JPH0629709B2 - Measuring method and device for three-dimensional curved surface shape - Google Patents

Description

TECHNICAL FIELD The present invention relates to a method and an apparatus for measuring the shape of a three-dimensional curved surface in a non-contact manner.

[Prior Art] Various methods have been proposed for the measurement of three-dimensional curved surface shapes because they can be applied to a wide range of fields such as three-dimensional CAD input, robot vision, or human body shape measurement for medical or clothing design. Has been done.

Among them, a method generally known as a light-section method is described in, for example, “Image Processing Handbook” (Shokoido Co., Ltd.), No. 398,
As described on page 399, as shown in FIG. 2, the slit light from the slit light source (52) with respect to the DUT (51) is measured.
The phenomenon that the light beam pattern formed on the object surface when irradiating (53) corresponds to the cross-sectional shape of the measured object (51) at the slit light irradiation position when observed from a direction different from the irradiation direction. Is a method that has been widely used from the past because of its simplicity, non-contact property, and quantitative property.

When measuring the shape of a three-dimensional curved surface using this light-section method, while moving the slit light (53) in the direction of the arrow (54) in FIG.
By observing in (55) and processing the obtained video signal, the light cutting line (the shape of the light beam pattern) in the screen is extracted every moment (56), and it is reconstructed (57) ,
Construct a curved shape.

As a configuration of the optical system, an optical spot scanner is used as a light source instead of the slit light source (52), and as an imaging system, for example, a PSD (Position Sensit) is used instead of the television camera (55).
There is also a method of using a high-speed optical spot position detecting device known as an ive detector sensor, but the basic principle is the same as that of FIG.

[Problems to be Solved by the Invention] Although the above-described light-section method has various advantages, in order to detect and identify each point on the object to be measured, it is necessary to The process of extracting the light section line is essential, and this causes a problem in measurement accuracy or reliability as described below.

(1) Deterioration of measurement accuracy and spatial resolution due to the shape of the object to be measured; in the optical cutting method, the object to be measured (51) with respect to the optical axis of the slit light (53) as shown in FIG. When the surface is a slope having an angle close to a right angle, the width W of the light beam pattern on the surface of the object is narrow, and therefore highly accurate measurement is possible. However, as shown in FIG. 3 (b), when the surface of the object to be measured becomes a slope having an angle close to parallel to the optical axis of the slit light (53), the width w of the light beam pattern on the surface of the object widens, and The uncertainty of the position when extracting the cutting line increases, the accuracy deteriorates, and the amount of movement of the light beam pattern on the object surface when moving the slit light source (53) increases, which results in spatial measurement. Resolution also deteriorates at the same time.

(2) Decrease in measurement reliability due to the surface reflectance of the object to be measured; in the optical cutting method, it is assumed that the light beam pattern is sufficiently brighter than the surroundings in the process of extracting the light cutting line in the screen. Therefore, for example, if there is a large unevenness in the reflectance on the object surface, or if the slope angle of the object surface is close to the optical axis of the slit light and the reflected light intensity is low, the light cutting line is often extracted. As a result, a break point may occur, or a completely different point may be erroneously detected as an optical cutting line. Such a phenomenon also occurs when background light other than the slit light is present at the time of measurement, and both of them cause a reduction in the reliability of the measurement and a restriction on the measurement environment to which it is applied.

As described above, the light-section method has many application restrictions such as the shape of the object to be measured, the surface texture, or the measurement environment because of some measurement problems caused by the light-section line extraction process. Its applications are limited in spite of its advantages such as simplicity, non-contact, and quantitativeness, and it has been assembled as a general-purpose three-dimensional curved surface shape measuring device and put to practical use. There wasn't.

This invention has been made in order to solve the above-mentioned problems of the light cutting method, and while using an optical system similar to the light cutting method, the slit light source position on the surface to be measured using slit light as a medium. It is an object of the present invention to obtain a method and apparatus for measuring a three-dimensional curved surface shape based on a new measurement principle that does not require an optical cutting line extraction process at all by introducing a new method of coding.

