult to do so.

The present invention solves the drawbacks of the prior art as described above, and provides an automatic tracking measurement device that can continuously and easily measure the three-dimensional coordinate values of the ridge lines of an object with a complex and varied three-dimensional shape. The purpose is to

The configuration of the present invention is roughly as follows.

First, the edge apex detection device for automatic measurement of edge coordinates according to the present invention includes: a deflected thin light beam projecting means for deflecting and scanning a thin light beam with a constant width and projecting the deflected thin beam onto an area including the ridgeline of the object to be measured; a light receiving means that focuses the reflected light from the object to be measured to form an image at a predetermined angle that is perpendicular to the direction of deflection; an image forming position vibrating means for vibrating in the direction; and determining whether or not the image forming light of the light receiving means is within a region having a predetermined length in the direction of deflection of the thin light beam and a minute width in the vertical direction. a detection system head having a light spot position detecting means for detection; and sampling the deflection signal of the deflected thin beam projecting means and the vibration signal of the imaging light oscillating means using a signal from the light spot position detecting means as a synchronization signal. The present invention is characterized by comprising an edge line coordinate calculating means for calculating two-dimensional coordinate values of the edge section line by calculating the two-dimensional coordinate values, and calculating a second-order difference of each of the two-dimensional coordinate values to calculate the edge apex coordinates.

Furthermore, the automatic edge line coordinate measuring device according to the present invention includes: a three-dimensional drive mechanism that moves the detection system head in the direction in which the edge line extends; and a distance measuring means that measures the amount of movement of the detection system head by the three-dimensional drive mechanism. , absolute coordinate system edge coordinate calculation means for calculating edge line coordinates in an absolute coordinate system from the output of the edge line coordinate calculation means and the output of the distance measuring means; and recording means for recording the edge line coordinates in the absolute coordinate system. It is characterized by the fact that it is equipped with Furthermore, from the trend of changes in the ridge apex coordinates measured previously,
A main control device that calculates the positional relationship between the direction in which the ridgeline extends and the detection system head, controls the movement of the three-dimensional drive mechanism and the two-axis rotation mechanism, and causes the detection system head to follow the ridgeline. It is characterized by

According to an embodiment of the present invention, the sensor coordinate system edge coordinate calculating means is configured to calculate the deflection signal of the deflected narrow beam projecting means and the imaging position vibrating means based on the synchronization signal of the light spot position detecting means. An X-axis sampling circuit and a Y-axis sampling circuit that sample and accumulate vibration signals, an inflection point detection circuit that detects an inflection point of an output signal of the Y-axis sampling circuit, and an output synchronization signal of the inflection point detection circuit. The apparatus further includes an X-coordinate sampling circuit and a Y-coordinate sampling circuit that sample and store output signals of the X-axis sampling circuit and the Y-axis sampling circuit.

Further, the three-dimensional drive mechanism includes an Y-axis servo means for changing the distance between the detection optical system head and the object to be measured so that the output signal of the coordinate sampling circuit is within a certain range; and Z-axis servo means for moving at a predetermined speed in a direction perpendicular to the direction.

Further, the absolute coordinate system ridgeline coordinate calculation means adds a correction to the movement amount of the detection optical system head by the X-axis servo means by an output signal value of the X-coordinate sampling circuit to obtain the X-axis coordinate, The amount of movement of the detection optical system by the servo means is
A correction for the output signal value of the coordinate sampling circuit is added to make the Y-axis coordinate, and the amount of movement of the detection optical system head by the Z-axis servo means is taken as the Z-axis coordinate,
Calculate the edge coordinates in the absolute coordinate system.

Embodiments will be described in detail below with reference to the drawings.

FIG. 1 shows an embodiment of the present invention, in which numeral 1 indicates a thin light beam source with good directivity, such as a laser. The light deflection unit 2a and the oscillation unit 2b deflect and scan the light in the x-axis direction (hereinafter simply referred to as x-scanning) x
It constitutes an axial light deflector 2. Reference numeral 3 denotes a projection lens arranged with its focus aligned with the deflection center of the x-axis optical deflector 2, 4 an object to be measured having a ridge, and 5 an optical axis of the projection lens 3 (hereinafter referred to as projection optical axis). y axis) and 0~
This is a light receiving lens that has a constant angle within 90 degrees and has an optical axis (hereinafter referred to as a light receiving axis) that is perpendicular to the x scanning direction, and that focuses reflected light from the object to be measured 4 and forms an image.

