JP2012127675A - Method and apparatus for evaluating front-surface shape - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately evaluate a front-surface shape of an object to be evaluated by eliminating influences of a reflection image made by being reflected by a second front surface opposed to a first front surface.SOLUTION: An apparatus for evaluation is configured to capture a stripe pattern 1 projected on a front surface 3a of an object to be evaluated 3 such as a plate glass by a CCD camera 2. The stripe pattern 1 is provided on a light-emitting surface of a light source. The wavelength of light emitted from the light source is 200 to 380 nm. The image captured by the CCD camera 2 is taken into a computer 4 that functions as an arithmetic unit, and the computer 4 analyzes the image.

Description

  The present invention relates to a surface shape evaluation method and an evaluation apparatus.

  As a method of inspecting the surface shape of an object, the surface of the object to be inspected is irradiated with a stripe pattern having periodic brightness and darkness (for example, a pattern in which strip-shaped black pattern portions are arranged at regular intervals). There is a method for evaluating the surface shape of the object to be inspected based on the shift of the light-dark cycle in the reflected image formed by reflecting (see, for example, Patent Document 1). However, when such a method is applied to a transparent plate-like body such as a plate glass, not only a reflection image by the surface of the transparent plate-like body but also a reflection image by the back surface of the transparent plate-like body is captured simultaneously. Hereinafter, a reflection image from the surface of the object to be inspected such as a transparent plate-like body is referred to as a surface reflection image, and a reflection image from the back surface of the object to be inspected such as a transparent plate-like body is referred to as a back surface reflection image.

  FIG. 30 is an explanatory diagram showing a state in which a front surface reflection image and a back surface reflection image are formed simultaneously. As shown in FIG. 30, the light emitted from the point 5 on the stripe pattern is reflected by the surface 3 a of the object to be evaluated 3, passes through the optical path 8, passes through the lens center 30, and the imaging point 10 on the light receiving surface 7 of the camera. Is imaged. The light transmitted through the evaluation object 3 is reflected by the back surface 3 b of the evaluation object 3 and forms an image at the imaging point 11 on the light receiving surface 7 via the optical path 9 through the lens center 30.

  Here, depending on the period or width of the stripe pattern, the following problems may occur in the captured image signal. FIG. 31 is a waveform diagram schematically showing an example of an image signal output from the camera. FIG. 31A shows the image signal of the front surface reflection image, and FIG. 31B shows the image signal of the back surface reflection image. The low level indicates the level of the image signal based on the dark portion in the stripe pattern, and the high level indicates the level of the image signal based on the bright portion in the stripe pattern. If the width of the dark portion in the stripe pattern is wide, the width of the low level portion in the image signal also increases, and the low level in the image signal of the front surface reflection image and the low level in the image signal of the back surface reflection image may overlap. . Then, the image signal output from the camera becomes a signal as shown in FIG. 31C, and the signal is different from the image signal of the surface reflection image that is originally required (see FIG. 31A). Based on this, the surface shape is inspected.

  Further, as shown in FIG. 32, even if the width of the dark part in the stripe pattern is sufficiently narrow, the difference T between the position of the dark part in the image signal of the front surface reflection image and the position of the dark part in the image signal of the back surface reflection image is When it is close to an integer multiple of the light / dark cycle, an image signal as shown in FIG. 32C is output from the camera. 32A shows the image signal of the front surface reflection image, and FIG. 32B shows the image signal of the back surface reflection image. Also in this case, the surface shape is inspected based on a signal different from the image signal of the surface reflection image, which causes a problem that the accurate surface shape cannot be evaluated.

  There has been proposed a surface shape evaluation method in which the influence of the back surface reflection image is reduced so as not to lower the accuracy of the surface shape evaluation (see, for example, Patent Document 2). FIG. 33 is a schematic diagram showing an outline of an evaluation apparatus for evaluating the flatness of the surface of an object to be evaluated such as a plate glass described in Patent Document 2.

As shown in FIG. 33, the evaluation apparatus uses the stripe pattern 1 projected on the surface 3a of the object 3 to be evaluated such as a plate glass to be inspected placed on a mounting table (not shown) as an imaging means. The CCD camera 2 is configured to take an image. The stripe pattern 1 is provided on the light emitting surface of a light source (not shown). FIG. 34 is an explanatory diagram showing an example of the stripe pattern 1. In Figure 34, _{L 1} is the width of the dark portion, _{L 2} is the width of the bright portion. L ₁ + L ₂ corresponds to a light / dark cycle. When the stripe pattern 1 is realized by coloring the black portion on the transparent resin film, the bright portion corresponds to the transparent portion and the dark portion corresponds to the black portion.

  In the surface shape evaluation method described in Patent Document 2, as a first step, a stripe pattern determining step for determining a stripe pattern 1 suitable for the object to be evaluated 3 is executed, and then in a second step, in the first step. Using the determined stripe pattern 1, a surface shape inspection process is performed in which the surface shape of the evaluation object 3 is evaluated by image analysis based on the reflected image of the stripe pattern 1 by the evaluation object 3. In the second step, only the reflection image by the surface 3a of the evaluation object 3 is used among the reflection images by the surface 3a and the back surface 3b of the evaluation object 3 of the stripe pattern 1 determined in the first step.

  FIG. 35 is a flowchart illustrating an example of processing for determining a stripe pattern described in Patent Document 2. In the processing, first, a plurality of stripe patterns on which different patterns are printed for each sheet are prepared (step S31). That is, a plurality of stripe patterns having different stripe periods and widths are prepared. The stripe pattern is formed by printing a pattern on a transparent resin film by, for example, ink jet printing. Next, one stripe pattern is attached to the light source (step S32). Then, a reflection image of the stripe pattern by the object to be evaluated 3 is picked up by the CCD camera 2 (step S33).

  Next, the arithmetic device (for example, a computer) 4 inputs an image signal of an image captured by the CCD camera 2 and executes image analysis processing for analyzing the image signal in accordance with an image analysis processing program (step S34). In the image analysis processing, it is determined whether or not the output signal of the CCD camera 2, that is, the image signal input from the CCD camera 2 is in a state as shown in FIG. Specifically, it is determined whether or not the levels corresponding to the two dark portions are not superimposed. If the image signal does not indicate such a state (in the case of NG), another stripe pattern is attached to the light source, and the processes of steps S32 and S3 are executed again.

  When the arithmetic device 4 confirms that the levels corresponding to the two dark portions are not superimposed on the image signal, the stripe pattern attached to the light source at that time is used to evaluate the surface shape of the object 3 to be evaluated. Is determined as the stripe pattern 1 to be used (step S37). As described above, the stripe pattern 1 having a light and dark pattern set to be separated in the image signal obtained by the CCD camera 2 is determined as the stripe pattern 1 suitable for the evaluation of the surface shape of the object 3 to be evaluated. Is done.

  FIG. 36C is a waveform diagram showing an example of an output signal of the CCD camera 2 when the stripe pattern 1 determined in the first step is used. 36A shows an image signal of the front surface reflection image, and FIG. 36B shows an image signal of the back surface reflection image.

In the invention described in Patent Document 2, the CCD camera is optimized by optimizing the width of the dark part of the stripe pattern (corresponding to the signal widths W ₁ and W ₂ ) and the period of light and darkness (corresponding to the periods T ₁ and T ₂ ). 2 is adjusted so that the dark part in the front-surface reflection image and the dark part in the back-surface reflection image do not overlap (see FIG. 36C). Therefore, only the image signal of the surface reflection image can be easily extracted at the time of image analysis by the arithmetic device 4, and precise surface shape evaluation can be performed. That is, the surface shape can be accurately evaluated by removing the influence of the back reflection image with an inexpensive apparatus configuration.