[Means for Solving the Problem] A method for measuring a three-dimensional curved surface shape according to the present invention (claim 1)
Is the angle θ that is not orthogonal to the measurement reference plane on the surface of the object to be measured.
The step of linearly scanning the linear slit light projected from the direction that forms the entire surface of the object to be measured, and the surface of the object to be measured is imaged by the TV camera from an angle different from the slit light. The step of generating a video signal, the step of obtaining the position of the slit light, the signal of each pixel in the screen of the video signal is sequentially captured and stored, and the level of the signal input later for the same pixel is already stored. The level of the signal being input, and when the level of the signal input later is higher, the stored content of that pixel is updated by that level, and for each pixel in the screen of the video signal, When the stored content of a pixel is updated, the information about the position of the slit light at that time is fetched and stored so that the measured signal corresponding to each pixel in the screen of the video signal is stored. For each position on the target surface, a step of generating a composite image in which information on the position of the slit light at the moment when the slit light passes through the position is used as a value of the pixel, and 3 of the measured object based on the composite image. And a step of obtaining a dimensional curved surface shape.

A three-dimensional curved surface shape measuring device according to the present invention (claim 2) is a slit projection device that projects linear slit light onto a surface to be measured from a direction forming an angle θ that is not orthogonal to the first reference plane. Light light means, slit light scanning means for linearly scanning the slit light over the entire surface of the measured object, slit light position measuring means for measuring the position of the slit light, and the measured object surface of the slit light. A television camera that captures an image from a direction different from the light projecting unit to generate a video signal, and a signal of each pixel in the screen of the video signal is sequentially captured and stored, and the level of a signal input later for the same pixel and The stored level of the signal is compared, and when the level of the signal input later is higher, the stored content of that pixel is updated by that level, and the maximum level for each pixel is obtained. For the maximum brightness image calculation means and each pixel in the screen of the video signal, when the maximum brightness image calculation means updates the stored contents of the pixel, information about the position of the slit light at that time is fetched from the slit light position measuring means. , For each position of the surface to be measured corresponding to each pixel in the screen of the video signal, the information about the position of the slit light at the moment when the slit light passes through that position is obtained by calculating the pixel value. A composite image calculation means for generating a composite image having a value of, and an image calculation means for obtaining a three-dimensional curved surface shape of the object to be measured based on the composite image.

Further, in the three-dimensional curved surface shape measuring apparatus according to the present invention, the television camera images the object to be measured from a direction perpendicular to the reference plane (claim 3).

Further, in the three-dimensional curved surface shape measuring apparatus according to the present invention, the image calculation means includes a combined image u (x, y) obtained by the combined image calculation means when scanning the surface to be measured with slit light. Based on the combined image u ₀ (x, y) obtained by the combined image calculation means when the slit light is scanned on the reference surface, the three-dimensional shape f (x, y) of the surface to be measured is set to the reference surface. using a slit light projection angle θ with respect to f (x, y) = { u 0 (x, y) -u (x, y)} tanθ becomes determined based on equation (claim 4).

Further, in the three-dimensional curved surface shape measuring apparatus according to the present invention, the image calculation means is a composite image u (x, y) obtained by the composite image calculation means when scanning the surface to be measured with the slit light and the reference. Synthetic image u obtained by the synthetic image calculation means when scanning the slit light on the surface
_{A combination of 0} (x, y) and a second reference plane parallel to the reference plane and separated by a distance d (the side closer to the television camera is +, the side away from the television camera is −) is scanned. Composite image u ₁ (x, y) obtained by the image calculation means
And are used to calculate the three-dimensional shape f (x, y) of the surface to be measured. It is obtained based on the following formula (Claim 5).

Further, in the three-dimensional curved surface shape measuring apparatus according to the present invention, the image calculation means includes a combined image u (x, y) obtained by the combined image calculation means when scanning the surface to be measured with slit light. The angle θ between the horizontal axis of the TV camera screen and the scanning direction of the slit light with respect to the virtual reference plane
A composite image u ₀ (x, y) obtained by calculation processing using
And the three-dimensional shape f (x,
y) is obtained based on the formula: f (x, y) = {u ₀ (x, y) −u (x, y)} tan θ using the slit light projection angle θ with respect to the reference surface (claim 6). ).

