JP2007271300A - Surface profile measuring method and device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To precisely acquire measurement data of an internal surface profile of a cylindrical member with no defect even for a complicated structure such as a groove or a rib, and to extend a measurement range in a height direction in profile measurement without deteriorating measurement accuracy. <P>SOLUTION: Slit-shaped measurement light is obliquely incident toward a cylindrical internal wall surface from both sides of two openings existing in a cylindrical object to be inspected, and the cylindrical internal surface profile is measured by a light sectioning method. Profile data acquired from two sets of measurement systems provided in substantially facing directions with respect to a part to be inspected are used complementarily to complement the measured profile data with a defect part. A light convergence point shifted for each wavelength is generated using a multi-wavelength light source and a collector lens having large wavelength dispersion so as to be used in the measurement, and the measurement accuracy is made compatible with a wide measurement range. According to the means, the internal profile of the cylindrical part having a complicated profile can be measured precisely without omission. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a technique for measuring the surface shape of a test surface by projecting measurement light having a plurality of wavelengths onto the test surface and observing the reflection pattern.

  As a technique for measuring the inside of a conventional cylindrical shape, FIG. 15 shows a cross-sectional view of a conventional in-cylinder shape measuring apparatus (see, for example, Patent Document 1).

  In FIG. 15, laser light sources 2 </ b> A and 2 </ b> B are attached to the casing 1. The laser beams emitted from the laser light sources 2A and 2B are respectively bent by the plane mirror 3, and further bent by the plane mirrors 4A and 4B, and enter the test surfaces 6A and 6B from the windows 5A and 5B of the casing 1 at an angle θ. Is done. The laser light reflected after entering the test surfaces 6A and 6B returns to the windows 5A and 5B again and reaches the optical position detection element 9 through the lenses 7A and 7B and the mirror 8.

  The return position of the reflected laser light to the lenses 7A and 7B varies depending on the distance from the windows 5A and 5B to the test surfaces 6A and 6B. Therefore, the arrival position of the laser beam of the optical position detection element 9 also changes. Thus, the distance from the windows 5A, 5B to the test surfaces 6A, 6B is obtained by performing triangulation with the laser using the plane mirrors 4A, 4B, the test surfaces 6A, 6B, the lenses 7A, 7B and the mirror 8. Can be calculated.

  FIG. 16 is a diagram showing a measurement situation by a conventional in-cylinder shape measurement method.

Here, when the test surface 10 is not aligned with the axis of the casing 1, the locus of the light reflected from the test surface 10 of the light emitted from the window 5A does not enter the lens 7B as shown in FIG. Further, when the groove shape is on the test surface like the test surface 11, the locus of the light reflected by the test surface 11 does not enter the lens 7A as shown in FIG.
Japanese Patent Laid-Open No. 7-260439 (page 7, FIG. 1)

  In the conventional configuration, when the test surface is not aligned with the axis of the casing, the error in the reflection angle of the reflected light on the test surface increases, the lens movement width increases, and the detection error increases. Further, when the groove shape is on the test surface, the light reflected by the test surface does not always return to the lens, and there is a high possibility that measurement is impossible. Therefore, depending on the state of the test surface, the inner side of the cylindrical shape may not be measured.

  In addition, when trying to improve the measurement accuracy inside the cylindrical shape, it is necessary to collect the measurement light. However, if the measurement light is collected, the measurement light will be in a defocused state at a position far from the focus of the measurement light. , Measurement accuracy decreases. In the conventional configuration, when there is a large error in the shape of the measurement surface, there is a high possibility that the measurement surface will be in a defocused state, and the measurement accuracy tends to decrease.

  An object of the present invention is to provide a measurement technique that can measure the structure of various test surfaces and has high measurement accuracy.

  In order to achieve the above object, the surface shape measurement method of the present invention irradiates the test surface with light having different wavelengths from each other along phase-parallel optical axes, and splits the light reflected by the test surface for each wavelength. Then, after measuring the distance from the reference surface of the test surface for each wavelength based on the brightness of the dispersed light for each wavelength, the shape data for each wavelength by the measurement is relatively shifted and overlapped, It is characterized by measuring the surface shape of the inspection surface.

  The surface shape measuring device of the present invention includes a light projecting device, an image capturing device that captures light emitted from the light projecting device and reflected by a test surface, and a processing device that processes data obtained by the image capturing device. The surface shape measuring device includes a light projecting device having a plurality of light sources having different wavelengths from each other and an optical system that converts light from the plurality of light sources into phase-parallel optical axes. A plurality of imaging elements having different effective wavelengths and an optical system that splits light incident on the imaging device into a plurality of lights having different wavelengths.

  As described above, according to the surface shape measurement method and the measurement apparatus of the present invention, it is possible to perform measurements on various structures of the test surface, and provide a measurement technique with high measurement accuracy.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
FIG. 1 is a schematic diagram of a cylindrical inner side measuring apparatus according to the first embodiment.

  In FIG. 1, the in-cylinder shape measuring device of the present embodiment is composed of light projecting devices 12A and 12B and imaging devices 13A and 13B. Moreover, the measurement cylinder 14 which is a test object has two openings, openings 15A and 15B. In the present embodiment, the first light projecting device 12A and the second imaging device 13B are arranged on the first opening 15A side, and the first imaging device 13A and the second imaging device 13B are arranged on the second opening 15B side. Are arranged.