Japanese Patent Laid-Open No. 11-148813 (paragraphs 0082-0083, FIG. 24) Japanese Patent Laying-Open No. 2005-345383 (paragraphs 0020-0024, FIG. 3)

  However, in the surface shape evaluation method described in Patent Document 2, in order to determine a stripe pattern that can appropriately separate the dark part based on the front-surface reflection image and the dark part based on the back-surface reflection image, at one time. As long as it cannot be determined, the operation of applying the stripe pattern to the light source must be performed a plurality of times, and it takes time to prepare for the actual surface shape inspection. In addition, referring to the example shown in FIG. 36C, the lower one of the two adjacent low-level portions corresponds to the dark part based on the front-surface reflection image, but the back-surface reflected light flux is large. In this case, the level difference between the two low levels becomes small, and it becomes difficult to extract only the image signal of the surface reflection image, and as a result, the accuracy of the evaluation of the surface shape may be lowered. Further, when the plate thickness of the object to be evaluated is as thin as 0.5 mm or less, adjacent stripes in the stripe pattern may be observed in an overlapping manner. Since the surface shape is inspected based on a signal different from the image signal of the surface reflection image, there arises a problem that the accurate surface shape cannot be evaluated.

  Therefore, the present invention provides a surface shape that can accurately evaluate the surface shape of the object to be evaluated without taking the effort of preparation work and without being affected by the reflected image of the second surface facing the first surface. An object is to provide an evaluation method and an evaluation apparatus.

  The surface shape evaluation method according to the present invention irradiates a first surface of an object to be evaluated with a pattern having periodic brightness and darkness, receives a pattern reflected from the first surface, and light and darkness in the pattern irradiated to the first surface. In order to detect the shift of the light / dark period in the light-receiving image with respect to the basic period of the light, the brightness of the region in the light-receiving image with respect to the basic period of light / dark in the pattern irradiated on the first surface is averaged, and A surface shape evaluation method for evaluating a surface shape of one surface, wherein a wavelength of light of the pattern irradiated on the first surface is 200 to 380 nm.

  Another aspect of the surface shape evaluation method according to the present invention is to form each reflection image at different points on the first surface of the object to be evaluated, and to determine the position, and the moving bright point with a known moving speed. Alternatively, each reflection image of the reference object that is an object point is observed through a camera, and a deviation amount of each observation reflection image with respect to each reflection image when the first surface is an ideal plane is obtained. Using the position information of the body and the lens center position information of the camera, the inclination of the undulation shape of the first surface is obtained and the inclination of the undulation shape is integrated by constraining that the first surface is almost flat. A surface shape evaluation method for evaluating a waviness shape, wherein a wavelength of light of the reflected image formed on the first surface is 200 to 380 nm.

  According to another aspect of the surface shape evaluation method of the present invention, each reflection image is formed at different points on the first surface of the object to be evaluated, the position can be specified, and the pattern has periodic brightness. Each reflection image of a reference object is observed through a camera, and the amount of deviation of each observation reflection image with respect to each reflection image when the first surface is an ideal plane is obtained. Then, the inclination of the undulation shape of the first surface is obtained using the lens center position information of the camera, and the undulation shape of the first surface is obtained by integrating the inclination of the first surface under the constraint that the first surface is substantially flat. In which the wavelength of the light of the reflected image formed on the first surface is 200 to 380 nm.

  The surface shape evaluation apparatus according to the present invention includes a light source that irradiates a first surface of an object to be evaluated with a pattern having periodic brightness and darkness, a light receiving means that receives a pattern reflected from the first surface, and a light and darkness in the pattern of the light source. Evaluation means for evaluating the surface shape of the first surface on the basis of the shift of the light-dark cycle in the light-receiving image by the light-receiving means with respect to the basic period, and the evaluation means is based on the basic period of light and darkness in the pattern irradiated on the first surface. Averaging means for averaging the brightness of the region in the corresponding received light image, and a signal for specifying the deformation location and the deformation amount of the surface shape on the first surface using the average signal output from the averaging means are output. A surface shape evaluation apparatus including a processing means for performing light emission, wherein the light emitted from the light source has a wavelength of 200 to 380 nm.

  In another aspect of the surface shape evaluation apparatus according to the present invention, each reflected image is formed at different points on the first surface of the object to be evaluated, the position can be specified, and the moving bright point whose movement speed is known. Or, a reference body that is an object point, a camera that obtains each reflection image by the first surface of the reference body, and a deviation amount of each reflection image obtained by the camera with respect to each reflection image when the first surface is an ideal plane. The slope of the undulation shape of the first surface is calculated using each displacement amount, the position information of the reference body, and the lens center position information of the camera, and the slope of the undulation shape is obtained with the constraint that the first surface is substantially flat. Is a surface shape evaluation apparatus including a calculation means for calculating the waviness shape of the first surface, wherein the wavelength of light forming the reflected image is 200 to 380 nm.

  The surface shape evaluation apparatus according to another aspect of the present invention is a pattern that forms each reflection image at different points on the first surface of the object to be evaluated, can specify the position, and has periodic light and dark. Calculating a deviation amount of each reflected image obtained by the camera with respect to each reference image when the reference surface, the camera that obtains each reflection image by the first surface of the reference body, and the first surface is an ideal plane; The slope of the undulation shape of the first surface is obtained using the amount, the position information of the reference body, and the lens center position information of the camera, and the slope of the undulation shape is integrated under the constraint that the first surface is substantially flat. An apparatus for evaluating a surface shape including a computing means for obtaining a waviness shape of one surface, wherein the wavelength of light forming the reflected image is 200 to 380 nm.

  In the present invention, since the wavelength of the light arriving at the first surface of the object to be evaluated is 200 to 380 nm, most of the light beam that has traveled into the object to be evaluated is absorbed. Further, the light flux of the reflected image from the first surface of the object to be evaluated is not absorbed by the optical member constituting the camera that receives the reflected image. Therefore, it is possible to accurately evaluate the surface shape of the object to be evaluated without being affected by the reflected image from the second surface facing the first surface.

Explanatory drawing which shows the relationship between the light which injects into a to-be-evaluated object, and the light which reflects the surface and back surface of a to-be-evaluated object. The wave form diagram which shows typically the example of the image signal which a camera outputs. The schematic diagram and perspective view which show the evaluation apparatus for evaluating the flatness of the surface of the to-be-evaluated object with a test object. Explanatory drawing for demonstrating the evaluation method of surface shape. Explanatory drawing which shows an example of a stripe pattern. The block diagram which shows the example of the functional block implement | achieved by the arithmetic unit. Explanatory drawing which shows the mode of the imaging surface corresponding to a stripe pattern. The flowchart which shows the evaluation method of the surface shape by this invention. The wave form diagram which shows an example of the image signal which shows the contrast of a light reception pattern, and an average signal. Explanatory drawing which shows the result of having measured the surface shape of the single side | surface with the contact-type measuring machine about one cross section of each sample. Explanatory drawing which shows the output of the averaging circuit about each sample. Explanatory drawing which shows the correlation of a shape value and a measured value. The block diagram which shows the other structural example of the evaluation apparatus of the surface shape by this invention. The flowchart which shows the schematic process of the evaluation method of the surface shape by this invention. Explanatory drawing which shows the measurement condition of the wave | undulation shape of a to-be-measured object. Explanatory drawing which shows the condition where the locus | trajectory of the observed reflected image precedes the reflected image by an ideal plane. Explanatory drawing which shows the condition where the locus | trajectory of the observed reflected image is delayed with respect to the reflected image by an ideal plane. Explanatory drawing which shows the relationship between the inclination | tilt degree and the inclination of a wave shape when the locus | trajectory of the observed reflected image precedes the reflected image by an ideal plane. Explanatory drawing which shows the relationship between the grade of a delay and the inclination of waviness shape when the locus | trajectory of the observed reflected image delays with respect to the reflected image by an ideal plane. Explanatory drawing which shows the definition of x-axis and z-axis. Explanatory drawing which shows the optical system used in simulation. Explanatory drawing which shows the waviness shape obtained by the simulation of an example of waviness shape and the evaluation method of the surface shape by this invention. Explanatory drawing which shows the waviness shape obtained by the simulation of the other example of waviness shape and the evaluation method of the surface shape by this invention. Explanatory drawing which shows one implementation example of the moving bright spot. Explanatory drawing which shows one Example of the measurement of the waviness shape over the whole surface of a to-be-measured object. The block diagram which shows the schematic structural example of the measuring apparatus of the waviness shape of other embodiment of this invention. (A) shows the result of measuring the surface shape of the object to be evaluated with a contact-type measuring machine, and (B) is the surface obtained by carrying out the method according to the present invention on the same object to be measured in Example 2. Explanatory drawing which shows a shape. (A) shows the result of measuring the surface shape of the object to be evaluated using a contact-type measuring machine and calculating the absolute value of the derivative, and (B) shows the deviation from the basic period when an image is taken in Example 3. Explanatory drawing which shows the absolute value of quantity. (A) shows the measurement result of the surface shape of the object to be evaluated by the contact-type measuring machine, and (B) shows that the result shown in FIG. 28 (B) is inverted for each minimum value and integrated. Explanatory drawing which shows the shape obtained by this. Explanatory drawing which shows a mode that a surface reflective image and a back surface reflected image are formed simultaneously. The wave form diagram which shows the example of the image signal which a camera outputs. The wave form diagram which shows the example of the image signal which a camera outputs. The schematic diagram which shows the conventional evaluation apparatus with a test object. Explanatory drawing which shows an example of a stripe pattern. The flowchart which shows an example of the process which determines a stripe pattern. The wave form diagram which shows the output signal example of a CCD camera.

