JP3161260U - Inspection device - Google Patents

Abstract

The present invention relates to a distortion inspection apparatus for inspecting distortion of a reflecting surface such as a mirror with high accuracy, and provides a distortion inspection apparatus capable of inspecting the distortion direction as well as the distortion amount. Projecting means for projecting a plurality of parallel line patterns onto an inspection object, reading means for reading a reflected parallel line pattern reflected from the inspection object by the parallel line patterns projected by the projection means, and the reading A data processing means for comparing and calculating the measurement pattern data 31 created based on the reflected parallel line pattern by the means and the predetermined reference pattern data 30, and the state of the inspection object is determined based on the result of the data processing means It has the determination means as described above. [Selection] Figure 4

Description

  The present invention relates to a distortion inspection apparatus for inspecting distortion of a reflecting surface such as a mirror.

Conventionally, methods for measuring optical distortion (distortion of a reflected image) in a reflecting mirror are stipulated in a reflection image distortion test (JIS R3220) of a mirror material and a distortion test (JIS D5705) of an automobile mirror.

The reflection image distortion test is performed by irradiating light from the projector from a direction of 45 degrees with respect to the surface of the specimen, and a parallel straight line black and white drawn on the seed plate in the projector on the projection surface provided in the reflected light path shape. This is a method of projecting stripes and measuring the stripe interval of the projected image. Therefore, in this method, the test can be performed only on the flat plate as the mirror material, and the distortion of the curved mirror cannot be detected.

On the other hand, in the distortion rate test, a concentric memory of a predetermined interval and an image of eight equal parts passing through the center are projected onto the mirror surface, and the image of the concentric memory of the mirror surface is projected by a camera or the like. The distortion rate is calculated based on the average radius of the concentric reflection image and the radius of the maximum or minimum reflection image. According to this method, since the concentric memory is used, the plane mirror and the curved mirror can be inspected.

In the distortion rate test, the average radius of the reflection image of the concentric circle and the radius of the maximum or minimum reflection image are obtained by measuring the interval between the intersection points of the concentric circle and the eight equal parts. Therefore, since only the displacement in the radiation direction is detected, the detection accuracy may differ with respect to the distortion direction. Further, since the interval (circumferential direction interval) in the peripheral portion is larger than that in the central portion, the inspection density is coarse, and there is a case where the distortion between the eight equal lines cannot be detected. is there.

According to Japanese Patent Laid-Open No. 9-175457, the distortion is defined by the difference between the lengths of two diagonal lines of the lattice pattern and the amount of angular distortion between each side of the lattice of the lattice pattern and the axis of the lattice pattern.

This method can also be measured when the distortion causes a diagonal difference, or when each side of the grid is distorted so that the angle is relative to the grid pattern, but the distortion is perpendicular to the grid. If it is out, it cannot be measured. Moreover, in the case of the measurement by these, the direction of distortion cannot be measured. In addition, since all distortion calculations are performed using measurement data, it is not possible to measure relative shifts such as offset distortion. Furthermore, since it is a lattice point, the restriction | limiting of the resolution that the inspection resolution is neglecting the detection accuracy below the distance between lattice points comes out.
Japanese Patent Laid-Open No. 9-175457

The present invention has been made to solve the above-described problems, and an object of the present invention is to a distortion inspection apparatus that inspects distortion of a reflecting surface of a mirror such as a plane mirror and a curved mirror with higher accuracy than before. Therefore, it is an object of the present invention to provide a distortion inspection apparatus that can inspect the direction of distortion as well as the amount of distortion, and can perform a fine inspection at the peripheral portion.

In order to solve the above problems, a distortion inspection apparatus according to the present invention includes a projection unit that projects a plurality of parallel line patterns on the inspection target, and is reflected from the inspection target by the parallel line pattern projected by the projection unit. A reading means for reading the reflected parallel line pattern, a data processing means for comparing and calculating measurement pattern data created based on the reflected parallel line pattern by the reading means and predetermined reference pattern data, and a result of the data processing means And determining means for determining the state of the inspection object based on the above.

 In the inspection apparatus according to the present invention, the parallel line pattern is composed of at least two types of parallel line patterns in different directions, and the reference pattern data corresponds to at least the different types of parallel line patterns. It comprises a plurality of reference pattern data, and further comprises identifying means for identifying different types of parallel line patterns in the parallel line patterns.