  The first light-projecting device 12A and the first image-capturing device 13A use the first image-capturing device to generate reflected light generated by irradiating the measurement spot 17 with the first measurement light 16 emitted from the first light-projecting device 12A. It arrange | positions so that it can observe by 13A. In addition, the second light projecting device 12B and the second imaging device 13B provide the second reflected light generated by irradiating the measurement location 17 with the second measurement light 18 emitted from the second light projecting device 12B. It arrange | positions so that it can observe with the imaging device 13B.

  In addition, the first light projecting device 12A, the first image capturing device 13A, the second light projecting device 12B, and the second image capturing device 13B are disposed at positions where no mechanical interference occurs.

  The angle between the first measurement light 16 until the measurement point 17 is irradiated and the second measurement light 18 until the measurement point 17 is irradiated is the first measurement light 16 and the second measurement light. Of the conditions in which 18 is not obstructed by the cylinder 14 to be measured, it is desirable to take an angle that minimizes the value.

  In this case, the same measurement point 17 is observed using the first measurement light 16 and the second measurement light 18 by reducing the angle formed by the first measurement light 16 and the second measurement light 18. It is equivalent to making the angle formed by the different observation directions close to 90 °. By doing so, the first measurement light 16 and the second measurement light 17 are measured with respect to the measurement point 17 constituted by grooves and irregularities. This is because at least one of the measurement lights 18 can be projected without being interrupted by the grooves or irregularities.

  FIG. 2 is a schematic cross-sectional view of the in-cylinder shape measuring apparatus of FIG.

  In FIG. 2, the straight line connecting the first light projecting device 12 </ b> A and the first imaging device 13 </ b> A and the straight line connecting the second light projecting device 12 </ b> B and the second imaging device 13 </ b> B are They are arranged so as not to be the same as the measured cylindrical shaft 19 determined from the cylindrical shape.

  The first measurement light 16 emitted from the first light projecting device 12A is collected at the measurement location 17 and becomes the first line-shaped measurement light 20A. Similarly, the second measurement light 18 emitted from the second light projecting device 12B is condensed at the measurement location 17 and becomes the second linear measurement light 20B.

  The first linear measurement light 20 </ b> A and the second linear measurement light 20 </ b> B are arranged so as to cross each other at the measurement location 17. The region of the measurement location 17 is configured to be a position that can be observed from both the first imaging device 13A and the second imaging device 13B.

  The first imaging device 13A and the second imaging device 13B are connected to the first image data synthesis device 21A and the second image data synthesis device 21B, respectively. The first image data synthesizing device 21A and the second image data synthesizing device 21B are further connected to an image data analyzing device 22, and the image data analyzing device 22 is connected to an image display device 23.

  Here, it is desirable not to directly capture the reflected light that is regularly reflected among the reflected light generated by reflecting the respective measurement lights 16 and 18 at the measurement location 17 with the respective imaging devices 13A and 13B. For example, the first imaging device 13A is arranged at a position where the optical axis after the optical axis of the first measurement light 16 is regularly reflected at the measurement location 17 and the optical axis of the first imaging device 13A do not coincide. This is realized.

  This is based on the power of the emitted light from each of the light projecting devices 12A and 12B, but it is intended to prevent the brightness of each of the line-shaped measurement lights 20A and 20B from changing significantly due to the unevenness of the measurement location 17. This is because.

  In addition, when adjustment of each imaging device 13A and 13B can absorb the change in luminance between the reflected light when specularly reflected at the measurement location 17 and the reflected light when not regularly reflected, each of the imaging devices described above is used. There is no need to consider the relative positional relationship between 13A and 13B and the optical axis of the reflected light.

  The cylinder 14 to be measured is fixed to a moving stage 24 capable of fixing a cylindrical member, and the moving stage 24 is parallel to the cylinder axis 19 to be measured (direction of arrow C in FIG. 2). It is configured to be able to perform a straight movement operation and a rotation operation in the circumferential direction of the cylindrical shaft 19 to be measured (direction of arrow D in FIG. 2). Furthermore, the moving stage 24 includes a scale that manages the linear movement amount in the direction of arrow C and the rotation angle in the direction of arrow D, and can grasp the movement amount of the moving stage 24 from a predetermined position.

  The moving stage 24 is connected to a stage operating device 25, and the stage operating device 25 is connected to the image data analyzing device 22.

  FIG. 3 is a schematic cross-sectional view of the in-cylinder shape measuring apparatus of FIG.

  In FIG. 3, the second light projecting device 12 </ b> B and the second imaging device 13 </ b> B are omitted for explanation.

  In FIG. 3, the first projector 12A includes a first laser light source 26 (first red laser light source 26R, first green laser light source 26G, first blue laser light source 26B) and a first collimator. Lens 27 (first red collimating lens 27R, first green collimating lens 27G, first blue collimating lens 27B), first cylindrical lens 28, first prism mirror 29, and first And the polarizing element 30.

  Here, although described as a collimating lens for each color, this is a lens for converting incident light in the wavelength region of each color into parallel light.

  Further, the first prism mirror 29 includes a first prism mirror red reflecting surface 29R and a first prism mirror blue reflecting surface 29B in order to align the optical axes of incident light for each frequency of red, blue, and green. Have.