A y-axis optical deflector 6 is constituted by the optical deflection sections 6a and 6b and the oscillation section 6c, and this y-axis optical deflector is arranged on the optical receiving axis between the object to be measured 4 and the imaging plane formed by the optical receiving lens 5. It deflects and scans the reflected light from the object to be measured 4 in a direction orthogonal to the x-scanning direction (hereinafter simply referred to as y-scanning).
A light spot position detector 7 is constituted by an optical slit 7a having minute intervals 7d parallel to the x-scanning direction, a photodetector 7b, and a differential processing circuit 7c. A predetermined trigger signal is generated the moment the point passes.

8 is an x-axis sampling circuit that samples and stores the x-scanning signal of the x-axis optical deflector 2 at the moment when the trigger signal of the light spot position detector 7 is generated; 10 is a y-axis sampling circuit that samples and stores the y-scanning signal of the y-axis optical deflector 6 at the moment of occurrence;
This is an inflection point detection circuit that detects an inflection point of the output signal of the axis sampling circuit 9 and generates a predetermined trigger signal at that moment.

11 is an x-coordinate sampling circuit that samples and stores the signal of the x-axis sampling circuit 8 at the moment when the trigger signal of the inflection point detection circuit 10 is generated;
Reference numeral 12 denotes a y-coordinate sampling circuit that samples and stores the signal of the y-axis sampling circuit 9 at the moment when the trigger signal of the inflection point detection circuit 10 is generated.
3 is an x-axis comparator that generates a signal when the output signal value of the x-coordinate sampling circuit 11 deviates from a certain range, and 14 is a y-axis comparator that generates a signal when the output signal value of the y-coordinate sampling circuit 12 deviates from a certain range.
It is an axis comparator.

Reference numeral 15 is a detector which is operated by a signal from the x-axis comparator 13 and includes the light source 1, the x-axis optical deflector 2, the light projecting lens 3, the light receiving lens 5, the y-axis optical deflector 6, and the light spot position detector 7. A three-dimensional drive mechanism 24 that has the function of moving the optical system head 23 in three dimensions of X, Y, and Z.
Of the three axes, 16 is an X-axis movement mechanism that drives the x scanning direction, that is, the x-axis direction in the example of FIG. In the example, a Y-axis moving mechanism drives in the y-axis direction, and 17 is a Z-axis moving mechanism that operates upon receiving instructions from the main controller 18.

Reference numeral 19 is attached to the X-axis moving mechanism 15 and measures the amount of movement of the detection optical system head 23 in the x-axis direction. Reference numeral 20 is attached to the Y-axis moving mechanism 16 and measures the amount of movement of the detection optical system head 23 in the y-axis direction. A Y-axis distance measuring device 21 for measuring the amount of movement in the axial direction is attached to the Z-axis servo mechanism 17, and the detection optical system head 2
This is a Z-axis distance measuring device that measures the amount of movement in the Z-axis direction of No. 3.

18 is the main control device of the measuring device according to the present invention; 2
2 is a display/recording device, and 25 is a trajectory of the x-scanning light on the object 4 to be measured.

Next, the operating principle of this embodiment will be explained.

First, the operating principle of the cross-sectional shape measurement function of the detection optical system head will be explained.

The narrow light source 1 is a light source that emits parallel light with good directionality and little spread, and may be a combination of a point light source and a lens, but typically uses a gas laser, semiconductor laser, etc. The light enters the x-axis optical deflector 2 and is deflected and scanned at a repetition frequency _{of 1} by a predetermined angle in one axis direction perpendicular to the projection optical axis, that is, in the x-axis direction when the projection optical axis is the y-axis. As the x-axis optical deflector 2, a device using a well-known electro-optic effect, a tuning fork vibrating mirror, a rotating polygon mirror, etc. can be used, but FIG. 1 shows an example in which the galvanometer mirror 2a is driven by an external It is shown.

The deflection-scanned light passes through a projection lens 3 whose focal point O′ coincides with the center of deflection of the x-axis optical deflector 2.
The ridge line of the object to be measured 4 is cut at a right angle and projected so as to scan the vicinity of the ridge line. With this configuration, the projection lens 3 transforms the polarized light that is diffused fan-shaped from the deflection center into a beam array that performs band-shaped polarization scanning parallel to the projection optical axis, and the light spot position in the x-axis direction is independent of the distance. can be uniquely known from the deflection angle of the x-axis optical deflector 2, and by converging parallel light beams, the diameter of the light spot on the object to be measured 4 can be made minute, thereby increasing the detection resolution. As an example, if the deflection angle of the optical deflector 2 is 0.1 rad and the focal length of the projection lens 3 is 100 mm, the deflection scanning width will be 10 mm regardless of the distance, and the focal point O of the projection lens 3 on the side opposite to the optical deflector 2 The diameter of the light spot on nearby objects is approximately 0.1 mm.