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

FIG. 1 is an explanatory diagram showing a relationship between light incident on the evaluation object 3 and light reflected on the front surface 3a and the back surface 3b of the evaluation object 3. Here, a plate glass is taken as an example of the object to be evaluated 3. FIG. 1A shows an example in which the wavelength of light incident on the evaluation object 3 is 400 nm to 780 nm, that is, the light is visible light. The light beam incident on the object to be evaluated 3 is reflected by the surface 3a of the object to be evaluated 3 and reflected as light A, but part of the light travels inside the object to be evaluated 3, and many light beams are incident on the object to be evaluated 3. Reflected by the back surface 3b. Many light beams are emitted as light B from the surface 3 a of the object to be evaluated 3.
In addition, although plate glass is illustrated as the to-be-evaluated object 3 here, what is necessary is just the material which absorbs wavelength 200-380 nm. Moreover, the kind of plate glass is not specifically limited, Alkali glass and an alkali free glass are illustrated.

When the wavelength of the light arriving at the surface 3a of the object to be evaluated 3 is set to 380 nm or less, which is outside the wavelength region of visible light, the light proceeds to the inside of the object to be evaluated 3 as shown in FIG. Most of the luminous flux is absorbed by the evaluation object 3 and hardly arrives at the back surface 3b. Therefore, the amount of light B is small. In other words, the reflected light from the back surface 3 b of the object to be evaluated 3 is hardly emitted from the surface 3 a of the object to be evaluated 3.
Further, if the wavelength of the light arriving at the surface 3a of the object 3 to be evaluated is 200 nm or more, the light A of the reflected image from the surface 3a of the object 3 to be evaluated is transmitted by an optical member constituting the camera that receives the light A. Not absorbed.

  In FIG. 1, a thick object to be evaluated 3 is shown. If light A is light from a stripe pattern, the object to be evaluated 3 is a thin plate glass having a thickness of 0.5 mm or less. In the example shown in FIG. 1A, adjacent stripes in the stripe pattern may be observed in an overlapping manner. However, in the example shown in FIG. 1B, the reflected light on the back surface 3b of the object 3 to be evaluated does not substantially exit from the surface 3a of the object 3 to be evaluated. Alternatively, the influence of the reflected light on the back surface 3b of the object to be evaluated 3, that is, the influence (degree of superposition) of the back surface reflection image is reduced.

  In the present invention, by utilizing the relationship illustrated in FIG. 1B, the influence of the back reflection image is reduced, so that the surface shape can be accurately evaluated.

  FIG. 2 is a waveform diagram schematically showing an example of an image signal output by a camera that captures a stripe pattern reflected by the evaluation object 3 in the present invention. FIG. 2A shows the image signal of the front surface reflection image, and FIG. 2B shows the image signal of the back surface reflection image. The low level indicates the level of the image signal based on the dark portion in the stripe pattern, and the high level indicates the level of the image signal based on the bright portion in the stripe pattern.

FIG. 2C shows an image signal in which a front surface reflection image and a back surface reflection image are superimposed. FIG. 2 (B)
As shown in FIG. 2, since the difference between the high level and the low level in the back-surface reflection image becomes small, the waveform shown in FIG. 2C is close to the waveform shown in FIG. In the conventional example shown in FIG. 31, the difference between the waveform shown in FIG. 31C and the waveform shown in FIG.

  That is, in the present invention, even if the surface shape of the object to be evaluated 3 is evaluated using the image signal output from the camera as it is, the evaluation in which the influence of the back reflection image is eliminated is performed. Further, it is not necessary to perform a process for determining an appropriate stripe pattern as described in the conventional example.

  FIG. 3A is a schematic diagram showing an evaluation apparatus for evaluating the flatness of the surface of an object to be evaluated such as a plate glass as an example of the object to be evaluated together with the object to be evaluated. As shown in FIG. 3A, the evaluation apparatus is configured to capture an image of the stripe pattern 1 projected on the surface 3a of the evaluation object 3 such as a plate glass to be evaluated by a CCD camera 2 as an imaging means. Has been. The stripe pattern 1 is provided on the light emitting surface of a light source (not shown).

  If the wavelength of the light emitted from the light source is 200 to 380 nm, most of the light beam that has traveled into the object under evaluation 3 is absorbed by the object under evaluation 3 and hardly arrives at the back surface 3b. Further, the luminous flux of the reflected image from the surface 3a of the object to be evaluated 3 is not absorbed by the optical member constituting the CCD camera 2 that receives the reflected image. Therefore, it is possible to accurately evaluate the shape of the front surface 3a of the evaluation object 3 without being affected by the reflected image of the back surface 3b of the evaluation object 3.

  The light source is exemplified by a xenon lamp, a mercury lamp, a metal halide lamp, and the like, and its light emitting surface is formed of a member that transmits light having a wavelength of 200 to 380 nm. Further, the optical member such as the lens and filter of the CCD camera 2 and the stripe pattern are also configured by members that transmit light having a wavelength of 200 to 380 nm. Examples of the member that transmits light having a wavelength of 200 to 380 nm include quartz glass, ultraviolet transmissive acrylic (methacrylic resin), fluorine resin, and polystyrene resin.

  Although the plate | board thickness of the to-be-evaluated object 3 is not specifically limited, 0.5 mm or less is preferable and 0.3 mm or less is further more preferable. When the thickness of the object to be evaluated is more than 0.5 mm, the light beam is easily absorbed inside the object to be evaluated 3 or attenuated by multiple reflection. Therefore, the influence of the back surface 3b on the shape evaluation of the front surface 3a is small.

  An image picked up by the CCD camera 2 is taken into an arithmetic device 4 such as a personal computer, and image analysis is performed by the arithmetic device 4.

  Although the CCD camera 2 is illustrated here, any type of camera such as an area camera, a line camera, a video camera, or a still camera can be used instead of the CCD camera 2. In addition, any light receiving device may be used as long as a reflected image can be specified, such as an array of photosensors.

The optical axis of the CCD camera 2 and the normal line of the stripe pattern 1 (specifically, the plane on which the stripe pattern 1 exists) have the same angle θ ₁ with respect to the normal direction of the surface 3 a of the object 3 to be evaluated. In addition, a CCD camera 2 and a stripe pattern 1 are installed. The angle θ ₁ is preferably 45 °.

  FIG. 3B is a perspective view showing the evaluation apparatus shown in FIG. As shown in FIG. 3B, in the evaluation apparatus, the object to be evaluated 3 is irradiated with light from the light source 100 in which the stripe pattern 1 is provided on the light emitting surface.

Embodiment 1 FIG.
First, the premise of the surface shape evaluation method used in the present embodiment will be described. FIG. 4 is an explanatory diagram for explaining a surface shape evaluation method.

  Light to be evaluated 3 is irradiated on the object to be evaluated 3 from the light source 100, and reflected light from the surface 3 a of the object to be evaluated 3 is received by the image sensor of the CCD camera 2. “Light from the stripe pattern 1 from the light source 100” means light that has passed through the stripe pattern 1, but hereinafter, the light that has passed through the stripe pattern 1 is irradiated. "

  In the present embodiment, the reflected light from the back surface 3 b of the evaluation object 3 hardly enters the CCD camera 2.