  The first imaging device 13A includes a first imaging element 31 (a first red imaging element 31R, a first green imaging element 31G, and a first blue imaging element 31B), and a first spectral prism 32. And a first imaging lens 33, a first objective lens 34, and a first polarizing filter 35. Here, the first spectral prism 32 has a first spectral prism red reflecting surface 32R and a first spectral prism blue reflecting surface 32B in order to divide the incident light into light of red, blue and green frequencies. Have

  The lights emitted from the first red laser light source 26R, the first green laser light source 26G, and the blue laser light source 26B in the first light projecting device 12A are respectively the first red collimating lens 27R and the first red laser light source 26B. The light is refracted by the green collimating lens 27 </ b> G and the first blue collimating lens 27 </ b> B to become parallel light, and is incident on the first prism mirror 29. The light from the first red laser light source 26R is reflected by the first prism mirror red reflecting surface 29R, the light from the first blue laser light source 26B is reflected by the first prism mirror blue reflecting surface 29B, and the first The light from the green laser light source 26G travels straight through the first prism mirror 29. Each light from the first prism mirror 29 passes through the first cylindrical lens 28 with the same optical axis, and is condensed in a direction parallel to the first measurement light source optical axis 36 direction.

  The first prism mirror 29 is provided with the first prism mirror red reflecting surface 29R and the first prism mirror blue reflecting surface 29B. If the purpose of aligning the optical axes of the light for each frequency is satisfied, these are provided. A green reflecting surface may be provided by replacing any one of the above.

  The light reflected from the measurement location 17 becomes the first linear measurement light 37. Specifically, the first red measurement light 16R emitted from the first red laser light source 26R through the first prism mirror 29 is a thin first red line shape parallel to the Y direction at the measurement location 17. It becomes the measurement light 37R. Similarly, the first green measurement light 16G from the first green laser light source 26G becomes the first green line-shaped measurement light 37G, and the first blue measurement light 16B from the first blue laser light source 26B is The first blue line-shaped measurement light 37B is obtained.

  The first cylindrical lens 28 uses an optical material having a large wavelength dispersion, such as an optical material having an Abbe number of 40 or less. By using such a material, the respective condensing positions of the first red measurement light 16R, the first green measurement light 16G, and the first blue measurement light 16B are changed in the direction of the first measurement light source optical axis 36. Can be made.

  In this manner, the first imaging device 13 </ b> A captures an image in which each of the linear measurement light 37 irradiated to the measurement location 17 is irradiated in the vicinity of the measurement location 17.

  Specifically, the image irradiated with the first red linear measurement light 37R near the measurement location 17 is reflected near the measurement location 17, and the first polarizing filter 35, the first objective lens 34, the first The light enters the first spectroscopic prism 32 through one imaging lens 33. The light reflected by the first red line-shaped measurement light 37R in the vicinity of the measurement location 17 is reflected by the first spectral prism red reflecting surface 32R and forms an image on the first red image pickup device 31R. Similarly, the light reflected by the first green line-shaped measurement light 37R in the vicinity of the measurement point 17 travels straight through the first spectral prism 32 and forms an image on the first green image pickup device 31G, thereby forming the first blue color. The light reflected by the line-shaped measurement light 37B in the vicinity of the measurement location 17 is reflected by the first spectral prism blue reflection surface 32B and forms an image on the first blue image pickup device 31B.

  Further, the first spectral prism 32 is provided with the first spectral prism red reflecting surface 32R and the first spectral prism blue reflecting surface 32B. If the purpose of dividing the light into each frequency is satisfied, these may be included. A green reflective surface may be provided by replacing either one.

  The first red imaging device 31R has a first red line at the center of the first red imaging device 31R when the measurement location 17 is observed through the first imaging lens 33 and the first objective lens 34. The measurement light 37R is arranged so as to form an image. Similarly, the first green image sensor 31G is arranged so that the first green line-shaped measurement light 37R forms an image at the center of the first green image sensor 31G, and the first blue image sensor 31B is The first blue line-shaped measurement light 37B is arranged in the center of the first blue image pickup device 31B. The three image sensors are individually optical axis aligned. Thereby, it becomes possible to capture only the linear measurement light of one wavelength light with respect to one imaging device.

  By adopting such a configuration, it becomes possible to capture only one type of line-shaped measurement light of wavelength light with respect to one imaging apparatus.

  The direction of the polarization axis of the first polarizing element 30 and the direction of the polarization axis of the first polarizing filter 35 are matched.

  The first red image sensor 31R, the first green image sensor 31G, and the first blue image sensor 31B are each connected to the first image data composition device 21A. In the first image data synthesis device 21A, the image data sent from the first red image sensor 31R, the first green image sensor 31G, and the first blue image sensor 31B are combined into one data.

  The first image data synthesizing device 21A is further connected to the first image display device 23. As shown in the first image display device 23 of FIG. An image irradiated on the measurement spot 17 is displayed.

  Here, the first light projecting device 12 </ b> A and the first imaging device 13 </ b> A exist at positions shifted in the Y-axis direction with the measured cylindrical shaft 19 in between.

  FIG. 4 is a perspective view of the in-cylinder shape measuring apparatus of FIG.

  FIG. 4 is shown in order to clarify the positional relationship between the first light projecting device 12A and the first imaging device 13A, and the vicinity of the measurement point 17 and the first location from a direction different from FIG. It is a figure which shows the outline of the positional relationship of 12 A of light projection apparatuses, and the 1st imaging device 13A.

  FIG. 5 is a schematic cross-sectional view of the in-cylinder shape measuring apparatus of FIG. However, in FIG. 5, the first light projecting device 12 </ b> A and the first imaging device 13 </ b> A are omitted for explanation. In FIG. 5, description of the same reference numerals as those in FIG. 3 is omitted.