Further, in the inspection apparatus according to the present invention, the parallel line pattern includes at least two types of parallel line patterns in the same direction, and the reference pattern data corresponds to at least the different types of parallel line patterns. And further comprising identifying means for identifying different types of parallel line patterns in the parallel line patterns.

Further, in the inspection apparatus according to the present invention, the identification unit is reflected from the inspection target by the projection unit that projects different types of parallel line patterns at different timings, and the parallel line pattern projected by the projection unit. A reading means for sequentially reading different types of parallel line patterns, the measurement pattern data created based on the reflected parallel line patterns of the different types of parallel line patterns by the reading means, and the predetermined different types of parallel line patterns And a data processing means for comparing and calculating the reference pattern data, and a determination means for determining the state of the inspection object based on the result of the data processing means.

  Moreover, in the inspection apparatus according to the present invention, the identification unit is projected by the projection unit that projects different types of parallel line patterns in different colors (including wavelength ranges outside the visible light region), and the projection unit. It consists of reading means that identifies and reads different types of parallel line patterns reflected from the inspection object by the parallel line pattern, and measurement pattern data created based on the different types of parallel line patterns by the reading means, and predetermined Data processing means for comparing and calculating the reference pattern data of the different types of parallel line patterns, and determination means for determining the state of the inspection object based on the result of the data processing means .

  In the inspection apparatus according to the present invention, the identification unit includes a projection unit that projects different types of parallel line patterns and portions other than the parallel lines with different luminances, and a parallel line pattern projected by the projection unit. It comprises reading means that identifies and reads different types of parallel line patterns reflected from the inspection object, and measurement pattern data created based on the different types of parallel line patterns by the reading means, and the different types of predetermined types Data processing means for performing a comparison operation with reference pattern data of parallel line patterns, and determination means for determining the state of the inspection object based on the result of the data processing means.

  The inspection apparatus according to the present invention is characterized in that, in the inspection apparatus according to claims 2 to 6, the projection means finely translates at least one type of parallel line pattern to be projected.

The inspection apparatus according to the present invention is the inspection apparatus according to any one of claims 2 to 6, wherein the reading unit reads a parallel line pattern reflected from at least one type of inspection object while finely moving the parallel line pattern. It is characterized by.

Further, the inspection apparatus according to the present invention compares the amount of positional deviation between the measurement pattern data created based on each type of reflection parallel line pattern by the reading unit and each predetermined type of reference pattern data. A data processing means for creating a distortion vector by calculation and further generating a combined vector by combining each type of distortion vector, and a distortion vector or a combined distortion vector as a result of the comparison operation And determining means for determining the state of the inspection object based on the above.

In the inspection apparatus according to the present invention, in the distortion inspection apparatus according to claim 1 or claims 4 to 6, the determination unit includes at least a distortion vector or a maximum value, an average value, and a standard deviation of a distortion vector. Any one of the statistical numerical values is calculated.

Moreover, the inspection apparatus according to the present invention is the distortion inspection apparatus according to claim 1 or claims 4 to 6, wherein a plurality of adjacent distortion vectors or composite distortion vectors calculated by the determination means are determined according to a specific length. By performing clustering (grouping as data with a short distance) using thresholds, strain clusters (lumps of strain areas) are classified as aggregate clusters (lumps) that have a specific strain strength or higher, and the strain clusters are classified. It is characterized by determining whether each is convex or concave with respect to the reference surface.

Further, the inspection apparatus according to the present invention is the distortion inspection apparatus according to claim 1 or claims 4 to 6, wherein the determination unit calculates the statistical value of the distortion vector amount, and calculates the distortion from the distortion cluster. It is characterized in that the determination of the uneven area and the calculation of the intensity of the unevenness are performed, and it is determined whether or not at least one is larger or smaller than a preset threshold value corresponding to each, and the pass / fail is determined.

  The effect of the present invention is that a quantitative inspection can be applied to a distortion inspection of an inspection object made of, for example, a plane or curved mirror. According to the present invention, it is possible to quantitatively detect a distortion amount and a direction of a plane or curved mirror body, which could only be determined sensuously until now, with high accuracy. In addition, if the distortion vector introduced in the present invention is used, the magnitude and type of the distortion can be known visually and can be used effectively not only for product inspection but also for product development. In addition, by analyzing the inspection data, grasping the trend and taking measures, it is possible to prevent the occurrence of defects in advance and improve the quality and the yield.