  The arithmetic device 4 evaluates the surface shape based on an electrical signal (image signal) based on the received light. If the surface of the object to be evaluated 3 is completely flat, the contrast state of light and dark obtained from the image signal matches the known contrast state of the stripe pattern 1. FIG. 4A shows an example of a light reception pattern (image) of the CCD camera 2, a pixel arrangement in the image pickup device of the CCD camera 2, and an image signal indicating light and darkness from the image pickup device. When there is no surface deformation in the evaluation object 3, no contrast change appears as exemplified in the first three detection regions in the image signal (see FIG. 4A).

  However, a contrast change occurs in the image signal due to the reflected light from the surface deformation portion of the object 3 to be evaluated. In other words, the pattern of light received from the surface deformation portion of the object to be evaluated 3 is enlarged or reduced compared to the pattern from the non-deformation portion. For example, when the image is enlarged, the contrast increases, but the pattern may protrude from the detection area.

  Therefore, as shown in FIG. 4B, the original maximum value portion 120a and the minimum value portion 120b do not enter the same detection region, and the contrast in the detection region becomes small. The contrast is defined by (a−b) / (a + b) in each detection region. Further, when the image is reduced, it cannot be resolved by the image sensor, and the contrast becomes small as shown in FIG. Therefore, by comparing the difference between the white portion and the black portion in the image signal with the reference value, it is possible to evaluate whether the evaluation object 3 has surface deformation.

  Even when the pattern is not enlarged or reduced, the value of the bright part in the image signal may decrease or the value of the dark part may increase. Even in that case, it is possible to evaluate whether or not the object to be evaluated has surface deformation by comparing the difference between the bright part and the dark part in the detection area of a predetermined size with the reference value. In FIG. 4A, the halftone portion is a white portion with reduced brightness, and the contrast in the detection region is small in the detection region including that portion.

  However, for example, even if surface distortion that causes enlargement of the pattern occurs, the contrast in the detection region differs if the relationship between the enlargement mode and the detection region is different. As an example, the pattern shown in FIG. 4D and the pattern shown in FIG. 4E are enlarged in the same way, but since the way to enter the detection area is different, there is a difference in contrast in the detection area. There is. Originally, the same evaluation result should be obtained when the image signal is as shown in FIG. 4D and when the image signal is as shown in FIG. 4E. Attempting to evaluate the surface shape based solely on the differences may result in different results. That is, there is a limit in trying to evaluate the surface shape based only on the difference in contrast within the detection region.

  In the present embodiment, as described below, the surface shape can be evaluated with high accuracy based on the difference in contrast in the detection region, and as a result, the detectable range of distortion can be expanded.

FIG. 5 is an explanatory diagram showing an example of the stripe pattern 1. In FIG. 5, L ₁ indicates the width of the dark portion, and L ₂ indicates the width of the bright portion. L ₁ + L ₂ corresponds to a light / dark cycle (1 / f). The reciprocal of the period is the spatial frequency f of the pattern. Hereinafter, the spatial frequency f is called a fundamental frequency, and its reciprocal is called a fundamental period. When the stripe pattern 1 is realized by coloring the black portion of the transparent resin film, the bright portion corresponds to the transparent portion and the dark portion corresponds to the black portion. As an example, the period (1 / f) is about 1.0 mm, and the dark portion width L ₁ is about 100 μm.

  FIG. 6 is a block diagram illustrating an example of functional blocks realized by the arithmetic device 4. In the arithmetic device 4 shown in FIG. 3, the input circuit 41 inputs an image signal of a light receiving pattern indicating light and dark from the CCD camera 2 and stores it in the memory 42. The shading correction circuit 43 performs shading correction on the image signal in the memory 41 to remove the influence of the light amount distribution of the light source 100 on the image pickup by the CCD camera 2. The averaging circuit 44 performs an averaging filter process on the brightness of each pixel in each detection region after shading correction, and outputs an averaged value to the maximum value extraction circuit 45 and the minimum value extraction circuit 46.

  The detection area for averaging corresponds to the reciprocal (basic period) of the fundamental frequency of the pattern. Therefore, the averaging circuit 44 is configured by an integral filter or the like that averages the image signal for each basic period. The maximum value extraction circuit 45 outputs the maximum value of the output of the averaging circuit 44, and the minimum value extraction circuit 46 outputs the minimum value of the output of the averaging circuit 44. Then, the difference calculation circuit 47 outputs a signal indicating the difference between the output value of the maximum value extraction circuit 45 and the output value of the minimum value extraction circuit 46 and the position of the detection region. These maximum and minimum values are extracted for a region corresponding to the swell period assumed for the evaluation object 3. For example, the width of this region is 2 to 10 times the fundamental frequency.

  Next, the operation of the surface shape evaluation apparatus will be described with reference to the explanatory diagram of FIG. 7, the flowchart of FIG. 8, and the explanatory diagram of FIG.

  FIG. 7 is an explanatory diagram showing the state of the imaging surface of the CCD camera 2 corresponding to the stripe pattern 1. However, FIG. 7 shows only pixels in the rows E0 to E4 among the pixels on the imaging surface of the CCD camera 2. In the present embodiment, the optical system is set so that the basic period in the stripe pattern 1 corresponds to 5 pixels in the CCD camera 2. Therefore, here, the detection area for averaging is composed of 5 pixels.

  FIG. 8 is a flowchart showing the operation of the evaluation apparatus. An image (specifically, an image signal) captured by the CCD camera 2 is input to the memory 42 via the input circuit 41 in the arithmetic device 4. The shading correction circuit 43 obtains the average value of the image signals for each pixel in the several areas before and after the target detection area (including the target detection area) (step S11). For example, if the c region in the E1 column shown in FIG. 7 is used as the detection region, the average value of each pixel in the b, c, and d regions in the E1 column is obtained. By this processing, an image corresponding to the background light amount distribution image (shading image) around the detection region is obtained. Of course, the target area for taking the average value may be further expanded, and the areas in the columns before and after the E1 row may be included in the target area for taking the average value.

  The shading correction circuit 43 further divides the image signal of each pixel by the obtained average value (step S11). As a result, even if there is a difference in the amount of light emission depending on the position on the surface of the light source 100, the influence of the difference on imaging is reduced.

  The shading correction method is not limited to the method described above. For example, the following method can also be used.

(1) In the method described above, instead of taking the average of each pixel in each region, an intermediate value of the image signals of each pixel in each region is used.

(2) In the above-described method, instead of dividing each image signal by the obtained average value, the average value is subtracted from each image signal.

(3) Instead of obtaining an average value for shading correction for each detection region each time, a shading image is obtained in advance by a method such as directly receiving light from the light source 100. Then, when performing shading correction for each detection region, division processing and subtraction processing are performed using the image signal in the already obtained shading image.

  FIG. 9 is a waveform diagram showing an example of an image signal and an average signal indicating the brightness of the light receiving pattern. On the left side of FIGS. 9A, 9 </ b> B, and 9 </ b> C, an example of an image signal indicating the brightness of the light receiving pattern is shown. Since the shading correction circuit 43 is provided in the present embodiment, specifically, the output signal of the shading correction circuit 43 is shown. When the object to be evaluated 3 is flat, enlargement or reduction of the light receiving pattern does not occur, so that the period of the light receiving pattern coincides with the basic period as shown in FIG.

  If the surface of the object to be evaluated 3 has irregularities, the pattern is enlarged or reduced in the light receiving pattern by the CCD camera 2 due to refraction. For example, pattern enlargement occurs in transmitted light from a portion having a convex surface, and pattern reduction occurs in transmitted light from a portion in which the surface is concave. As a result, as shown on the left side of FIGS. 9B and 9C, the period of the light receiving pattern deviates from the basic period. Therefore, if it can be detected that the pattern period of the light receiving pattern is deviated from the basic period, it can be detected that the surface of the evaluation object 3 has irregularities.