  In FIG. 5, the second light projector 12B includes a second red laser light source 38R, a second green laser light source 38G, a second blue laser light source 38B, a second red collimating lens 39R, It comprises a green collimating lens 39G, a second blue collimating lens 39B, a second cylindrical lens 40, a second prism mirror 41, and a second polarizing element 42. Further, the second prism mirror 41 includes a second prism mirror red reflection surface 41R and a second prism mirror blue reflection surface 41B in order to align the optical axes of incident light for each frequency of red, blue, and green. Have.

  The second imaging device 13B includes a second red imaging element 43R, a second green imaging element 43G, a second blue imaging element 43B, a second spectral prism 44, and a second imaging lens. 45, a second objective lens 46, and a second polarizing filter 47. Here, the second spectral prism 44 has a second spectral prism red reflecting surface 44R and a second spectral prism blue reflecting surface 44B in order to divide the incident light into light of red, blue and green frequencies. Have

  The second red measurement light 18R emitted from the second red laser light source 38R through the second prism mirror 41 becomes a thin second red line-shaped measurement light 48R parallel to the Y direction at the measurement location 17. . Similarly, the second green measurement light 18G from the second green laser light source 38G becomes the second green line-shaped measurement light 48G, and the second blue measurement light 18B from the second blue laser light source 38B is The second blue line-shaped measurement light 48B is obtained.

  However, the direction of the polarization axis of the second polarizing element 42 and the direction of the polarization axis of the second polarizing filter 47 are adjusted so as to be orthogonal to the directions of the first polarizing element 30 and the first polarizing filter 35.

  As a result, the first measurement light 16 is blocked by the first polarizing filter 35 before entering the second imaging device 13B, so that the first measurement light 16 enters the second imaging device 13B, or Conversely, when the second measurement light 18 is incident on the first imaging device 13A, it is possible to prevent a decrease in accuracy of the shape measurement data caused by detection of light not used for measurement in each imaging device.

  First, a method of expanding the measurement range in the depth direction of the cylindrical inner surface shape using measurement systems using three different wavelengths will be described with reference to FIGS.

  FIG. 6 is an enlarged schematic view of the vicinity of the measurement location 17 in FIG.

  In FIG. 6, when acquiring the shape data of the groove structure 54 in the vicinity of the measurement location 17, the first red line-shaped measurement light 37R, the first green line-shaped measurement light 37G, and the first blue line-shaped measurement light 37B. As shown in FIG. 4, the image of the measurement light having different wavelengths has a different condensing position. It is necessary to use depending on the type.

  In FIG. 6, the first image pickup device 13A is adjusted so that the first red line-shaped measurement light 37R forms an image near the center of the first red image pickup device 31R. The first green line-shaped measurement light 37G is adjusted to form an image near the center of the first image pickup element 31B, and the first blue line-like measurement light 37B is formed near the center of the first blue image pickup element 31B. .

  Therefore, when the inner surface shape of the cylinder 14 to be measured is observed with the first imaging device 13A, the height range of the cylinder inner surface shape observed with the first red image pickup device 31R in FIG. 6 is the first red observation. Similarly, the cylinder inner surface shape height range observed by the first green image pickup device 31G is the region 56R, and the cylinder inner surface shape observed by the first green image pickup region 31G and the first blue image pickup device 31B. The first height range is the blue observation region 56B.

  Here, the first red observation area 56R, the first green observation area 56G, and the first blue observation area 56B are overlapped by about 10 to 15% in the height direction (h direction in FIG. 12). The imaging device 13A is adjusted in advance.

  FIG. 7 is a diagram showing a coupling state of three cylinder inner surface shape data in the first embodiment. FIG. 7A is a diagram showing measurement data of each of the three wavelengths in the first embodiment, and FIG. 7B shows a combination state of the three data obtained in FIG. 7A. FIG. 7C is a diagram in which the combined data obtained in FIG. 7B is integrated into one data.

In FIG. 7A, assuming that the in-cylinder shape data calculated from the three image sensor image data is red shape data 57R, green shape data 57G, and blue shape data 57B, respectively, red shape data 57R (W _R , h _R ), green shape data 57G (W _G , h _G ), and blue shape data 57B (W _B , h _B ).

Since the region observed by the image sensor of each wavelength varies depending on the adjustment of the first image pickup device 13A and the installation error of each image sensor, the red shape data 57R (W _R , h _R ) and the green shape data 57G ( Between W _G , h _G ) and blue shape data 57B (W _B , h _B ), offsets are included in both h (height) coordinates and W (position) coordinates. The offset amount of the green shape data 57G (W _G , h _G ) with respect to the red shape data 57R (W _R , h _R ) is (W _P , h _P ), and the blue shape with respect to the red shape data 57R (W _R , h _R ) The offset amount of the data 57B (W _B , h _B ) is assumed to be (W _Q , h _Q ). FIG. 7B shows the relative relationship between the red shape data 57R, the green shape data 57G, and the blue shape data 57B. The offset values (W _P , h _P ), (W _Q , h _Q ) are added to the green shape data 57G (W _G , h _G ) and the blue shape data 57B (W _B , h _B ), and (W _G + W _P , h _G + h _P ), (W _B + W _Q , h _B + h _Q ), and the red shape data 57R, the green shape data 57G, and the blue shape data 57B are collected into an observation region as shown in FIG. Cylinder inner surface shape data (W, h). Here, the red shape data 57R, the green shape data 57G, and the blue shape data 57B are superimposed while being shifted relative to each other in the height direction of the measurement surface 17 (the depth direction of the test surface). The amount of shift is determined according to the focal length of light for each color, and the shift is performed so that the deeper the focal length, the deeper the depth.