[Operation] In the present invention, the state in which the linear reflection pattern of the slit light moves on the surface of the object to be measured is imaged by the television camera, and for each pixel in the screen, the object corresponding to that pixel is captured. A composite image is created in which the position of the slit light source at the moment when the slit light passes through the position of the surface of the measuring object is the value of the pixel. Then, for example, the three-dimensional curved surface shape of the object to be measured is obtained by obtaining the difference in the value of the corresponding pixel between the above-mentioned combined image and the combined image created in the same manner for the reference plane from which the object to be measured is removed. To measure.

[Embodiment] Prior to the description of an embodiment of the present invention, the measurement principle of the present invention will first be conceptually described based on FIGS. 1 (A) and 1 (B).

As shown in Fig. 1 (A), a slit light (3a) is projected onto the surface of the object to be measured (2) placed on the reference surface (1) from diagonally above and perpendicular to the paper surface. , This slit light (3a)
While moving in the horizontal direction of the paper, for example, the measured object (2)
Images are taken with a TV camera (8) from directly above. At this time, on the monitor television (8a) connected to the television camera (8), it is observed that the linear reflection pattern of the slit light on the object surface moves in the horizontal direction of the screen.

As described above, the line shape of the reflection pattern of this slit light (3a) reflects the unevenness information of the object surface, and in the conventional light cutting method, the line shape of the reflection pattern is extracted from moment to moment. The three-dimensional shape of the object to be measured was measured by reconstructing.

In the present invention, a television camera showing a state in which a linear reflection pattern of slit light (3a) moves on the surface of an object.
Based on the video signal output from (8) and the position signal of the slit light source that changes moment by moment, for each pixel in the screen,
An image having the position of the slit light source at the moment when the slit light passes through the position of the object surface corresponding to the pixel as the value of the pixel is synthesized.

The image thus synthesized represents the height profile of the object surface with the value at each pixel as a reference with respect to the plane A (hereinafter referred to as a temporary reference plane) indicated by the alternate long and short dash line in FIG. There is. In this way, the height profile of the object surface with reference to the temporary reference plane A is measured.

However, the three-dimensional shape measurement of an object is generally not a profile for the temporary reference plane A in FIG. 1 (A), but a surface position on which the measured object 1 is placed (reference plane in FIG. 1 (A), hereinafter reference The profile must be measured with reference to the surface).

In order to meet such a requirement, first, the above-mentioned measurement is performed on the object surface to measure the height profile with reference to the temporary reference plane A, and then the object to be measured is removed and the reference surface is removed. The same measurement is performed for (1) to measure the height profile with reference to the temporary reference plane A. Then, as shown in FIG. 1 (B), these two height profile images, that is, the object plane composites. The difference between the values of corresponding pixels in the image and the reference plane composite image is calculated. By this calculation, the reference plane
A height profile image based on (1) is obtained, and the value of each pixel of this height profile image is proportional to the height of the measurement target surface position corresponding to that pixel with respect to the reference plane (1). It will be what you did.

4 (a), (b), and (c) are configuration diagrams of the optical system of the present invention. FIG. 4 (a) is a perspective view, FIG. 4 (b) is a plan view, and FIG. 4 (c) is a side view. Is.

In FIG. 4 (a), an object to be measured having an xy plane as a reference plane (1) and having an object plane z = f (x, y) on the reference plane (1)
(2) is placed. The television camera (8) observes the object plane at a predetermined angle, for example, directly above, with the z axis as the optical axis. The slit light source (3) projects the slit light (3a) spreading in the y-axis direction at an angle θ with respect to the x-axis and is scanned in the x-axis direction. At this time, change the position of the slit light source (3) to slit light (3a).
Is defined as a position x = x ₀ corresponding to the reference plane (1) (The position of the slit light source (3) can be defined as any plane as long as it is parallel to the xy plane. Here, the position of the slit light source is defined on the xy plane for simplification.) Therefore, the object plane and the slit ray plane are respectively defined by the following equations.

Object plane: z = f (x, y) (1) Ray plane: z = (x-x0) tan θ (2) On the line where the ray hits the object plane, equations (1) and (2) are established at the same time. Therefore, the following relationship holds.