Hereinafter, an embodiment of the present invention will be specifically described with reference to the drawings.

  As shown in FIG. 1, the distortion inspection apparatus 10 uses a display 15 as a parallel line pattern projector, and a liquid crystal display with particularly little distortion in the peripheral portion is desirable. Further, the half mirror 14 and the inspection object table 17 for allowing the mirror of the inspection object to be installed in a predetermined posture, the parallel line pattern is reflected on the half mirror 14 and is reflected on the inspection object 13. , And a camera 11 that captures the image. The camera 11 can be a CMOS or a CCD camera. The camera can be a two-dimensional image area sensor or a one-dimensional line sensor. In the case of a line sensor, a camera moving device 12 that can move linearly in a direction perpendicular to the scanning line of the line sensor is further required. In the case of a line sensor, high resolution can be acquired as in one line of 5000 pixels, but in order to obtain two-dimensional image data, the camera moving device 12 is required, and thus time for scanning is required. It becomes. This may be selected depending on the intended use. The camera 11 side is shielded from the half mirror 14 of the inspection apparatus so that external light does not enter.

The half mirror 14 is installed at a predetermined angle. In this embodiment, the display 15 is installed at a 45 degree posture. Also, the optical axis of the camera 11 and the half mirror 14
It is installed so that its normal vector matches.

  As shown in FIG. 2, the image processing apparatus 20 projects a parallel light emission pattern on the display 15, the light is half reflected by the half mirror 14, is reflected by the inspection object 13, and again the half mirror 14. Then, the reflected light is transmitted, and the reflected light travels straight and enters the optical axis of the camera. The image acquired by the camera is temporarily stored in the memory of the image processing apparatus, and after performing a predetermined calculation process, the calculation result is displayed on the result output monitor 22.

  Here, in the present embodiment, the installation position of the display 15 needs to pay attention to the luminance unevenness of the device. Since the brightness is usually brightest at the center, when the parallel line pattern 16 of the display 15 is projected onto the inspection object 13 via the half mirror 14, the center position of the inspection object 13 and the center position of the display 15 are matched. Install. In addition, when the luminance center is shifted, it is preferable to install the luminance centers so as to match. The luminance unevenness is preferably converted so that the whole becomes uniform when the optical system is determined by image processing. What is necessary is just to determine the fineness of the line of the parallel line pattern 16 with the precision which test | inspects the test target object 13. FIG. In addition, the display method of the inspection pattern on the display 15 can be displayed as it is when the graphic board of the personal computer 20 for image processing apparatus is a dual channel. Presentation software can be used for simple display. The display 15 uses a display 15 having a fine dot pitch when the distortion of the inspection object 13 is viewed in detail. In the case of a display of 1600 × 1200, an output of about 0.25 dot pitch is possible, and when a detection pattern is output in steps of about 10 dots, a line width of 2.5 mm is obtained.

For example, when the inspection target is a plane mirror, the image captured by irradiating the inspection target 13 with the parallel line pattern 16 as described above is as shown in FIG. In this example, the half mirror surface is set to 45 degrees with respect to the imaging element surface and the display display surface of the camera 11.
In FIG. 3, when it is assumed that the parallel line pattern 16 is an ideal inspection symmetry plane without distortion, the reference pattern data 30 is set horizontally at equal intervals. (In FIG. 3, there are six parallel straight lines provided at equal intervals.) In contrast to the reference pattern data 30, the actually detected measurement pattern data 31 approaches the reference pattern data 30 as long as there is no distortion. To go. A deviation between the reference pattern data 30 and the measurement data pattern 31 is represented by a distortion vector 32. In this case, the distortion vector 32 has a direction and a length (up and down). The distortion vector 32 is recorded with all pixels on the reference pattern data 30 (if the inspection is rough, skipping pixels) as a start point and a point on the detection line orthogonal to the straight line as an end point vector. When this vector is V, the absolute value | V | is the distortion length.
The total distortion amount D _S is expressed by the following mathematical formula.