  FIG. 8 is a diagram showing a processing flow of measurement using a light source of three wavelengths in the first embodiment. Details of specific shape data acquisition in FIG. 8 will be described later, and will be omitted here.

In FIG. 8, first, in step S1 to step S3, the cylinder inner surface shape data of the cylinder 14 to be measured is acquired by using the first measurement system. At this time, the first imaging device 13A captures projection images of the first red line-shaped measurement light 37R, the first green line-shaped measurement light 37G, and the first blue line-shaped measurement light 37B on the inner surface of the cylinder. Thus, image data reflecting the shape of the measurement location 17 is obtained from the first red image sensor 31R, the first green image sensor 31G, and the first blue image sensor 31B, and the red shape data 57R is obtained from the image data. (W _R , h _R ), green shape data 57G (W _G , h _G ), and blue shape data 57B (W _B , h _B ) are obtained.

Next, in steps S4 and S5, an offset amount (W for green shape data 57G (W _G , h _G ), blue shape data 57B (W _B , h _B ) with respect to red shape data 57R (W _R , h _R ). Green shape data 57G ′ (W ′ _G , h ′ _G ) = (W _G + W _P , h _G + h _P ) after offset correction by adding _P , h _P ), (W _Q , h _Q ), after offset correction The blue shape data 57B ′ (W ′ _B , h ′ _B ) = (W _B + W _Q , h _B + h _Q ) is obtained.

The red shape data 57R (W _R , h _R ) obtained in this way, the offset green shape data 57G ′ (W ′ _G , h ′ _G ), and the offset blue shape data 57B ′ (W ′ _B , In step S6, h ′ _B ) are superimposed on each other as shown in FIG. 7 to form an actual surface shape, and the measurement is completed.

  As described above, by measuring the cylinder inner surface shape using a light source of three wavelengths in the measurement system, the height direction is compared with the case where the same measurement is performed using a monochromatic light source to obtain the cylinder inner surface shape data. The measurement range is approximately three times larger, and even if the measurement range in the height direction of the cylinder inner surface shape is wide, there is no need to readjust the measurement optical system to match the cylinder inner surface shape height, thus increasing the measurement time It is possible to measure the shape of the inner surface of the cylinder with large irregularities without causing the deformation.

  In addition, when measuring a cylindrical inner surface shape with large irregularities using a monochromatic light source, there is a means to increase the measurement range in the height direction by increasing the depth of focus of the measurement light, but in that case the line shape The width of the measurement light is widened and the measurement resolution of the in-cylinder shape is impaired. However, if the shape of the cylinder inner surface is measured using a light source of 3 wavelengths in the measurement system and attention is paid to each wavelength, it is possible to keep the depth of focus without intentionally increasing the depth of focus. The measurement range in the height direction of the cylinder inner surface shape can be expanded without impairing the measurement resolution of the shape.

  Next, a method for obtaining the shape data of the measurement location 17 from the projection image of the line-shaped measurement light onto the measurement location 17 will be described.

  FIG. 9 shows a schematic diagram of a projected image of the linear measurement light in the vicinity of the measurement location 17.

  Here, FIG. 9A is a perspective view of the projection image of the linear measurement light near the measurement location 17, and FIG. 9B is a cross-sectional view of the projection image of the line measurement light near the measurement location 17. .

  In FIG. 9A, the measurement light 50 incident from the light projecting device at an angle θ along the longitudinal direction of the rib structure 49 having a height h existing at the measurement location 17 is changed into a line-shaped measurement light 51 having a width L. Forming. The line-shaped measurement light 51 is reflected at an angle ω and enters the imaging apparatus.

  Here, the plane 52a where the optical axis of the measurement light 50 from the light projecting device exists and the plane 52b where the optical axis of the light incident on the imaging device exists have an angle φ.

  Here, it is assumed that the image data shown in the shape of FIG. 9B is displayed on the image display device 23 as a projection image of the line-shaped measurement light 51 of FIG. An image area 53 indicated by a broken line corresponds to the size of the image sensor incorporated in the image pickup apparatus.

  The image data displayed in the image area 53 is obtained from any of the first blue image sensor 31B, the first red image sensor 31R, and the first green image sensor 31G shown in FIG. The same method is used for obtaining shape data for image data.

  Assuming that the image magnification of the imaging device is M, the height h of the rib structure 49 from the image obtained from the image display device 23 is the height of the image data corresponding to the reflected images from the upper and lower portions of the rib structure 49. Assuming that the difference is H, (Equation 1) holds.

  When the width of the rib structure 49 is W and the width of the image data corresponding to the reflected image from the upper part of the rib structure 49 is I, (Equation 2) holds.

  In FIG. 9, after each image is taken into the image data analyzing device 22 via the first image data synthesizing device 21A and the second image data synthesizing device 21B, the processing until obtaining the shape data from the image data is shown in FIG. This is performed by the data analysis device 22. The obtained image data and the processing result by the image data analysis device 22 are displayed on the image display device 23.

  FIG. 10 is a diagram showing a processing flow for obtaining shape data from image data in the first embodiment.

  In FIG. 10, first, in step S <b> 7, a reflection image of the measurement light projected onto the inner surface of the cylinder 14 to be measured is displayed on the image display device 23.

  Next, in step S8, pixel coordinates having the maximum luminance value among the pixels in the vertical direction (screen vertical direction) of the image data are obtained for each pixel vertical column in order from the pixel vertical column at the screen edge. The position of the maximum value of the obtained pixel in the vertical direction is defined as coordinates (I, H).