The reflected light from the object to be measured 4 is on the yz plane in FIG. ) and is focused by a light receiving lens 5 having an optical axis passing through ). The value of θ is typically on the order of π/6 to π/4 rad.

The light that has passed through the light receiving lens 3 is directed to the y-axis optical deflector 6
It is reflected by the deflection mirror 6a, and an image is formed near the imaging point 0'' by the light receiving lens 3 at the origin O.

As the y-axis optical deflector 6, several types can be used like the x-axis optical deflector 2, but in the example shown in FIG. An example of driving with The y-axis optical deflector 6 has its deflection center at the origin O,
It is on the straight line connecting the imaging point O'', and the deflection axis is arranged parallel to the x scanning direction.The position is between the origin O and the imaging point O'', in front of the light receiving lens (deflection on the object space side). In the example shown in FIG. 1, it is placed at a predetermined position between the light-receiving lens 3 and the image-forming point O'', although the rear (image space side deflection) does not matter.

With such a configuration, the y-axis optical deflector 6 can be
When the axial light deflector 2 is driven at a repetition frequency ₂ that is sufficiently higher than the driving frequency ₁ , the imaging light reflected by the object to be measured 4 and focused by the light receiving lens 5 moves the light receiving axis in a direction parallel to the y-axis. It is deflected around the center (hereinafter referred to as y-scanning).

The light deflected by the y-axis optical deflector 6 includes an image O'' formed by the light receiving lens 3 at the origin O, and a detection surface 7a of the light spot position detector 7, which has a detection surface perpendicular to the light receiving axis.
It forms an image as a light spot above.

The light spot position detector 7 has one axis that is parallel to the x-axis and perpendicular to the light-receiving axis that includes the image formation O'' (hereinafter referred to as x _i
axis), there are two areas divided by this axis, in the example in Figure 1 +y _i area and -y _i area, and there is a light spot on the x _i axis,
Alternatively, in the case of a spreading light point, a predetermined signal is generated when the center of gravity of the light is present.
It may be a differential slit, differential photodiode, differential prism, etc. as shown in Figure 2.
In the illustrated example, an optical slit 7a having an aperture that simply coincides with the x _i axis, a photodetector 7b, and a differential circuit 7c are shown.
It is composed of These light spot position detection means are
There is no afterimage at all, and even if the light spot image is blurred, the center of gravity of the light amount can be detected accurately.

The function of light spot position detection in this embodiment will be explained with reference to FIG. Figure 3a shows the minute gap 7.
This figure shows how a light spot having a size moves on the optical slit 7a having a size d in the order of t ₀ →t ₁ →...→t ₆ . Correspondingly, as shown in FIG. 3b, at time _t3 when the center of the light spot coincides with the minute gap 7d, the amount of light passing through the minute gap 7d becomes maximum, and the output of the photodetector 7b also becomes maximum. FIG. 3c shows a signal waveform obtained by time-differentiating the output of the photodetector 7b,
It crosses 0VOLT at the time _t3 . The differential processing circuit 7c performs the differentiation and zero cross detection shown in FIG. 3c and outputs a signal, an example of which is shown in FIG. 3d. In the example of FIG. This shows that it coincides with the center of the minute gap 7d.

The signal generated from the light spot position detection device simultaneously activates the x-axis sampling circuit 8 and the y-axis sampling circuit 9 to sample the input. x
The x-deflection oscillator 2b output of the x-axis optical deflector 2 is input to the axis sampling circuit 8, and the output of the oscillator 6 of the y-axis optical deflector 6 is input to the y-axis sampling circuit 9.
The output of c is input.

With this configuration, the outputs of the x-axis sampling circuit 8 and the y-axis sampling circuit 9 are irradiated with the x-axis scanning light of the object to be measured 4 during one period of x-scanning by the signal of the light spot position detection device 7. The cross-sectional shape of the area is output. For example, if you connect the outputs of the x-axis sampling circuit 8 to the x-input of an x-y oscilloscope and the outputs of the y-axis sampling circuit 9 to the y-input, the x-y cross section of the object to be measured 4 with the origin O as the origin will be displayed. be done.

The functions up to this point (cross section measurement function) will be explained in more detail.