In the case of a curved surface, the reference pattern data 30 in FIG. 3 is not a straight line but a curved line. This curve varies with the curved mirror R, camera, irradiation pattern, half mirror, and inspection symmetry position, so the degree of the curve changes. The reference pattern data 30 needs to be obtained. This relationship may be made possible by simulation.

Although the distortion vector can specify a certain distortion position even by processing only for the parallel line pattern 16 in the horizontal direction, the parallel line pattern 16 in the vertical direction is captured with respect to the parallel line pattern 16 in the case of more detailed inspection. There is a need.
The parallel line patterns 16 in the vertical direction and the horizontal direction (or the horizontal and vertical relations such as the first direction and the second direction are not necessarily required) are switched at a fixed interval by a projector including a display. Alternatively, it may be output simultaneously in the horizontal and vertical directions.
In the case of simultaneous output, in the case of the same color tone and luminance, it becomes a grid pattern and it is necessary to identify vertical and horizontal lines. In the case of image processing, in order to specify the position, processing different from other positions It is necessary to do. In the case of simultaneous detection, cross point detection or the like may be performed in addition to parallel line detection. Since the processing is complicated, in the case of a color image, different colors (wavelengths) are irradiated simultaneously, or in the case of monochrome, conflicting luminances, for example, the lowest luminance in the horizontal direction, the highest luminance in the vertical direction, and the background luminance are When the intermediate luminance is used, the processing becomes simple.

By combining these two measurement pattern data 31, a two-dimensional vector is completed. The distortion vectors are clustered according to the size. The vector group detects a region facing outward as shown in FIG. 5 as a convex region, and conversely detects a region facing inward as a concave region. The inspection object 13 can perform the same processing on a flat surface or a curved surface by calculating the combined reference pattern data 30 in the horizontal and vertical directions. Thereby, it is possible to determine which part of the inspection object 13 has an uneven surface with respect to the reference surface (ideal surface of the ideal inspection object 13). Thereby, it becomes possible to use for the judgment of the unevenness | corrugation of the foundation incorporated when the test object 13 was incorporated in the product.

The distortion vector is calculated over the entire surface of the inspection target 13. At this time, the statistical amount of calculation, the average distortion vector amount, the distortion vector standard deviation amount, and the maximum distortion vector are acquired, and a pass / fail decision is made by comparing with a preset threshold value.

Regarding the unevenness determination, when the distortion vectors are connected in groups in the same direction as shown in FIG. 6, it is determined that the case is gently inclined. Further, when facing inward, it is determined as a concave region, and when facing outward, it is determined as a convex region. In order to make such a determination, it is necessary to determine the direction of the vector of each cluster after performing clustering based on the distortion intensity in advance. Further, when the cluster vector covers a wide range, the distortion is moderate, and when the cluster vector is a narrow range, the distortion is abrupt. Put it in the criteria for judgment.

A threshold value serving as a reference is also set in advance for the unevenness determination, and the pass / fail determination is performed by comparing with the threshold value for each product.

  In the embodiment, the distortion inspection of the inspection target made up of a plane or curved mirror is described. However, the inspection target is not limited to the mirror and is an object having a reflectance that allows the parallel irradiation pattern to be detected by the camera. Can measure the distortion of flat surfaces and curved surfaces. Especially when the reflectance of the inspection object is low, it can be applied to various inspection objects by improving the detection ability such as increasing the amount of irradiation pattern light. In addition, it can be used widely. In addition, the surface shape of the object to be inspected does not need to be a flat surface or a single curved surface. Can be used industrially.

Schematic diagram of a distortion inspection apparatus according to an embodiment Flow chart of processing of distortion inspection device of one embodiment Distortion horizontal direction detection diagram of one embodiment Distortion horizontal direction vertical direction detection diagram of one embodiment Schematic diagram of distortion unevenness display of one embodiment Schematic diagram of distortion uneven area display of one embodiment

DESCRIPTION OF SYMBOLS 10 Distortion inspection apparatus 11 Camera 12 Camera moving apparatus 13 Inspection object 14 Half mirror 15 Display 16 Parallel irradiation line 17 Inspection object stand 20 Image processing apparatus 22 Result output monitor 30 Reference pattern data 31 Measurement pattern data 32 Distortion vector