  In step S9, the extracted luminance of the pixel having the maximum vertical luminance is compared with a preset threshold value.

  Thereafter, different processing is performed depending on the relationship between the extracted maximum luminance value and a preset threshold value.

  If the maximum luminance value is equal to or greater than the threshold value, the pixel having the maximum luminance value is set as the maximum value (I, H) of the vertical pixel row in step S10.

  Thereafter, in step S11, the position of the maximum value of the pixel (I, H) is calculated from the above-described calculation formulas (Expression 1) and (Expression 2), and the height of the point where the measurement light is actually projected. Convert to information (W, h).

  On the other hand, if the maximum luminance value is smaller than the threshold value, it is determined in step S12 that the reflected image of the measurement light has not been obtained in the pixel having the maximum luminance value, and is determined as a defect point where the in-cylinder shape data cannot be acquired.

  After performing Step S10 and Step S11 or Step S12, in Step S13, it is determined whether or not processing has been performed for all the vertical pixel columns of the obtained image. If all pixel columns have been processed, the process ends. If there is a pixel column that has not yet been processed, the process moves to the next pixel vertical column in step S14, and from step S8. Repeat the series of operations.

  In this way, when the processing is performed for all the vertical pixel columns of the acquired image data, the shape information of the location where the measurement light is projected can be obtained from the image data shown in FIG.

  Next, a method for measuring the shape of the entire surface in the cylinder 14 to be measured will be described.

  As described with reference to FIG. 2, since the cylinder 14 to be measured is fixed to the moving stage 24, the first measuring light 16 and the measuring beam 16 are operated by combining the linear movement mechanism and the rotational movement mechanism of the movement stage 24. The second measurement light 18 can be scanned over the entire inner surface of the cylinder.

  FIG. 11 is a diagram showing how to take coordinates on the inner surface of the cylinder 14 to be measured.

  FIG. 11A is a diagram showing coordinates on the inner surface of the cylinder 14 to be measured, and FIG. 11B is a schematic development view of the inner surface of the cylinder 14 to be measured in FIG.

  In FIG. 11A, in order to explain how to take coordinates on the inner surface of the cylinder 14 to be measured, a cutting line CC ′ is defined, and corner portions near the end of the cutting line CC ′ are represented by D, D ′, Let E, E ′. Here, assuming that D is the origin and the direction along the cutting line CC ′ is the X axis, and the direction along the edge of the cylinder from the angle D to the angle D ′ is the Y axis, as shown in FIG. In addition, the cylindrical inner surface, which is a cylindrical surface, can be represented by a plane defined by the X axis and the Y axis. D, D ′, E, E ′ shown in FIG. 11A and D, D ′, E, E ′ shown in FIG. 11B indicate the same positions.

  A method for obtaining the shape data of the entire inner surface of the cylinder 14 to be measured shown in FIG.

  The coordinate numerical value determined in advance by the image data analyzing device 22 is transmitted to the stage moving device 25, and the stage operating device 25 causes the moving stage 24 to rotate and move straight according to the instruction from the image data analyzing device 22. The image data analyzing device 22 is moved through the stage moving device 25 by the amount of movement of the moving stage 24 by the straight movement (corresponding to the movement in the X-axis direction in FIGS. 11A and 11B) and the rotating operation (FIG. a) and the rotation amount according to the operation in the Y-axis direction in FIG. For this reason, the shape data (W, h) of the inner surface of the measured cylinder 14 obtained through the processing flow shown in FIG. 10 and the position (X, Y) on the inner surface of the measured cylinder 14 can be associated with each other. The shape data (W, h, X, Y) of the entire 14 inner surface is obtained.

  Next, a method for acquiring the shape data of the inner surface of the cylinder 14 to be measured over the entire surface of the cylinder 14 to be measured using two sets of measurement systems will be described with reference to FIGS. 12 (a) and 12 (b). .

  FIG. 12 is a diagram showing a positional relationship between the measurement system and the structure of the object to be measured in the first embodiment.

  FIG. 12A is a diagram illustrating a case where measurement light is not visible by the light cutting method according to the first embodiment, and FIG. 12B is a diagram illustrating data interpolation of the light cutting method according to the first embodiment. is there.

  When the shape measurement in the cylinder 14 to be measured is measured by the above-described method, if only the result obtained from one measurement system is used, the incident angle of the measurement light and the structure on the inner surface of the cylinder 14 to be measured (rib, It is assumed that the line-shaped measurement light is hidden behind the structure and cannot be seen when viewed from the imaging optical system due to the position and angle relationship with the groove). For example, as shown in FIG. 12A, when the first linear measurement light 37 is projected onto the groove structure 54, the portion of the first linear measurement light 37 that is a broken line is a shadow of the groove structure 54. And cannot be observed from the imaging optical system. In such a case, the shape data of the part that cannot be observed becomes the above-described missing point, and the shape data cannot be measured.

  In such a case, the shape data of the shaded portion cannot be obtained by using only one measurement system. However, if the two sets of measurement systems arranged as shown in FIG. 2 have a structure in which the measurement light is incident from the opposite directions, the locations hidden in the structure when measured from the first measurement system are If the measurement is performed from the measurement system, it is possible to measure a portion that is hidden by the structure when measured from the first measurement system.

  FIG. 12B shows a flow in the case of measuring the shape data of the inner surface of the cylinder 14 to be measured using two measurement systems. Details of the shape data acquisition already described are omitted.

  By performing the above processing on the shape data for all the positions (X, Y) at which the shape data is acquired on the inner surface of the cylinder, the shape data is measured even if there are shapes such as ribs and grooves on the inner surface of the cylinder. 14 can be obtained over the entire inner surface.

  For example, as shown in FIG. 12B, only the cylinder inner surface shape data 55a (coarse broken line) can be obtained from the first measurement system for the cylinder inner surface shape data 55 to be measured, but the measurement is performed from the second measurement system. Then, cylinder inner surface shape data 55b (fine broken line) is obtained. When the processing for measuring the shape data of the cylinder inner surface of the cylinder 14 to be measured is performed using two measurement systems, the missing portions of the cylinder inner surface shape data 55a and the cylinder inner surface shape data 55b are compensated. The cylinder inner surface shape data 55 with no measurement data leakage is obtained.

  FIG. 13 shows a process flow diagram of data interpolation of the light section method in the first embodiment.

In FIG. 13, as shown in steps S <b> 15 and S <b> 16, the moving stage 24 is operated to acquire the shape data of the inner surface of the cylinder 14 to be measured. At this time, since the shape data of the inner surface of the cylinder 14 to be measured is acquired using two sets of measurement systems arranged as shown in FIG. 2, the shape data at the same position is acquired at the same time as the shape data is acquired by the first measurement system. Data can be acquired by the second measurement system. When the shape data acquisition operation is performed in this way, in-cylinder shape data (W, h ₁ , X, Y) obtained from the first measurement system and in-cylinder shape data obtained from the second measurement system (W, h ₂ , X, Y) are obtained simultaneously. As described above, depending on the shape of the inner surface of the cylinder, the shape data cannot be acquired in the first measurement system and h ₁ becomes a defect point, or the shape data cannot be acquired in the second measurement system. There are cases where ₂ is a missing point.

Next, in step S17, in-cylinder shape data (W, h ₁ , X, Y) obtained from the first measurement system for all positions (X, Y) at which shape data was acquired on the inner surface of the cylinder 14 to be measured. ) And in-cylinder shape data (W, h ₂ , X, Y) obtained from the second measurement system.

When h ₁ obtained from the first measurement system becomes a defect point, the process proceeds to step S18, and the in-cylinder shape data (W, h ₂ , X, Y) obtained from the second measurement system is used as the cylinder inner surface. It is adopted as shape data (W, h, X, Y) at the position.

If h ₂ obtained from the second measurement system becomes a defect point, the process proceeds to step S19, and the in-cylinder shape data (W, h ₁ , X, Y) obtained from the first measurement system is used as the cylinder inner surface. It is adopted as shape data (W, h, X, Y) at the position.

If both h ₁ obtained from the first measurement system and h ₂ obtained from the second measurement system are not defective, the process proceeds to step S20, and the shape data (W, h, X at the cylinder inner surface position) is advanced. Y) is obtained by averaging the height data h of Y) as the average value of the measurement result in the first measurement system and the measurement result in the second measurement system, h = (h ₁ + h ₂ ) / 2 adopt.

  As shown in step S21, the data obtained in this way is performed for all positions on the inner surface of the cylinder, and the measurement is completed.

It should be noted that if h ₁ obtained from the first measurement system and h ₂ obtained from the second measurement system are both missing points, they cannot be measured as they are, and in step S22, the light projecting device or imaging In some cases, measurement can be performed by performing measurement again after changing the position of at least one of the apparatuses.

In the flow of FIG. 13, if neither the in-cylinder shape data obtained from the first measurement system nor the in-cylinder shape data obtained from the second measurement system is a missing point, the weighting is performed. An averaging means other than a simple average such as an average may be taken, or height data of either h ₁ or h ₂ may be adopted as a representative.

  Next, a processing flow in the case where the conventional methods are performed together will be described.

  FIG. 14 is a diagram showing a measurement processing flow in the first embodiment.

  In FIG. 14, steps S23 to S27 are the same as steps S1 to S5 of FIG.

After performing Steps S23 to S27, in Step S28, the red shape data 57R (W _R , h _R ), the green shape data 57G ′ (W ′ _G , h ′ _G ) after offset correction, and the blue color after offset correction For all the shape data in the shape data 57B ′ (W ′ _B , h ′ _B ), the maximum value (Wmax) and the minimum value (Wmin) of the position W are searched.

Next, in step S29, it is checked whether height data (any one of h _R , h ′ _G , and h ′ _B ) that is not a missing point exists for all Ws that have acquired position data satisfying Wmin ≦ W ≦ Wmax. .

Here, when all of the height data h _R , h ′ _G , h ′ _B corresponding to the position W are missing points, the position W is defined as a missing point in step S30.

If there is one or more height data that is not a missing point among the height data h _R , h ′ _G , h ′ _B corresponding to the position W, the height data that is not the missing point in step S31. Is the height data of the position W.

  In step S32, this operation is performed for possible positions W (Wmin ≦ W ≦ Wmax) of all the acquired shape data.

  As described above, in step S33, the cylinder inner surface shape data of the measurement location 17 is obtained by superimposing the red shape data 57R, the green shape data 57G, and the blue shape data 57B, which are obtained separately at the time of data acquisition, with an offset added. It can be obtained as (W, h).

  Next, a method for correcting shape data in the in-cylinder rotation direction will be described.

  In FIG. 2, it has been described that the cylinder to be measured 14 is fixed to the moving stage 24 when acquiring the shape data of the cylinder inner surface. May not match due to errors.