FIG. 4 shows the locus of the imaged light spot on the optical slit 7a. In FIG. 4, x _i and ±y _i are set as in FIG. 1. Due to x-scanning and y-scanning that is sufficiently faster than x-scanning, the imaged light spot starts from the black spot shown in FIG. 4, reaches the position of the white spot after half a period of x-scanning, and returns to the black spot by following the reverse path again. The above movement is repeated one after another, but the period of the y-scanning is sufficiently shorter than the period of the x-scanning, and as long as it is only about one period of the y-scanning, the imaged light spot will not move in the x _i- axis direction on the optical slit 7a. It's hardly moving.

The distance measuring function performed in this state will be explained below with reference to FIG.

To simplify the explanation, it is assumed that the x-axis optical deflector 2 is stopped and the light emitted from the light source 1 is projected onto the object to be measured 4 along a path that coincides with the projection optical axis, that is, the y-axis. Figure 5a shows the intersection point P of the object to be measured 4 with the y-axis.
indicates a state in which O coincides with the origin O. Naturally, the projection light point is also at the same position as P (hereinafter simply referred to as projection light point P). 5b and 5c show the state in which the object to be measured 4 is displaced from the state shown in FIG. 5a, respectively, in the direction of the detection optical system and in the opposite direction. Figure 5d,
e and f represent the positions of the projection light spot Q formed on the optical slit 7a, corresponding to FIGS. 5a, b, and c, respectively.

When the projected light point P and the origin O coincide as shown in Fig. 5d, the respective images Q and O'' also coincide,
It is located above the minute gap 7d of the optical slit 7a. Here, since the y-axis optical deflector 6 performs optical deflection scanning around the light receiving axis, the image Q actually moves repeatedly between Q' and Q'' at a frequency of ₂ . The Q shown in FIG. The time course of the signal that was optically converted by the photodetector 7b and passed through the differential processor 7c is shown in the lower row, with the output of the y-axis sampling circuit 9 activated by this signal aligned with the time axis origin. As is clear from the figure, when the origin O and the projected light point P coincide as shown in Figures 5a and d, Figure 5g
At the time when the deflection angle shown in the upper row is φ, the image Q of P passes through the optical slit 7a and produces an output effective as a signal of the light spot position detection device 7. A y-axis sampling circuit 9 activated by the signal of the light spot position detection device 7 has a y-deflection oscillator 6 connected to its input.
Sample and store the output of c. Since the output of the y-deflection oscillator 6c has a one-to-one correspondence with the deflection angle of the y-axis optical deflector, the output of the y-axis sampling circuit 9 is zero in this case.

In this state, the measured object 4 is displaced in the positive direction of the y-axis as shown in Fig. 5 b and e, or in the negative direction of the y-axis as shown in Fig. 5 c and f. In this case, as is clear from the figure, if the y-axis optical deflector 6 is not deflected by the amount that the image Q does not coincide with the image point O'' of the origin O, the image Q will be shifted to the optical slit 7.
It does not pass through the minute gap 7d of a. Therefore, as shown in FIG. 5h and FIG. 5i, the y-axis optical deflector 6
The image Q passes through the minute gap 7d of the optical slit 7a at a time when the deflection angle of A y-deflection oscillator 6 that has a one-to-one correspondence with the deflection angle of the y-axis optical deflector 6
Sample and output the non-zero output value of c.
That is, when the object to be measured 4 is displaced in the positive direction of the y-axis as shown in FIG. , the y-axis sampling circuit samples and stores the positive voltage. Figure 5 e,
In the case of displacement in the negative direction of the y-axis, such as f and i, the opposite will only occur, so the explanation will be omitted. This y
The output voltage of the axis sampling circuit can be easily converted into the deflection angle of the y-axis optical deflector 6 in order to match the image formation Q with the reduced gap 7d of the optical slit 7a.
It is clear that this can be converted into the distance in the y-axis direction between the point of intersection P with the y-axis on the surface of the object to be measured 4 and the origin O, or the y-coordinate. Also,
The distance OH between the origin O and the detection optical system head 23, for example, the distance OH from the center H of the projection lens 3, is physically fixed, and the output voltage of the y-axis sampling circuit 9 is determined by the distance OH on the object to be measured 4. It is clear that the distance between the point P and the detection optical system head 23 can also be easily converted.