  Therefore, the averaging circuit 43 performs a filtering process that averages the brightness of each pixel subjected to the shading correction in the detection region in order to detect a shift of the period of the light receiving pattern from the basic period (step S12). . In the present embodiment, since one detection region is constituted by five pixels corresponding to the basic period in the stripe pattern 1, the averaging circuit 43 averages the image signal for the five pixels. Examples of output signals from the averaging circuit 44 are shown on the right side of FIGS. When the period of the light receiving pattern coincides with the basic period, the averaging circuit 44 calculates a signal having a constant value as shown on the right side of FIG. 9A as a result of averaging each signal for one period. Output.

When there is a convex part on the surface of the object 3 to be evaluated, a signal in which a peak part and a valley part are spread appears in the input signal of the averaging circuit 44 as shown on the left side of FIG. Further, when the surface of the object to be evaluated 3 has a concave portion, a signal in which the peak portion and the valley portion are narrowed as shown on the left side of FIG. 9C appears. As shown in FIG. 9A, the averaging circuit 44 outputs a constant predetermined value when a signal whose peak portion and valley portion are not changed is input, but on the left side of FIG. 9B. When the signal when the received light pattern is enlarged is input, the averaging circuit 44 outputs a signal having irregularities with respect to a predetermined value, as shown on the right side of FIG. 9B. .

  In addition, even when a signal when the light receiving pattern is reduced as shown on the left side of FIG. 9C is input, as shown on the right side of FIG. Output the resulting signal. In the input signal of the averaging circuit 44, the greater the degree of expansion or narrowing of the crests and valleys, that is, the greater the degree of unevenness in the surface 3a of the object 3 to be evaluated, The degree of is great.

  In the present embodiment, a maximum value extraction circuit 45, a minimum value extraction circuit 46, and a difference calculation circuit 47 are provided in order to evaluate the output of the averaging circuit 44. The maximum value extraction circuit 45 receives the averaged signal in each detection region from the averaging circuit 44 and extracts the maximum value in that region (step S13). The maximum value extraction circuit 45 includes, for example, a filter that outputs the maximum value in one detection region, that is, the basic period. Further, the minimum value extraction circuit 46 receives the averaged signal in each detection region from the averaging circuit 44, and extracts the minimum value in that region (step S13). The minimum value extraction circuit 46 is constituted by, for example, a filter that outputs the minimum value in one detection region, that is, the basic period.

  The difference calculation circuit 47 calculates the difference between the output of the maximum value extraction circuit 45 and the output of the minimum value extraction circuit 46 (step S14). The target of the difference calculation is one detection area or a plurality of neighboring detection areas. For example, two or three adjacent detection areas. When two adjacent detection areas are targeted for difference calculation, the difference between the maximum value and the minimum value in the two detection areas is output from the difference calculation circuit 47.

  Of course, the size of the detection area handled by the averaging circuit 44 and the size of the range handled by the difference calculation circuit 47 may be different. For example, the difference calculation circuit 47 may obtain the difference between the maximum value and the minimum value for a range of 2 to 10 times the detection area handled by the averaging circuit 44. Further, the target area of the maximum value extraction circuit 45 and the minimum value extraction circuit 46 may be different from the detection area handled by the averaging circuit 44.

  And the difference signal which is a calculation result, and the position signal which shows the position of the detection area | region made into object at that time are output. When the surface 3a of the object to be evaluated 3 is not distorted, the output of the averaging circuit 44 is constant as shown in FIG. 9A, so that the difference signal output from the difference calculating circuit 47 shows the value 0. . However, when the surface 3a of the object to be evaluated 3 has irregularities, the difference signal indicates a value other than zero. In addition, the value of the difference signal corresponds to the degree of unevenness. Therefore, based on the value of the difference signal output from the difference calculation circuit 47 and the position signal, the degree of deformation of the surface 3a of the evaluation object 3 and the position of the deformation can be specified. In other words, the surface shape of the evaluation object 3 can be evaluated based on the value of the difference signal output from the difference calculation circuit 47 and the position signal (step S15).

  The evaluation of the surface shape in step S15 may be executed by an inspector, but the calculation device 4 includes an evaluation function for determining and outputting an evaluation result based on the value of the difference signal and the position signal. May be. The output of the evaluation result by the evaluation function is, for example, screen display or print output. In addition, the evaluation function compares the value of the difference signal with a predetermined threshold value, and if the threshold value is exceeded, it indicates that there is a defect on the surface 3a of the object 3 to be evaluated. It is a function that outputs together with data indicating the position.

  Further, for example, even in the case shown in FIGS. 4D and 4E, a range in which the difference calculation circuit 47 extends over a plurality of neighboring detection areas or exceeds the detection areas is a calculation target range. If the difference between the minimum value and the maximum value is output, the same difference signal is output in the cases shown in FIGS. 4D and 4E. Therefore, according to the method of the present invention, the detectable range of unevenness can be further expanded.

  Further, when the surface shape of the object to be evaluated 3 is changed but the light receiving pattern is not enlarged or reduced, and the value of the bright part or the value of the dark part in the image signal output from the CCD camera 2 decreases. In addition, irregularities appear in the signal averaged by the averaging circuit 44. Therefore, such a change in surface shape is also detected by the method of the present embodiment.

  Note that the averaging circuit 44, the maximum value extraction circuit 45, the minimum value extraction circuit 46, and the difference calculation circuit 47 sequentially process image signals corresponding to the entire area of the object 2 to be evaluated, stored in the memory 42. Alternatively, the two-dimensional image in the memory 42 may be collectively processed. Further, instead of the method of averaging the signals indicating the light and darkness in the detection region, the standard deviation of the light and darkness in the region may be output, or the surface of the object 2 to be evaluated is obtained by differentiating the averaged signal. The shape can also be evaluated.

[Example 1] Next, the result of comparison between the measurement result obtained by the contact-type measuring instrument and the evaluation by the evaluation apparatus according to the present embodiment will be described. FIGS. 10A to 10E show the results of measuring the surface shape of one side with respect to one cross section using a contact-type measuring machine, using five plate glasses as samples A to E, respectively. Each sample A to E is a plate glass having a thickness of 0.7 mm and a 300 mm square. Further, each sample A to E, which is an object to be evaluated, has a swell with a half pitch of about 10 to 15 mm.

  FIGS. 11A to 11E show the outputs of the averaging circuit 44 under the conditions that the optical axis is tilted by 30 ° with the samples A to E and the lens F25 is used in the CCD camera 2 and the aperture is F16. The stripe pattern 1 used was such that one pitch of black and white was 6 mm, one pixel was slightly over 1 mm on the sample surface, and one pitch was about 5 pixels on the image. Each sample A to E from which the results shown in FIGS. 11 (A) to (E) were obtained is the same as each sample A to E from which the results shown in FIGS. 10 (A) to (E) were obtained. is there.

  As described above, in FIGS. 11A to 11E, there are crests or troughs in the part where the period of the light receiving pattern is shifted from the basic period, that is, in the part corresponding to the part where the surface of the plate glass is uneven. Appears. Actually, depending on the relationship between the surface shape of the plate glass and the phase of the pattern, the concave portion on the surface of the plate glass corresponds to the peak portion or the valley portion in FIGS. However, there is a correlation between the absolute value of the difference between the peak portion and the valley portion and the degree of unevenness on the surface of the plate glass in both cases corresponding to the peak portion and the valley portion.

  Further, FIGS. 11A to 11E show the distribution of a portion in which the period of the light receiving pattern in one band-like region of the sample is shifted from the basic period, and the cross section is based on such distribution. The surface flatness of can be immediately evaluated. If the distribution of the entire cross section of the sample is output, the surface flatness of the entire surface of the sample can be immediately evaluated.

  FIG. 12 is an explanatory diagram showing the correlation between the shape value and the measured value. As the shape value, the maximum value (true value) of the differences between adjacent convex portions and concave portions in the surface shape of each sample is used. Moreover, in the evaluation apparatus of the embodiment described above, the maximum value of the difference signals output from the difference calculation circuit 47 for each sample is used as the measurement value. As shown in FIG. 12, the shape value and the measured value are well correlated. The correlation coefficient is 0.81.