  When the measured cylinder shaft 19 and the rotation axis of the moving stage 24 do not coincide (eccentricity occurs), the measured cylinder 14 is rotated around the measured cylinder axis 19 by the moving stage 24. The cylinder 14 rotates with eccentricity.

  If it does so, the influence of eccentricity will be added to the shape data of the rotation direction around the to-be-measured cylindrical axis | shaft 19 of the cylinder inner surface acquired by the 1st measurement system and the 2nd measurement system. The circumferential rotation angle of the measured cylindrical shaft 19 is Θ, the height of the inner surface of the cylinder at the position of the angle Θ when the measured rotating shaft 19 and the rotation axis of the moving stage 24 coincide with each other is H (Θ), and the measured The eccentricity of the cylindrical shaft 19 and the rotational axis of the moving stage 24 is A, the angular direction of the eccentricity is δ, and the height of the cylindrical inner surface including the influence of the eccentricity is H ′ (Θ). The influence of the eccentricity on the height measurement data can be expressed by (Equation 3) if the eccentricity amount A is an amount that is about 10% or less of the diameter of the cylinder 14 to be measured.

  Accordingly, by subtracting the second term of (Equation 3) from the height data in the rotational direction of the measured cylindrical shaft 19, the influence of the eccentricity of the rotational axes of the measured cylindrical shaft 19 and the moving stage 24 can be removed. Is possible. Here, A and δ can be calculated using mathematical processing (for example, Fourier transform).

  With the method and procedure as described above, the inner surface shape of the cylindrical part can be measured over the entire inner surface of the cylinder with a wide measurement range and high measurement accuracy in the height (thickness) direction.

  In the present embodiment, the surface shape is measured by performing the spectrum for each color. However, the reference for the spectrum here is not limited to the color, and the wavelength can also be used as the reference.

  The present invention is not limited to the tube shape of the test object, and can be applied to general three-dimensional shape measurement applications.

  According to the surface shape measuring method and measuring apparatus of the present invention, it is possible to measure the shape of the inner surface of the cylinder while maintaining the measurement accuracy over the inner peripheral front surface. It can be applied to thread shape evaluation measurement.

Schematic of in-cylinder shape measuring apparatus in Embodiment 1 FIG. 1 is a schematic cross-sectional view of the in-cylinder shape measuring apparatus shown in FIG. FIG. 2 is a schematic cross-sectional view of the in-cylinder shape measuring apparatus shown in FIG. 3 is a perspective view of the in-cylinder shape measuring apparatus of FIG. FIG. 2 is a schematic cross-sectional view of the in-cylinder shape measuring apparatus shown in FIG. Schematic diagram enlarging the vicinity of the measurement point 17 in FIG. (A) The figure which shows each measurement data of three wavelengths in Embodiment 1 (b) The figure which shows the coupling | bonding condition of three data (c) The figure which consolidated the coupling data into one data The figure which shows the processing flow of the measurement using the light source of 3 wavelengths in Embodiment 1. (A) Perspective view of projection image of line-shaped measurement light near measurement point 17 (b) Cross-sectional view of projection image of line-like measurement light near measurement point 17 The figure which shows the processing flow which acquires shape data from the image data in Embodiment 1. (A) A diagram showing coordinates on the inner surface of the cylinder 14 to be measured (A) The figure which shows the case where measurement light becomes invisible by the light cutting method in Embodiment 1, (b) The figure which shows the data complement of the light cutting method in Embodiment 1 The figure which shows the processing flow of the data complement of the light cutting method in Embodiment 1 The figure which shows the processing flow of the measurement in Embodiment 1. Sectional view of a conventional in-cylinder shape measuring device The figure which shows the measurement situation by the conventional cylinder shape measurement method

Explanation of symbols

12A First light projecting device 13A First imaging device 14 Cylinder to be measured 16 First measurement light 17 Measurement location 19 Cylinder shaft to be measured 21A First image data synthesis device 22 Image data analysis device 23 Image display device 25 Stage Moving device 26 First laser light source 27 First collimating lens 28 First cylindrical lens 29 First prism mirror 30 First polarizing element 31 First imaging element 32 First spectral prism 33 First imaging Lens 35 First polarizing filter 36 First measurement light source optical axis 37 First linear measurement light

Claims (4)

Irradiate the surface to be tested with optical axes parallel to each other with different wavelengths,
Spectroscopy the light reflected by the test surface for each wavelength,
After measuring the distance from the reference surface of the test surface for each wavelength based on the brightness of the dispersed light,
A surface shape measuring method, wherein the surface shape of the surface to be measured is measured by relatively shifting and superimposing the shape data for each wavelength by the measurement. 2. The surface shape measuring method according to claim 1, wherein the surface shape of the test surface is measured using light from two different directions across the test surface. By measuring the entire inner peripheral surface of the cylindrical test object by rotating the cylindrical test object,
2. A subtracting Acos (θ−δ) from the shape data, wherein θ is the rotation angle of the cylindrical specimen, A is the eccentricity of the rotation mechanism, and δ is the phase of the eccentricity. The surface shape measuring method according to claim 2. A surface shape measuring apparatus comprising a light projecting device, an image capturing device that captures light emitted from the light projecting device and reflected by a test surface, and a processing device that processes data obtained by the image capturing device,
The light projecting device includes a plurality of light sources having different wavelengths and an optical system that converts light from the plurality of light sources into phase-parallel optical axes,
The imaging apparatus includes: a plurality of imaging elements having different effective wavelengths; and an optical system that splits light incident on the imaging apparatus into a plurality of lights having different wavelengths.

Images