As described above, the distance between the detection optical system head 23 and the object to be measured 4 can be determined in the minimum half period of the y scan. From this state, if the x-axis optical deflector 2 is driven at a repetition frequency ₁ that is sufficiently slower than the y _- axis optical deflector 6 as shown in FIG. One cycle of y-scanning is completed during a slight movement, and at time t when the image Q passes through the gap 7d, a sampling pulse is generated from the light spot position detection device 7, and the y-axis sampling circuit 9 receives a detection pulse at that time t. A voltage related to the distance between the optical system head 23 and the object to be measured 4 is maintained.

Also, the sampling pulse of the light spot detection device 7 is x
The axis sampling circuit 8 is also activated, and the output voltage of the x-deflection oscillator 2b of the y-axis optical deflector 2 at that time t is sampled and accumulated. Similarly to the case of the y-axis sampling circuit 6, the output voltage of the x-axis sampling circuit 2 can also be easily converted into the deflection angle of the x-axis optical deflector 2, and as a result, the projected light spot P on the object to be measured 4 can be It is clear that the distance from the y-axis, which is the projection optical axis, can be converted into the x-coordinate.

In the above manner, as shown in FIG.
The origin is O of the trajectory of the projection light point P that moves over the object to be measured 4 in time series during the period, and the
The x-coordinate and y-coordinate with the direction of the projection optical axis as the y-axis are output for each period of y-scanning. That is, a cross-sectional shape obtained by cutting the object to be measured 4 along a plane formed by x-scanning is output. The above are details of the cross section measurement function.

Note that for the purpose of explanation in the above embodiments, an xy-axis oscillator, a sampling circuit, etc. were used, but a clock oscillator, counter, latch circuit, or
Like the A/D converter and latch circuit, it is activated by the sampling pulse of the light spot position detection device 7 and samples information corresponding to the deflection angle of each of the x-axis optical deflector 2 and the y-axis optical deflector 6. It goes without saying that it is good as long as it is something that can be accumulated. In this case, x coordinate, y
The coordinates are sequentially stored in a storage device as digital data, and subsequent processing is performed by a digital computer.

Next, the function of measuring the coordinates of the ridge apex using the above cross-sectional shape output will be explained with reference to FIG.

FIG. 6a is a diagram showing the time course of the deflection angle of the x-axis optical deflector 2, and simply shows the case of driving by a triangular waveform. In this case, since the polarization angle of the x-axis optical deflector 2 is proportional to time, the differential operation with respect to x can be conveniently replaced with time differential.
T represents the period of the half cycle. Figure 6 b, c
6A and 6B are respectively drawn with the time origins of the outputs of the x-axis sampling circuit 8 and the y-axis sampling circuit 9 aligned with those of FIG. 6a, and the following figures d to g are also aligned in the same manner. As described above, the output values at the corresponding times in FIGS. 6b and 6c represent the x and y coordinates in the cross-sectional shape of the object to be measured 4. In the arrangement shown in FIG. 1, the position of the ridge apex is taken as a point where the rate of change in the distance between the detection optical system head 23 and the object to be measured 4 shows a rapid change, and as is clear from FIG. It is easy to see that the point in time when the second-order differential value of the sampling circuit 9 exceeds a significant value indicates the position of the ridge apex. The inflection point detection circuit 10 includes a second-order differential circuit, a zero-cross detection circuit, and a gate circuit, and the outputs of each are shown in FIGS. 6d, e, and f. FIG. 6d shows the time differential waveform of the output of the y-axis sampling circuit 9 of FIG. 6c, and FIG. 6e shows the pulse waveform whose zero-crossing point was detected. If this continues, meaningless pulses will be generated at the top dead center and bottom dead center of the x-axis optical deflection.
The signal is passed through a gate circuit that generates a blank pulse shown in FIG.

FIG. 6g shows the output of the inflection point detection circuit 10. The output pulse of the inflection point detection circuit 10 activated the x-coordinate sampling circuit 11 and the y-coordinate sampling circuit 12, and was connected to their respective inputs. The values of the outputs of the x-axis sampling circuit 8 and the y-axis sampling circuit 9 at the moment when the output pulse of the inflection point detection circuit 10 is generated are sampled and stored. x
It is clear from the time correspondence in FIG. 6 that the values sampled by the coordinate sampling circuit 11 and the y-coordinate sampling circuit 12 are information regarding the x-coordinate and y-coordinate of the edge apex of the object to be measured 4.

As described above, the coordinates of the ridge apex of the object to be measured 4 on which the light is projected are measured every half cycle of the x-scan.