Embodiment 2. FIG.
FIG. 13: is a block diagram which shows the structural example of 2nd Embodiment of the surface shape evaluation apparatus by this invention. In this embodiment, a waviness shape measuring device as a surface shape evaluating device is shown. As shown in FIG. 13, the measuring apparatus receives a light emitted from a moving bright spot 6 and reflected by the surface of a plate glass as an object to be evaluated 3 to form a reflected image, and a CCD camera 2. And an arithmetic unit 4 such as a computer that calculates the waviness shape by inputting the locus of the reflected image.

  Although the CCD camera 2 is illustrated here, any system such as an area camera, a line camera, a video camera, or a still camera can be used instead of the CCD camera 2. Further, any light receiving device may be used as long as it can identify the reflected image of the bright spot 6 such as a photo sensor array.

  FIG. 14 is a flowchart showing a schematic process of the surface shape evaluation method in the present embodiment. As shown in FIG. 14, in the surface shape evaluation method according to the present invention, first, the CCD camera 2 receives the reflected light from the surface of the object 3 to be evaluated at the moving bright spot 6 whose movement speed is known, and moves. A reflected image is obtained (step S21). When the surface of the object to be evaluated 3 has undulations, the locus of the reflected image captured by the CCD camera 2 deviates from the locus of the reflected image provided by an ideal plane having no undulation. That is, it precedes or lags behind the locus of the reflected image given by the ideal plane.

  Therefore, the inclination (differential value) of the swell shape of the surface of the object to be evaluated 3 is calculated from the deviation (preceding information or delay information) of the obtained reflected image with respect to the reflected image by the ideal plane (step S22). Then, with the constraint that the surface of the object to be evaluated 3 is substantially flat, a wavy shape is obtained by integral calculation (step S23).

  Next, the apparatus shown in FIG. 13 and the surface shape evaluation method shown in FIG. 14 will be described with reference to the explanatory diagrams of FIGS.

  FIG. 15 is an explanatory diagram illustrating a measurement state of the undulation shape of the object to be measured. As shown in FIG. 15, the reflected image 19 of the bright spot 6 is formed on the light receiving surface 7 by the light receiving element of the CCD camera 2. The light emitted from the bright spot 6 is reflected by the surface of the evaluation object 3 through the optical path 8 and reaches the light receiving surface 7. Here, the case where the bright spot 6 moves at a predetermined constant speed is taken as an example.

  As described above, when there is a undulation on the surface of the plate glass, the position of the obtained reflected image 19 at each time may precede or lag behind the position of the reflected image by the light reflected on the ideal surface. To do. FIG. 16 is an explanatory diagram showing a situation where the locus of the reflected image 19 precedes the reflected image 26 by an ideal plane. FIG. 17 is an explanatory diagram showing a situation in which the locus of the reflected image 19 is delayed with respect to the reflected image 26 by an ideal plane. In FIG. 16 and FIG. 17, an optical path 8A indicated by a solid line indicates an actual optical path. The light emitted from the bright spot 6 is reflected by the reflection point 12 on the surface of the plate glass, then reaches the light receiving surface 7 through the lens center 30 of the CCD camera 2 and forms a reflected image 19.

  An optical path 8B indicated by a broken line is a light receiving surface 7 through the lens center 30 of the CCD camera 2 after the light emitted from the bright spot 6 is reflected by the reflection point 13 on the surface of the plate glass when the plate glass surface is an ideal plane. The optical path to reach is shown. In this case, the reflected image 26 shown in FIGS. 16 and 17 is formed on the light receiving surface 7. However, the reflection image 26 is an image formed when it is assumed that the plate glass surface is an ideal plane, and is not actually formed.

  Since the bright spot 6 is moved at a predetermined speed, if the plate glass surface is an ideal surface, the reflected image 26 also moves at a predetermined speed on the light receiving surface 7 of the light receiving element. If the moving speed of the bright spot 6 is determined, the arithmetic unit 4 can recognize the position of the reflected image 26 by the light reflected on the ideal surface at each time. Since the arithmetic unit 4 can recognize the position of the reflected image 26 by the light reflected on the ideal surface, the amount of deviation (preceding amount or delay amount) of the actually obtained position of the reflected image 19 from the reflected image 26 is calculated. I can know.

  FIG. 18 is an explanatory diagram showing the relationship between the degree of preceding and the inclination (differential value) of the wavy shape when the locus of the reflected image 19 precedes the reflected image 26 by an ideal plane. FIG. 19 is an explanatory diagram showing the relationship between the degree of delay and the inclination (differential value) of the wavy shape when the locus of the reflected image 19 is delayed with respect to the reflected image 26 by an ideal plane.

  In FIG. 18 and FIG. 19, α is an angle formed by a vector extending from the lens center 30 to the optical path 8A when a vector extending from the lens center 30 to the optical path 8B is used as a reference. β is an angle formed by a vector extending vertically downward from the lens center 30 when a vector extending from the lens center 30 to the optical path 8B is used as a reference. γ is an angle formed by a vector extending from the bright spot 6 to the optical path 8A when a vector extending vertically downward from the bright spot 6 is used as a reference. Δ is an angle formed by the normal vector of the waviness surface at the reflection point 12 with respect to the normal vector of the reflection point 12 (the inclination of the normal vector).

  Any angle assumes a positive value when tilted counterclockwise from the reference vector. Therefore, in the situation shown in FIG. 18, α <0, δ <0, and in the situation shown in FIG. 19, α> 0, δ> 0.

  The slope δ of the normal vector is expressed as in equation (1). When the undulation shape is expressed by a function z = f (x), the inclination (differential value) = tan δ of the undulation shape is expressed by equation (2). The x axis and the z axis are taken as shown in FIG. 20, and the point where x = 0 is set at the left end of the surface of the evaluation object 3, for example.

  Therefore, the undulation shape z is obtained as shown in Equation (3). In the equation (3), C is an integral constant. Assuming a plate glass used as a glass substrate used for a flat panel display as the object to be evaluated 3, the surface shape of the plate glass may be almost flat although there may be fine undulations. Therefore, it can be considered that the relationship shown in the equation (4) holds. That is, the condition that the average value of the undulation on the surface of the plate glass is 0 is added. Then, the integration constant C in the equation (3) can be determined so as to satisfy the constraint condition in the equation (4). However, in equation (4), the ideal surface is a plane with z = 0.

  Specifically, the following processing is performed. The arithmetic device 4 inputs the reflected image 19 on the light receiving surface 7 at each time from the CCD camera 2 and obtains position information of the reflected image 19 at each time. Since the moving speed of the bright spot 6 is a predetermined constant speed, the position of the bright spot 6 at each time and the position information of the reflected image 26 by the ideal plane can also be recognized. The positions of the reflected images 19 and 26 on the light receiving surface 7 correspond to the positions of the reflection points 12 and 13 on the plate glass surface.

  The position of the lens center 30 is also determined. Since the position of the bright spot 6 at each time can be recognized, the positions of the reflection points 12 and 13 at each time can be determined from the positions of the reflected images 19 and 26 on the light receiving surface 7, and the position of the lens center 30 is also known. The arithmetic unit 4 can calculate the angles α, β, and γ at each time. Therefore, δ at each time can be calculated based on the equation (1). The position of the reflection point 12 at each time is the value of x in the equations (2) to (4).

Since δ at each time is calculated, the arithmetic device 4 can easily calculate the value of tan δ (= f ′ (x)) at each time. Assume that n f ′ (x) at each time are obtained as f ′ (x1), f ′ (x2), f ′ (x3),..., F ′ (xn), and Δ (x ) Is defined as follows.
Δ (x1) = (f ′ (x1) + f ′ (x2)) × (x2−x1) / 2
Δ (x2) = (f ′ (x2) + f ′ (x3)) × (x3−x2) / 2
・
・
Δ (x (n−1)) = (f ′ (x (n−1)) + f ′ (xn)) × (xn−x (n−1)) / 2

The waviness shape can be obtained by numerical integration of f ′ (x). Specifically, the arithmetic unit 4 calculates f (xn) = Δ (x1) + Δ (x2) +... + Δ (x (n−1))
To obtain the height of the undulation at each x.