Value u (x, y) of the composite image corresponding to the (x, y) coordinates
Is given by the ray position x ₀ at that time, u
By setting (x, y) x ₀ , the following equation is established.

On the other hand, the reference plane (xy plane) from which the measurement object (1) is removed
Let u ₀ (x, y) be the composite image in f.
By setting (x, y) = 0, the following equation holds.

u ₀ (x, y) = x (5) Therefore, the object profile f (x, y) is expressed by equation (4) and (5).
From the equation, the following equation can be obtained.

f (x, y) = {u ₀ (x, y) −u (x, y)} tan θ (6) The following method can be easily considered as an application example of this measurement principle. First, as a first application example, a method of providing two reference planes is used. That is, in addition to the above-mentioned reference plane,
A second reference plane that is parallel to this and is separated by a distance d (the side closer to the television camera is +, the side away from the television camera is −),
A composite image is similarly calculated on this reference plane. If the composite image on this second reference plane is u ₁ (x, y),
By setting f (x, y) = d in equation (4), Therefore, the object plane f (x, y) is expressed by the equations (4), (5), and (7).
From the formula, it can be obtained in the form of the following formula.

Furthermore, as a second application example, the reference plane is not physically provided, and the composite image u ₀ (x, y) with respect to the virtual reference plane is displayed.
A method is conceivable in which the object plane f (x, y) is obtained by generating it by calculation using the equation (5) and substituting it into the equation (6). However, in this case, when the horizontal axis (raster direction) of the television camera screen and the scanning direction of the slit light source form an angle φ, an image obtained by rotating u ₀ (x, y) = x by the angle φ is obtained. u ₀ (x, y) = xcosφ + ysinφ (9) must be used.

Next, an embodiment of the present invention will be described with reference to FIGS.

FIG. 5 is a block diagram of the three-dimensional shape measuring apparatus according to the above embodiment. The object to be measured (2) on the reference surface (1) which is the standard of measurement
Is placed. The slit light source (3) is mounted on the linear stage (4) and projects the slit light (3a) on the reference surface (1) and the object to be measured (2) at a projection angle of θ. . The linear stage (4) equipped with the slit light source (3) is driven by the motor (6) controlled by the motor controller (5), and the slit light source (3) is parallel to the reference plane (1). Move to.

At this time, the position of the slit light source (3) is set to the linear stage (4).
It is measured by the position sensor (7) incorporated in the and is input to the shape measuring device (9) via the motor controller (5).

The reference plane (1) and the object to be measured (2) are photographed by the television camera (8) arranged so that the optical axis is orthogonal to the reference plane (1), and the obtained video signal is a shape measuring device ( Input to 9).

The shape measuring device (9) is roughly divided into a shape calculating circuit (10) as image calculating means for performing shape calculation by image synthesis, and a command to the motor controller (5) and a shape calculating circuit (10).
It is composed of a sequence controller (11) for controlling the operation timing for the.

When measuring the shape, the shape measuring device (9) is a sequence controller based on a start signal given from the outside.
The linear stage (4) is driven via (11) to set the slit light source (3) to the initial position. Then, the scanning of the slit light source (3) is started.

The shape calculation circuit (10) has an image synthesizing circuit (12) which will be described later in its input section, and at the same time when the slit light source (3) starts scanning, the video signal input from the television camera (8) is momentarily displayed. After processing, for each pixel in the screen, an image composition calculation is performed using the stage position signal at the moment when the slit light image passes through that pixel as the value of that pixel, and one scan of the slit light source (3) is completed. At the same time, the result u (x, y) is transferred to the object plane composite image memory (13).

Next, after removing the measured object (2) from the reference surface (1), the sequence controller (11) returns the slit light source (3) to the initial position, and then scans the slit light source (3) again. To start. The image combining circuit (1) performs the same image combining operation as that performed on the object to be measured (2) on the reference surface (1), and when the scanning of the slit light source is completed, the result u
₀ (x, y) is transferred to the reference plane composite image memory (14).