It is clear that the above explanation holds true for both convex and concave edges. In addition, although the above method for detecting edge vertices uses a differential processing circuit, a sampling circuit, etc. for the purpose of explaining the embodiment, generally, information regarding the y-coordinate is subjected to differential processing with respect to information regarding the x-coordinate. Of course, any method that extracts and outputs information regarding the x-coordinate and y-coordinate of a position that indicates a significant change in output is sufficient.

As another embodiment, FIG. 7 shows an example in which digital data is detected from x and y coordinates stored in a memory device. FIG. 7a shows an example of cross-sectional shape measurement in which the contents stored in the x-axis and y-axis digital memory devices corresponding to the x-axis sampling circuit 8 and the y-axis sampling circuit 9 are plotted with circles on the xy coordinates. express. FIG. 7b shows an example of calculation of the differential value and the second-order differential value. Inflection point detection circuit 10, x coordinate sampling circuit 1
1. The function corresponding to the y-coordinate sampling circuit 12 is realized as follows.

1 point 1 point 2 stored in digital memory
Dimensional coordinate data is (x _i , y _i ) (i=1,..., m)
shall be.

Now perform the following operation on this array.

f″={y _i+1 +y _i ― ₁ ―2y _i }／{x _i+1 ―x _i } ² Divide all data into regions from the point where this second-order differential value exceeds a certain predetermined threshold. , by finding the approximate straight line l ₁ and approximate straight line l ₂ that pass through the data points included in each using the method of least squares, etc., and output the coordinates of the intersection C ₁ of the two straight lines l ₁ and l ₂ as the edge apex. The above function is then realized.

In another case, the actual edge may be rounded at the apex. In such a case, add the following calculation. In addition to the above-mentioned areas and regions, an area where the second-order differential value f'' exceeds a predetermined threshold is set as a curved area, and an approximate biline (or approximate parabola) passing through the data points included in that area is set. Find the equal l ₃ using the method of least squares, etc. A straight line l passes through the intersection C _{1 of this approximated curve l 3 and} the two straight lines l ₁ and l ₂ , and bisects the angle formed by the straight lines l ₁ and l ₂ . By outputting the coordinates of the intersection point C ₂ with ₄ as the edge apex, the above-mentioned function of correcting the roundness of the edge apex portion is realized.

Next, the edge line measurement operation will be explained using FIGS. 1 and 8.

First, an edge line to be measured on the object to be measured 4 is selected, and the approximate coordinates of the measurement starting point and the measurement direction are stored in the main controller 18 of the three-dimensional drive mechanism 24 that moves the detection optical system head 23 three-dimensionally. The main controller 18, as shown in FIG.
4, the detection optical system head 23 is moved to the measurement point, and measurement is started. Detection optical system head 23
performs the above-mentioned cross-sectional line measurement and edge apex calculation, and outputs the edge apex coordinates x ₀ , y ₀ in the coordinate system (hereinafter referred to as sensor coordinate system) fixed to the detection optical system head 23 .
The main controller 18 includes an X-axis distance measuring device 19 and a Y-axis distance measuring device 2 attached to the three-dimensional drive mechanism 24.
The position of the detection optical system head 23, for example, the coordinates of the principal point H of the floodlight lens in FIG. Input the edge apex coordinates x ₀ , y ₀ in the coordinate system, and finally convert the coordinates to the position in the coordinate system fixed to the three-dimensional drive mechanism (hereinafter referred to as the absolute coordinate system) to obtain the edge apex coordinates X ₀ in the absolute coordinate system. , Y ₀ , Z ₀ to the display/recording device 22,
Displayed and recorded. This coordinate transformation can be performed in any coordinate system,
For example, it goes without saying that a coordinate system fixed to the object to be measured 4 (hereinafter referred to as a workpiece coordinate system) may be used. Next, the main controller 18
That is, the detection optical system head 23 is moved in the direction in which the ridgeline extends at a fixed pitch, at a constant speed, or at programmed intervals, and the measurement of the ridge apex coordinates is repeated again. Here, if the edge apex P deviates significantly from the origin O of the sensor coordinate system, for example, if it deviates greatly in the positive or negative direction of the x-axis, the output of the x-coordinate sampling circuit 11 will be a large positive or negative value. . This output is input to the x-axis comparator. The x-axis comparator has the function of a window comparator, and has a predetermined positive and negative threshold around φ, and when the input is within this range, the output is φ, and when it exceeds the negative threshold, it is a positive predetermined value. If the positive threshold is exceeded, a predetermined negative value is output for a certain period of time. The output of the x-axis comparator 13 is input to an The edge apex P is located near the origin O of the sensor coordinate system. Even when there is a deviation in the y-axis direction, the y-coordinate sampling circuit 12, y-axis comparator 14, and Y-axis drive mechanism 16 operate in the same way, moving the detection optical system head 23 to bring the ridge apex P to the origin O of the sensor coordinate system. be located near.