Claims (12)

Projection means for projecting a plurality of parallel line patterns onto the inspection object, reading means for reading the reflected parallel line pattern reflected from the inspection object by the parallel line pattern projected by the projection means, and reflection parallel by the reading means Data processing means for comparing and calculating measurement pattern data created based on a line pattern and predetermined reference pattern data, and determination means for determining the state of the inspection object based on the result of the data processing means Distortion inspection device characterized by having. The parallel line pattern includes at least two types of parallel line patterns in different directions, and the reference pattern data includes a plurality of reference pattern data corresponding to at least the different types of parallel line patterns, and further includes the parallel patterns. 2. The distortion inspection apparatus according to claim 1, further comprising identification means for identifying different types of parallel line patterns in the line patterns. The parallel line pattern includes at least two types of parallel line patterns in the same direction, the reference pattern data includes at least a plurality of reference pattern data corresponding to the different types of parallel line patterns, and the parallel lines 2. The distortion inspection apparatus according to claim 1, further comprising identification means for identifying different types of parallel line patterns in the patterns. The identification unit sequentially projects different types of parallel line patterns reflected from the inspection target by the projection unit that projects different types of parallel line patterns at different timings, and the parallel line patterns projected by the projection unit. Data processing for comparing and calculating measurement pattern data created based on reflection parallel line patterns of different types of parallel line patterns by the reading unit and reference pattern data of the different types of parallel line patterns set in advance 4. The distortion inspection apparatus according to claim 2, further comprising a determination unit configured to determine a state of the inspection target based on a result of the data processing unit. The identification unit is reflected from the inspection target by a projection unit that projects different types of parallel line patterns in different tones (including wavelength ranges outside the visible light region), and the parallel line pattern projected by the projection unit, It comprises reading means for identifying and reading different types of parallel line patterns, measurement pattern data created based on the different types of parallel line patterns by the reading means, and reference pattern data of the predetermined different types of parallel line patterns 4. The distortion inspection apparatus according to claim 2, further comprising: a data processing means for performing a comparison operation between the data processing means and a determination means for determining the state of the inspection object based on a result of the data processing means.   Projection means for projecting different types of parallel line patterns and portions other than parallel lines with different luminances, and different types of parallel lines reflected from the inspection object by the parallel line patterns projected by the projection means. Comparing the measurement pattern data created based on different types of parallel line patterns by the reading unit with the reference pattern data of the different types of parallel line patterns determined in advance, comprising a reading unit that identifies and reads line patterns 4. The distortion inspection apparatus according to claim 2, further comprising: a data processing unit that performs determination, and a determination unit configured to determine a state of the inspection target based on a result of the data processing unit.   The distortion inspection apparatus according to claim 2, wherein the projection unit finely translates at least one type of parallel line pattern to be projected.   The distortion inspection apparatus according to claim 2, wherein the reading unit reads a parallel line pattern reflected from at least one type of inspection object while finely moving the pattern. A distortion vector is created by comparing the measurement pattern data created based on each type of reflection parallel line pattern by the reading means and the positional deviation amount of each predetermined type of reference pattern data, Further, a data processing means for creating a combined vector by combining each type of distortion vector, and the state of the inspection object is determined based on the distortion vector as a result of the comparison operation or the combined distortion vector. The distortion inspection apparatus according to claim 2, further comprising a determination unit configured as described above. 7. The distortion according to claim 1, wherein the determination unit calculates at least one statistical value of a maximum value, an average value, and a standard deviation of a distortion vector or a combined distortion vector. Inspection device. A portion having a specific strain intensity or higher by performing clustering (grouping as data with a short distance) on a plurality of adjacent strain vectors or composite strain vectors calculated by the determination means with a threshold value of a specific length 5. A strain cluster (strain region chunk) is classified as a set cluster (cluster), and each of the classified strain clusters is determined to be convex or concave with respect to a reference surface. The distortion inspection apparatus of -6. The determination means calculates the statistical value of the distortion vector amount, performs determination of the uneven area and calculation of the intensity of the unevenness from the distortion cluster, and at least one threshold value set in advance corresponding to each The distortion inspection apparatus according to claim 1, wherein it is determined whether it is larger or smaller than that, and a pass / fail determination is made.

Images