After completion of these image composition operations, the shape operation circuit (10) uses the difference image operation circuit (15) based on the instruction of the sequence controller (11) and the image in the object plane composite image memory (13) and the reference image. After calculating the image {u ₀ (x, y) -u (x, y)} of the difference in the value of the corresponding pixel with the image of the surface composite image memory (14), the height correction circuit (16) is used. Then, the height profile is calibrated, and the resulting height profile data {u ₀ (x, y) -u (x, y)} tan θ is stored in the three-dimensional shape memory (17).

The height profile data stored in the three-dimensional shape memory (17) is appropriately transferred to a computer or CAD system based on a command from a host computer or CAD system.

In this embodiment, for example, as shown in FIG. 6, when the object to be measured (2) having a slope close to the projection angle of the slit light is measured, the slope of the slope becomes the projection angle of the slit light. Since it is very close, when the slit light comes to the position indicated by "1" in the figure, the entire slope becomes evenly bright, and the contents stored in the object plane composite image memory (13) are as shown by reference numeral (13a). Become. Reference plane synthesis memory (14) is stored as code (14a)
As shown in. Therefore, the data stored in the three-dimensional shape memory (17) after the image data is arithmetically processed by the difference image arithmetic circuit (15) and the height correction circuit (16) is the code (1
7a). From this, it is understood that a sufficiently high resolution is obtained even for a shape having a surface close to the angle of the slit light.

Conventionally, it is difficult to extract the light section line from this image (13a), as described above, and if the light section method is applied to such a slope, both measurement accuracy and spatial resolution can be expected. However, in this embodiment, even with respect to such an inclined surface, it is possible to perform measurement with the measurement accuracy and spatial resolution of the beam width of the slit light to the sampling pitch, and generally, depending on the shape of the object to be measured. It is possible to realize shape measurement that does not exist.

FIG. 7 is a block diagram showing an example of an image synthesizing circuit (12) which is a component of the shape measuring apparatus (9).

The image composition circuit (12) processes the video signal input from the television camera (8) and calculates the brightness at the moment when it becomes brightest for each pixel, and the maximum brightness image calculation unit (18) and each pixel. Is composed of an image composition operation unit (19) that performs an image composition operation using the stage position signal at the moment when the maximum luminance is taken as the value of that pixel, and a synchronization circuit (20) for controlling these. , A memory address generation circuit (21) and an output control circuit (22).

The maximum brightness calculation unit (18) A / D-converts the video signal based on the timing signal output from the synchronization circuit (20) centering on the maximum brightness image memory (23) which is a buffer memory for maximum brightness image calculation. A / D conversion circuit for digitization (2
4), maximum brightness image memory (25) that controls the reading and writing of data at the maximum brightness image memory address specified by the memory address generation circuit (21), and the image and maximum brightness memory input from the TV camera Circuit that compares the values of the corresponding pixels in the image
(26) and a switch circuit (27).

On the other hand, the composite image calculation unit (19) is mainly composed of a composite image memory (28) for storing the composite image calculation result,
Based on the output signal of the comparison circuit (26) in the maximum luminance image calculation unit (18), the signal level input from the television camera is higher than the pixel value at the address of the corresponding maximum luminance image memory (23). Synthetic image memory control circuit having a function of writing the stage position in the synthetic image memory (28) when it is large
It is equipped with (29).

This circuit starts from the state where the maximum brightness image memory (13) and the composite image memory (28) are cleared to zero at the timing of the start of calculation, and the video signal input from the television camera is A / D conversion circuit. While digitizing using (24), compare the value of the video signal with the value of the pixel of the maximum brightness image memory (13) corresponding to the position of that pixel, and only when the value of the video signal is larger, the maximum brightness Image memory (13)
At the same time as updating the value of that pixel of
It has a function of writing the stage position signal to the corresponding pixel of the composite image.

In this way, the above calculation is performed for the time designated by the calculation control signal from the outside. As a result, the predetermined image described above is generated in the composite image memory (28) at the end of the calculation. The composite image calculated in this way is transferred to the next arithmetic circuit via the output control circuit (22).

Although the composite image of the reference surface (1) is created at the time of measurement in the above-described embodiment, the composite image of the reference surface (1) only needs to be created once, and thus is created first at the time of the second and subsequent measurements. The synthetic image of the reference plane (1) may be used as it is. In addition, since the composite image of this reference plane (1) has a simple configuration, a calculation function is added to the shape calculation circuit (10), and the virtual reference plane is calculated by equation (5), and the composite image is obtained. make,
It may be stored in the reference plane composite image memory (14).

Further, in the above embodiment, an example in which one reference surface is provided is shown, but a second reference surface may be provided in addition to the reference surface (1). The second reference plane is installed between the reference plane (1) and the slit light source (3) with a distance d to the reference plane (1). The shape calculation circuit (10a) in this case, as shown in FIG.
₁ (x, y) is created by the image synthesizing circuit (12) in the same manner as described above, and the synthetic image memory (14a) for storing it is stored in addition to the object synthetic image memory (13) and the reference plane synthetic image memory (14). Further, in place of the difference image calculation circuit (15) and the height correction circuit (16), a calculation circuit (30) for calculating the above equation (8) is provided.

[Effects of the Invention] As described above, according to the present invention, even when an optical system similar to the light cutting method is used, for example, the object to be measured has a shape of an inclined surface close to the projection angle of the slit light. Even so, since the three-dimensional shape is obtained by calculating the difference between the values of the corresponding pixels in the combined image of the object to be measured and the combined image of the reference surface, the beam of slit light is obtained even on such a slope. It is possible to perform measurement with a measurement accuracy and a spatial resolution of a width to a sampling pitch, and it is possible to perform shape measurement that does not depend on the shape of the measurement target.

Further, according to the present invention, the state in which the linear reflection pattern of the slit light is moving on the surface to be measured is imaged by the television camera, and for each pixel in the screen, the object surface corresponding to the pixel is displayed. When the slit light passes through the position, an image composition calculation is performed using the position of the slit light source as the value of that pixel. The necessary conditions for this image composition calculation to be established and the shape information to be obtained correctly correspond to each pixel. The only condition is that the brightness at each position on the surface to be measured is maximized at the moment when the slit light passes through the position.

Therefore, not only the unevenness of the spatial surface reflectance of the object to be measured does not affect the measurement, but even if there is background light, the amount of light is constant over time and the signal of the TV camera is not saturated. As long as it is brightness, the brightness of each point on the surface of the object also becomes maximum at the moment when the slit light passes, so it is possible to perform measurement without being affected by the surface reflectance or background light of the measurement target. is there.

Furthermore, according to the present invention, by providing two reference planes for measurement, it is possible to perform measurement independent of the projection angle of the slit light, and the measurement accuracy is improved.

Further, according to the present invention, since the calculation is performed when obtaining the composite image of the reference plane, the work at the time of measurement is simplified.

[Brief description of drawings]

1 (A) and (B) are explanatory views showing the measurement principle of the present invention, FIG. 2 is a conceptual diagram of a conventional optical cutting method, and FIG. 3 (a).
FIG. 4B is an explanatory diagram showing a change in measurement accuracy depending on the slope angle of the conventional light cutting method, and FIGS. 4A, 4B and 4C are explanatory diagrams showing the configuration of the optical system of the present invention. (A) is a perspective view,
The figure (b) is a top view and the figure (c) is a side view. FIG. 5 is a block diagram of a three-dimensional shape measuring apparatus according to an embodiment of the present invention, FIG. 6 is an explanatory view showing an example of measuring a slope shape, and FIG. 7 shows details of the image synthesizing circuit of FIG. FIG. 8 is a block diagram showing another example of the shape calculation circuit. (1): Reference plane, (2): Object to be measured, (3): Slit light source,
(4): Linear stage, (5): Motor controller,
(6): Motor, (7): Position sensor, (8): TV camera,
(9): Shape measuring device, (10): Shape calculation circuit, (11): Sequence controller, (12): Image composition calculation circuit, (13):
Object plane composite image memory, (14): Reference plane composite image memory,
(15): Difference image calculation circuit, (16): Height correction circuit, (17): 3
Dimensional shape memory, (18): maximum brightness image calculation unit, (19): composite image calculation unit, (20): synchronization circuit, (21): memory address generation circuit, (22): output control circuit, (23) : Maximum brightness image memory, (24): A / D conversion circuit, (25): maximum brightness image memory control circuit, (26): comparison circuit, (27): switch circuit, (2
8): Composite image memory.