As is clear from the above description, the device according to the present invention does not require sophisticated servo operation and can track the edge by adjusting the position of the detection optical system head 23 step by step. This system is characterized in that sufficient speed can be achieved without using a double servo loop of a parent servo and a child servo. In the above description, for the purpose of explaining the embodiment, the x-axis comparator 13 and the y-axis comparator 14 were treated as independent, but the main controller 18 inputs information regarding the edge apex coordinates in a straight line, and It goes without saying that the comparison function described above may be realized by the judgment logic and the X-axis drive mechanism 15 and the Y-axis drive mechanism 16 may be directly controlled. By repeating the above operations, the detection optical system head 23 follows the ridgeline and moves in the direction in which the ridgeline extends. The main controller 8 at the appropriate time
9, Y-axis distance measuring device 20, Z-axis distance measuring device 21,
From the detection optical system head 23, the edge apex coordinates x ₀ , y ₀ in the sensor coordinate system are input, and the edge apex coordinates X ₀ , Y ₀ , in the absolute coordinate system are input.
Z ₀ is calculated and stored, and the data is transferred to an appropriate display or recording device 22 at each Z-axis measurement interval artificially instructed by a program or the like, and is recorded as a three-dimensional coordinate value of the ridgeline.

At the same time, the main controller 18 realizes that the direction in which the ridge line actually extends does not coincide with the Z-axis of the three-dimensional drive mechanism 24 as described above, and for example, In this case, the detection optical system head 23 may be moved in the direction in which the ridge line extends by controlling the drive shaft mechanism in an appropriate direction at the same time.

In addition, the main controller 18 can be used when the x-scan is not projected as if the ridgeline is cut at a right angle, or when the projection light axis and the normal to the ridgeline are tilted so much that the projected light no longer illuminates one side of the ridgeline. The condition can be determined from the tendency of change in the coordinates of the ridge apex measured previously, and the two-axis rotation mechanism 26 equipped with the detection head 23 can be driven to repeat the above-mentioned function again.

As described above, the present invention enables non-contact automatic measurement of the three-dimensional coordinate values of an arbitrary three-dimensional object by detecting the edge line of an arbitrary three-dimensional object and automatically tracking the edge line.

[Brief explanation of drawings]

Fig. 1 shows an embodiment of the present invention, and Fig. 2 shows an example of a light spot position detector 7, in which a shows a differential slit, b shows a differential photodiode, and c shows a differential prism. It is. FIG. 3 is a diagram for explaining the function of detecting the position of a light spot, and a in the same figure shows the optical slit 7a.
A diagram showing the movement of the light spot above, where b is the photodetector 7b.
Figure c is a diagram showing a waveform obtained by time-differentiating the output of the photodetector 7b, and Figure d is a diagram showing the output of the differential processing circuit 7c. FIG. 4 shows the locus of the imaged light spot on the optical slit 7a.
Figure 5 a, b, c, d, e, and f are diagrams showing changes in the position of the projected light point Q depending on the position of the object to be measured, and Figure 5 g, h, and i are diagrams corresponding to the above a, b, and c, respectively. The deflection angle of the corresponding y-axis optical deflector, the output of the light spot position detection device, the output of the y-axis sampling circuit, etc. are shown, respectively. In Fig. 6, a is the deflection angle of the x-axis deflector 2, b is the output of the x-axis sampling circuit 8, and c is the y
The output of the axis sampling circuit 9, d is the time differential of the above c, e is the output of the zero cross circuit, f is the output of the gate circuit, and g is the output of the inflection point detection circuit 10, respectively. Figure 7a shows the x-axis and y
The contents stored in the digital memory of the axis are
Shape data plotted on xy coordinates, Figure b shows the differential and second-order differential values of the data in Figure a, respectively. FIG. 8 shows a three-dimensional drive mechanism for moving the detection system head in the direction in which the ridgeline extends. DESCRIPTION OF SYMBOLS 1...Narrow light source, 2...x-axis optical deflector, 3...light projecting lens, 4...object to be measured, 5...light receiving lens, 6...y
Axial light deflector, 7... Light spot position detector, 7a... Optical slit, 7b... Photodetector, 7c... Differential processing circuit,
8...x-axis sampling circuit, 9...y-axis sampling circuit, 10...inflection point detection circuit, 11...x-coordinate sampling circuit, 12...y-coordinate sampling circuit, 13...x-axis comparator, 14...y-axis comparator, 15
...X-axis movement mechanism, 16...Y-axis movement mechanism, 17...Z
Axis movement mechanism, 18... Main control device, 19... X-axis distance measuring device, 20... Y-axis distance measuring device, 21... Z-axis distance measuring device, 22... Display recording device, 23... Detection optical system head, 24... 3 Dimensional drive mechanism, 25... Trajectory of scanning light, 26... Two-axis rotation mechanism.