The swell shape obtained in this way does not necessarily satisfy the equation (4), but the arithmetic unit 4 determines the integration constant C in the equation (3) as follows:
C = − (f (x1) + f (x2)) × (x2−x1) / 2
− (F (x2) + f (x3)) × (x3−x2) / 2
・
・
-(F (x (n-1)) + f (xn)) * (xn-x (n-1)) / 2
By defining as described above, a wavy shape satisfying the expression (4) can be obtained.

[Example 2] Next, an example is shown in which a plate glass is actually measured as an object to be evaluated by the method of the present embodiment. In this example, the optical system positional relationship shown in FIG. 21 was used. In FIG. 21, a is the distance between the end of the object to be evaluated 3 (in this example, a sheet glass) and a perpendicular drawn from the lens center 30 of the CCD camera 2, b is the width of the object to be evaluated 3, and c is a stripe. The distance between a point on the pattern and the lens center 30, h is the height of the lens center 30 from the ideal plane (the average plane of the undulation of the surface of the glass sheet).

In the simulation, the positional relationship among the bright spot 6, the ideal plane, and the lens center 30 is assumed to be known. Specifically, a = 450 mm, b = 250 mm, c = 1300 mm, and h = 28.43 mm.
Further, the waviness shape of the surface of the object 3 to be evaluated was expressed by a function represented by the equation (5) having amplitudes a1 and a2 and wavelengths L1 and L2.

  FIG. 22A is a waveform diagram showing a waviness shape when a1 = 0.00005 mm, L1 = 10 mm, a2 = 0.0 mm, and L2 = 20.0 mm in equation (5). 22B is obtained when the above-described measurement method is performed on the optical system having the positional relationship shown in FIG. 21 and the measurement object having the surface shape shown in FIG. It is a wave form diagram which shows a wave shape. It was confirmed that the undulation shape shown in FIG. 22 (B) substantially coincides with the undulation shape shown in FIG. 22 (A), that is, the undulation shape can be measured correctly by the method of the present invention.

  FIG. 23A is a waveform diagram showing a waviness shape when a1 = 0.00005 mm, L1 = 10 mm, a2 = 0.00005 mm, and L2 = 20.0 mm in the equation (5). FIG. 23B is obtained when the above-described measurement method is performed on the optical system having the positional relationship shown in FIG. 21 and the object to be measured having the surface shape shown in FIG. It is a wave form diagram which shows a wave shape. It was confirmed that the undulation shape shown in FIG. 23B substantially coincides with the undulation shape shown in FIG. 23A, that is, the surface shape can be measured correctly by the method according to the present invention.

  Further, in the surface shape evaluation method according to the present invention, the calculation amount of the arithmetic unit 4 is not so large because the angle calculation calculation based on the observed value and the addition / subtraction processing for integration may be performed.

  FIG. 24 is an explanatory diagram showing an example of realization of the moving bright spot 6. In this example, the light from the laser light source 21 is reflected by the movable mirror 22 so that the reflected light is applied to the screen 23. In this case, the irradiation point of the reflected light on the screen 23 can be used as the bright spot 6. Here, if the mirror 22 is moved so that the bright spot 6 moves at a constant speed on the screen 23, the bright spot 6 moving at a constant speed is realized. FIG. 24 shows a form in which the mirror 22 moves. Instead of the form in which the mirror 22 actually moves, the mirror movement is performed by using a rotating mirror such as a polygon mirror and a correction mirror. Can be easily realized.

  In the above-described embodiment, the bright spot 6 that moves at a constant speed is used as a reference body that can form the respective reflected images 19 at different points on the surface of the object 3 to be evaluated and whose position can be specified. The example used was described. However, if the moving speed is determined in advance, the present invention can be applied even if the moving speed of the bright spot 6 is not constant. In addition, the reference point is not limited to the bright spot 6 created as such, and a physical point (object point) to which the subject moves may be used.

  As the object point, for example, a point light source that moves itself or a movable point placed in an illumination environment can be used. When a point light source moving as an object point is used, the reflected image 19 becomes a bright point, and when a movable point placed under an illumination environment is used as an object point, the reflected image 19 becomes a dark point. Note that the bright spot and the object spot are not mathematical points, but are actually areas having a certain extent.

  In the above embodiment, the case of measuring the one-dimensional waviness shape has been described. However, when the waviness shape is measured over the entire surface of the object to be evaluated 3, for example, as shown in FIG. May be moved in a direction orthogonal to the optical path 8A. Then, the calculation device 4 performs the above-described calculation for each column on the surface of the object to be evaluated 3 (virtual columns shown between broken lines in FIG. 25). Although FIG. 25 shows the case where the DUT 1 moves, the bright spot 6 may be moved in a direction orthogonal to the optical path 8A.

  FIG. 26 is a block diagram showing a schematic configuration example of a surface shape evaluation apparatus according to another embodiment of the present invention. In the present embodiment, instead of the moving bright spot 6 as a reference body that can form each reflected image 19 at different points on the surface of the plate glass as the object to be evaluated 3 and whose position can be specified, A stripe pattern 61 having a known luminance pitch is used. Reflected images of the light from the high-intensity portions 62 of the stripe pattern 61 on the surface of the plate glass are captured by the CCD camera 2 and then input to the arithmetic unit 4.

  Since the luminance pitch is known in the stripe pattern 61, the arithmetic unit 4 can recognize each position of the reflected image 26 by the ideal plane on the light receiving surface 7 of the object 3 to be evaluated. The arithmetic device 4 calculates the waviness shape by the above-described calculation based on the deviation of the position of the actual reflected image 19 from each position of the reflected image 26 by the ideal plane. Thus, instead of the time information of the moving bright spot 6 (position information of the reflection images 19 and 26 at each time), the spatial information of the stripe pattern 61 (the reflection images 19 and 26 corresponding to the high-intensity portions 62). Even if position information) is used, the waviness shape on the surface of the plate glass can be measured.

[Example 3] Next, an example is shown in which a plate glass is actually measured as an object to be measured by the method of the present embodiment. In this example, the optical system positional relationship shown in FIG. 21 was used, and specifically, the apparatus shown in FIG. 26 was used. In FIG. 21, a is a distance between the end portion of the object to be evaluated 3 and a perpendicular drawn from the lens center 30, b is a width of the object to be measured 1, and c is a distance between the bright spot 6 and the lens center 30. , H is the height of the lens center 30 from the ideal plane (the average plane of the undulation of the surface of the object 3 to be evaluated).

  The positional relationship between the installation position of the stripe pattern 1, the ideal plane, and the lens center 30 is assumed to be known. Specifically, a = 225 mm, b = 300 mm, c = 750 mm, and h = 60 mm. In addition, one cycle of the high luminance portion and the low luminance portion of the used stripe pattern 1 was set to 1 mm. Therefore, one period is 0.5 mm on the center of the object 3 to be evaluated. In the CCD camera 2, a lens having an aperture F16 and a focal length of 55 mm was used. The pixel resolution in the stripe pattern 1 is about 0.09 mm, the pixel resolution is about 0.04 mm on the center of the object 3 to be evaluated, and one period in the stripe pattern 1 corresponds to 11 to 12 pixels.

  FIG. 27A shows the result of measuring the surface shape of the evaluation object 3 with a contact-type measuring machine. FIG. 27B shows a surface shape obtained by performing the method according to the present embodiment on the same object 3 to be evaluated. Since the surface shape shown in FIG. 27B substantially matches the surface shape shown in FIG. 27A, it can be seen that the surface shape can be measured correctly according to the present invention.

  In the example of measuring the surface shape of the sample of each shape shown in FIG. 10, the half pitch of each sample has a swell of about 10 to 15 mm, whereas the black and white pitch of the pattern is 6 mm. Measurements were made under relatively close conditions.

  Under such conditions, since there is a direct correlation between the deviation from the basic period and the surface shape, there is a high correlation in the level of unevenness. Therefore, in the example in which the surface shape of each shape sample shown in FIG. 10 is measured, one pixel is set to about 1 mm on the sample surface. However, the unevenness level is sufficiently evaluated even under such a low resolution condition. be able to.

  On the other hand, as in the above example (an example in which the waveform shown in FIG. 27B is obtained), compared with a half pitch of undulation such that one period on the center of the evaluation object 3 is 0.5 mm. If it is sufficiently small, it is not necessary to calculate the difference between the maximum value and the minimum value as described above, and the absolute value of the surface shape change (the absolute value of the slope of the wavy shape) is the amount of deviation from the basic period. There is also a correlation for the absolute value of. Therefore, in such a case, the surface shape can be calculated by integrating the absolute value of the deviation amount of the basic period calculated from the image by inverting it for each peak. Such a calculation method simplifies the calculation. Note that the positive / negative inversion is performed for each peak in order to return the data obtained as an absolute value to the data related to the original unevenness. Even if the square value of the deviation amount is used instead of the absolute value of the deviation amount, the same result is obtained although the value is different.

[Example 4] FIG. 28B shows the absolute value of the deviation from the basic period when an image is captured under the same conditions as in Example 3. FIG. However, in this example, the arithmetic unit 4 calculates a value averaged by performing a smoothing process in a basic cycle, and calculates an absolute value of a difference between the calculated value and a shading signal that is an average of five cycles. The absolute value of the deviation from the basic period. That is, the absolute value of the increase / decrease in the average signal is set as the amount of deviation from the basic period of the light / dark period.

  When a deviation from the basic period occurs due to the deformation of the evaluation object 3, there is a possibility that the smoothing width has many bright parts or many dark parts depending on the phase of the pattern. Therefore, the absolute value is taken so that it can be evaluated in either case. FIG. 28A shows the result of measuring the surface shape of the evaluation object 3 using a contact-type measuring machine and calculating the absolute value of the derivative.

  FIG. 29A shows the measurement result of the surface shape of the object 3 to be evaluated using a contact-type measuring machine. FIG. 29B shows a shape obtained by inverting the result shown in FIG. 29B for each minimum value and integrating the result. Since the two shapes are very similar, it can be seen that the surface shape can be evaluated by integrating the absolute value (or square value) of the deviation amount of the basic period by inverting the value for each peak.

  As described above, in the present embodiment, the surface shape of the object to be evaluated 3 is evaluated based on the deviation of the light / dark period from the basic period in the received light image by the reflected light from the surface 3a of the object to be evaluated 3. In addition, it is possible to more accurately evaluate the distortion in the surface shape in a short time. Further, by averaging the brightness and darkness of the region corresponding to the basic period in the received light image and evaluating the surface shape of the evaluation object 3 based on the averaged signal, the detectable range of distortion and the like can be expanded.

  Further, the wavelength of the light that irradiates the surface 3a of the object to be evaluated 3 is 200 to 380 nm that is absorbed inside the object to be evaluated 3 and not absorbed by the optical member that constitutes the CCD camera 2. Using the image signal from the CCD camera 2 in which the influence of the back reflection image is reduced, the surface shape of the evaluation object 3 can be accurately evaluated.

  The present invention is preferably applied when evaluating the flatness of the surface of an object to be evaluated.

DESCRIPTION OF SYMBOLS 1,61 Stripe pattern 2 CCD camera 3 Object to be evaluated 3a Front surface 3b Back surface 4 Arithmetic unit 5 Point on stripe pattern 6 Bright spot 7 Light receiving surface 8, 8A, 8B, 9 Optical path 10, 11 Imaging point 12, 13 Reflection point 19 , 26 Reflected image 21 Laser light source 22 Mirror 23 Screen 30 Lens center 31, 32 Reflection point 41 Input circuit 42 Memory 43 Shading correction circuit 44 Averaging circuit 45 Maximum value extraction circuit 46 Minimum value extraction circuit 47 Difference calculation circuit 62 High luminance portion 100 light sources

Claims (9)

Irradiating the first surface of the object to be evaluated with a pattern having periodic brightness,
Receiving a pattern reflected from the first surface;
In order to detect a shift of the light and dark period in the light receiving image with respect to the basic period of light and darkness in the pattern irradiated on the first surface, the brightness and darkness of the region in the light receiving image with respect to the basic period of light and darkness in the pattern irradiated on the first surface is determined. Average,
A surface shape evaluation method for evaluating the surface shape of the first surface based on an averaged signal,
The method for evaluating a surface shape, wherein a wavelength of light of the pattern irradiated on the first surface is 200 to 380 nm.   The surface shape evaluation method according to claim 1, wherein a shift portion and a shift amount of a light / dark cycle in a received light image are measured based on the averaged signal, and the surface shape of the object to be evaluated is evaluated based on the measurement result.   3. The surface shape evaluation method according to claim 1, wherein the surface shape of the object to be evaluated is evaluated based on a difference between a portion having a large amplitude in the averaged signal and a portion having a small amplitude in the vicinity thereof. The surface shape evaluation method according to claim 3, wherein an absolute value or a square value of an increase / decrease in an averaged signal is used as the shift amount of the light / dark cycle. Each reflected image of a reference object, which is a moving bright point or an object point, which is capable of specifying a position and whose movement speed is known, is formed by forming a reflected image at a different point on the first surface of the object to be evaluated. Through
Obtaining a shift amount of each observed reflected image with respect to each reflected image when the first surface is an ideal plane;
Using each displacement amount, the position information of the reference body and the lens center position information of the camera, the inclination of the undulation shape of the first surface is obtained,
A surface shape evaluation method for evaluating the undulation shape of the first surface by integrating the inclination of the undulation shape, with the first surface being substantially flat as a constraint.
The method for evaluating a surface shape, wherein a wavelength of light of the pattern irradiated on the first surface is 200 to 380 nm. Observation of each reflection image of the reference body, which is a pattern that forms a reflection image at different points on the first surface of the object to be evaluated, has a position that can be specified, and has periodic brightness and darkness. And
A displacement amount of each observation reflection image with respect to each reflection image when the first surface is an ideal plane is obtained, and the displacement amount, the position information of the reference body, and the lens center position information of the camera are used. Find the slope of the swell shape of the first surface,
A surface shape evaluation method for evaluating the waviness shape of the first surface by integrating the slope of the first surface, with the first surface being substantially flat as a constraint.
The method of evaluating a surface shape, wherein a wavelength of light of a reflected image formed on the first surface is 200 to 380 nm. A light source that irradiates the first surface of the object to be evaluated with a pattern having periodic brightness and darkness;
A light receiving means for receiving a pattern reflected from the first surface;
Evaluation means for evaluating the surface shape of the first surface based on a shift of a light-dark cycle in a light-receiving image by the light-receiving means with respect to a basic period of light-dark in the light source pattern,
The evaluation means uses an averaging means for averaging the brightness of regions in the received light image corresponding to the basic period of light and darkness in the pattern irradiated on the first surface, and an average signal output by the averaging means A surface shape evaluation apparatus including processing means for outputting a signal for specifying a deformation portion and a deformation amount of the surface shape on the first surface,
The surface shape evaluation apparatus, wherein the light emitted from the light source has a wavelength of 200 to 380 nm. A reference body that forms each reflection image at different points on the first surface of the object to be evaluated, is capable of specifying a position, and is a moving bright point or object point with a known moving speed;
A camera for obtaining each reflected image by the first surface of the reference body;
A deviation amount of each reflected image obtained by the camera with respect to each reflected image when the first surface is an ideal plane is calculated, and each deviation amount, position information of the reference body, and lens center position information of the camera And calculating means for obtaining the undulation shape of the first surface by integrating the undulation of the undulation shape on the condition that the inclination of the undulation shape of the first surface is obtained using A surface shape evaluation device including:
The surface shape evaluation apparatus, wherein a wavelength of light forming the reflection image is 200 to 380 nm. A reference body that forms each reflection image at different points on the first surface of the object to be evaluated and is capable of specifying a position and is a pattern having periodic brightness and darkness;
A camera for obtaining each reflected image by the first surface of the reference body;
A deviation amount of each reflected image obtained by the camera with respect to each reflected image when the first surface is an ideal plane is calculated, and each deviation amount, position information of the reference body, and lens center position information of the camera And calculating means for obtaining the undulation shape of the first surface by integrating the undulation of the undulation shape on the condition that the inclination of the undulation shape of the first surface is obtained using A surface shape evaluation device including:
The surface shape evaluation apparatus, wherein a wavelength of light forming the reflection image is 200 to 380 nm.

Images