Claims (6)

[Claims] 1. A linear slit light projected onto a surface of an object to be measured from a direction forming an angle θ which is not orthogonal to a measurement reference plane,
A step of linearly scanning the entire surface of the object to be measured; a step of generating a video signal by imaging the surface of the object to be measured by a television camera from an angle different from the slit light; and the position of the slit light And the step of obtaining the signal of each pixel in the screen of the video signal is sequentially captured and stored, and the level of the signal input later for the same pixel is compared with the level of the signal already stored, A step of updating the stored content of the pixel by the level of the input signal when the level is higher, and for each pixel in the screen of the video signal, when the stored content of the pixel is updated in the step, By capturing and storing information regarding the position of the slit light at that time, for each position of the measured surface corresponding to each pixel in the screen of the video signal, A step of generating a composite image in which information about the position of the slit light at the moment when the slit light passes through the position is a value of the pixel, and a step of obtaining a three-dimensional curved surface shape of the object to be measured based on the composite image A method for measuring a three-dimensional curved surface shape, comprising: 2. A slit light projecting means for projecting linear slit light from a direction forming an angle θ which is not orthogonal to the first reference plane on the surface to be measured, and slits over the entire surface of the object to be measured. Slit light scanning means for linearly scanning light, slit light position measuring means for measuring the position of slit light, and a surface to be measured from a direction different from that of the slit light projecting means to obtain a video signal. The TV camera to generate and the signal of each pixel in the screen of the video signal are sequentially captured and stored, and the level of the signal input later on the same pixel and the level of the signal already stored are compared, A maximum luminance image calculation means for obtaining the maximum level for each pixel by updating the stored content of the pixel according to the level of the signal input from the video signal, For each pixel in the screen of, when the maximum brightness image calculation means updates the stored contents of the pixel, the information about the position of the slit light at that time is fetched from the slit light position measuring means, and as the value of the pixel. By determining, for each position of the surface to be measured corresponding to each pixel in the screen of the video signal, information regarding the position of the slit light at the moment when the slit light passes through that position is set as the value of that pixel. A three-dimensional curved surface shape measuring device comprising: a composite image calculation means for generating an image; and an image calculation means for obtaining a three-dimensional curved surface shape of a measurement target based on the composite image. 3. The three-dimensional curved surface shape measuring apparatus according to claim 2, wherein the television camera images the object to be measured from a direction perpendicular to the reference plane. 4. The image calculation means scans the composite image u (x, y) obtained by the composite image calculation means when the surface to be measured is scanned with the slit light, and the reference surface with the slit light. The three-dimensional shape f (x, y) of the surface to be measured is calculated by using the slit light projection angle θ with respect to the reference plane, based on the composite image u ₀ (x, y) sometimes obtained by the composite image calculation means. The three-dimensional curved surface shape measuring apparatus according to claim 2, which is obtained based on an expression of f (x, y) = {u ₀ (x, y) −u (x, y)} tan θ. 5. The image calculation means, when the surface of the object to be measured is scanned with the slit light, when the composite image u (x, y) obtained by the composite image calculation means and the reference surface are scanned with the slit light. A composite image u ₀ (x, y) obtained by the composite image computing means, and a second reference plane that is parallel to the reference plane and is separated by a distance d (+ on the side closer to the television camera and − on the side away). The composite image u obtained by the composite image calculation means when the slit light is scanned with respect to
_{Using 1} (x, y), the three-dimensional shape f of the surface to be measured f
(X, y) The three-dimensional curved surface shape measuring device according to claim 2, which is obtained based on the following equation. 6. The image calculation means includes a composite image u (x, y) obtained by the composite image calculation means when the surface to be measured is scanned with slit light, and a television camera with respect to a virtual reference plane. A composite image u obtained by calculation using the angle θ formed by the horizontal axis of the screen and the scanning direction of the slit light
_{Based on 0} (x, y), the three-dimensional shape f (x, y) of the surface to be measured is calculated by using the slit light projection angle θ with respect to the reference plane: f (x, y) = {u ₀ ( The three-dimensional curved surface shape measuring apparatus according to claim 2, which is obtained based on an expression of (x, y) -u (x, y)} tan θ.