Claims (1)

[Scope of Claims] 1. A deflected thin beam projection means for deflecting and scanning a thin beam with a constant width and projecting the deflected beam onto an area including the ridge line of the object to be measured; a light receiving means that focuses reflected light from the object to be measured at an angle to form an image; and an imaging position vibrating means that vibrates the image formed by the light receiving means in a direction perpendicular to the direction of deflection of the narrow beam. , light spot position detection means for detecting whether or not the imaged light of the light receiving means is present in a region having a predetermined length in the deflection direction of the thin light beam and having a minute width in the vertical direction. A detection optical system head and a signal from the light spot position detection means are used as synchronization signals to sample the deflection signal of the deflection thin beam projection means and the vibration signal of the imaging light vibration means, thereby obtaining a two-dimensional image of the ridge cross-sectional line. Find the coordinate values, part 2
An edge apex detection device for automatically measuring edge coordinates, comprising: an edge coordinate calculation means for calculating edge apex coordinates by calculating a second-order difference of each dimensional coordinate value. 2. A deflected thin light beam projection means that deflects and scans a thin light beam with a constant width and projects it onto an area including the ridge line of the object to be measured; a light receiving device that focuses reflected light from an object to form an image; an imaging position vibrating device that vibrates the image formed by the light receiving device in a direction perpendicular to the deflection direction of the thin beam; and deflection of the thin beam. a detection optical system head having a light spot position detection means for detecting whether or not the imaged light of the light receiving means is present in a region having a predetermined length in the direction and a minute width in the vertical direction; A two-dimensional coordinate value of the ridge section line is obtained by sampling the deflection signal of the deflected narrow beam projection means and the vibration signal of the imaging light vibration means using the signal from the light spot position detection means as a synchronization signal. 2
a sensor coordinate system edge coordinate calculation means for calculating the edge apex coordinates by calculating the second order difference of each dimensional coordinate value; a three-dimensional drive mechanism for moving the detection system head in the direction in which the edge line extends; and the three-dimensional drive mechanism. distance measuring means for measuring the amount of movement of the head of the detection optical system according to the sensor coordinate system; An automatic edge coordinate measuring device comprising: a calculation means; and a recording means for recording edge coordinates in the absolute coordinate system. 3. A deflected thin beam projection means that deflects and scans a thin beam with a constant width and projects it onto an area including the ridgeline of the object to be measured; a light receiving device that focuses reflected light from an object to form an image; an imaging position vibrating device that vibrates the image formed by the light receiving device in a direction perpendicular to the deflection direction of the thin beam; and deflection of the thin beam. a detection optical system head having a light spot position detection means for detecting whether or not the imaged light of the light receiving means is present in a region having a predetermined length in the direction and a minute width in the vertical direction; A two-dimensional coordinate value of the ridge section line is obtained by sampling the deflection signal of the deflected narrow beam projection means and the vibration signal of the imaging light vibration means using the signal from the light spot position detection means as a synchronization signal. 2
a sensor coordinate system edge line coordinate calculation means for calculating the edge apex coordinates by calculating the second order difference of each of the dimensional coordinate values; and for changing the attitude of the detection system head.
a two-axis rotation mechanism capable of rotating in horizontal and vertical planes; a three-dimensional drive mechanism that moves the detection system head in the direction in which the ridgeline extends; and a distance for measuring the amount of movement of the detection optical system head by the three-dimensional drive mechanism. a measuring means; an absolute coordinate system ridge coordinate calculating means for calculating ridge coordinates in an absolute coordinate system from a calculation result of the sensor coordinate system ridge coordinate calculating means and an output of the distance measuring means; Based on the recording means and the trend of change in the previously measured ridge apex coordinates,
A main control device that calculates the positional relationship between the direction in which the ridgeline extends and the detection system head, controls the movement of the three-dimensional drive mechanism and the two-axis rotation mechanism, and causes the detection system head to follow the ridgeline. An automatic edge coordinate measuring device characterized by: