Parallel Flatness Measurement Method

Abstract

The present invention discloses a method for measuring the parallelism of a parallel flat plate, comprising the following steps: S1, collecting the self-interference fringes of the parallel flat plate through an interferometer; S2, preprocessing the collected self-interference fringes and filtering out the noise in the fringe pattern by using median filtering; S3, performing segmentation processing on the image by using the region growing algorithm; S4, calculating the parallelism by using the connected region algorithm to count the number of bright fringes. The present invention places the sample to be measured on the workbench and adjusts the optical path of the interferometer to collect the self-interference fringe pattern of the sample to be measured, and processes the self-interference fringe pattern. The region growing algorithm is used to perform image segmentation on the self-interference fringe pattern of the parallel flat plate, and then the connected region algorithm is used to count the number of bright fringes after image segmentation, so as to realize the measurement of the parallelism of the sample to be measured. This method is applicable to the measurement of parallel flat plates with good parallelism and has high measurement accuracy.

Description

Technical Field

[0001] The present invention relates to the field of measurement technology, and in particular, to a method for measuring the parallelism of a parallel flat plate. Background Art

[0002] In the manufacturing of optical and electronic devices, glass parallel flat plates composed of two planes or optical components equivalent to parallel flat plates are often used, and such optical components have high requirements for parallelism. Parallelism measurement is not only a key link in the manufacturing process, but also one of the means to ensure product quality, improve production efficiency and reliability. Therefore, it is very necessary to measure parallelism.

[0003] Currently, the common method for measuring parallelism is mainly to irradiate a beam of light onto the sample to be measured and then reflect to generate light spots, and judge the parallelism according to the positions of the light spots. When the two surfaces are parallel, the two light spots completely coincide; when the two surfaces are not parallel, the positions of the two light spots may intersect, be separated or be tangent. By using the method of image processing, the center distance between the two light spots is solved, and the angle between the two planes, that is, the parallelism, can be calculated according to the center distance between the two light spots. However, the method of measuring parallelism using the position of light spots is suitable for two planes with relatively large parallelism, and the measurement error is relatively large (the error is 2'), and there are limitations for the measurement of parallel flat plates with high requirements for parallelism.

[0004] In view of this, it is necessary to design a method for measuring the parallelism of a parallel flat plate to overcome the above defects. Summary of the Invention

[0005] The present invention aims to at least solve one of the technical problems existing in the prior art.

[0006] To this end, the present invention provides a method for measuring the parallelism of a parallel flat plate, which is applicable to the measurement of parallel flat plates with good parallelism and has high measurement accuracy.

[0007] A method for measuring the parallelism of a parallel flat plate according to the first aspect of the present invention includes the following steps:

[0008] S1. Collect self-interference fringes of the parallel flat plate through an interferometer;

[0009] S2. Preprocess the collected self-interference fringes and use median filtering to filter out the noise in the fringe pattern;

[0010] S3. Use the region growing algorithm to perform segmentation processing on the image;

[0011] S4. Use the connected region algorithm to count the number of bright fringes to calculate the parallelism.

[0012] Preferably, the self-interference fringes are obtained by detecting a sample to be measured through a measuring device, and the measuring device at least includes a laser, a beam expander, a diaphragm, a beam splitting prism, a collimating lens, an imaging lens, a CCD, and a computer.

[0013] Further preferably, the laser, the beam expander, the diaphragm, the beam splitting prism, and the collimating lens are all on the same horizontal plane. The sample to be measured is suitable for being placed on a workbench. If there is an inclination between the front and rear surfaces of the sample to be measured placed on the workbench, the self-interference fringes are formed on the CCD.

[0014] Preferably, when using median filtering to filter out the noise in the fringe pattern, the median filtering takes each pixel point in the self-interference fringe pattern as the center, selects a 3×3 filtering window, sorts the gray values of the 9 pixel points in the window, and takes the median value as the gray value of this pixel point.

[0015] Further preferably, when using the region growing algorithm to perform segmentation processing on the image, the following steps are adopted to extract the region of interest:

[0016] S31: Select appropriate initial seed points;

[0017] S32: Traverse the entire image to find the pixel points with the maximum and minimum gray values;

[0018] S33: Use the pixel points with the maximum and minimum gray values as the initial seed points for region growing to accurately capture the main part of the fringes;

[0019] S34: After obtaining the initial seed points, judge whether to merge the surrounding pixels into the current growing region according to the difference in gray values between pixels.

[0020] Preferably, to judge whether to merge the surrounding pixels into the current growing region, the following calculation formula is used for judgment:

[0021]

[0022] In the formula, is the gray value of the pixel point to be judged, is the pixel value of the initial seed point, is the set gray threshold.

[0023] Further preferably, when selecting the threshold, through experiments on multiple interference fringe patterns, observing the region growing effect, and gradually adjusting the threshold size, when the set threshold can completely separate the interference fringes, then this threshold is the optimal value.

[0024] Preferably, when performing region growing, the eight-neighborhood growing method is selected. The eight-neighborhood refers to the eight adjacent pixel points in the upper, lower, left, right, and diagonal directions of the current pixel point. The region growing rule based on the 8-neighborhood is as follows:

[0025] S1’ Determine the seed point, traverse the entire image, find the first pixel point that has not been grown, take this pixel point as the center, and judge whether the pixel values of the 8 neighborhoods of this pixel point meet the growth condition (the difference between the pixel value and the pixel value of the seed point is less than the set threshold). If it meets, mark this point as "1";

[0026] S2’ Take the newly grown point as the new seed point and repeat S1’;

[0027] S3’ Stop growing until the entire image is traversed.

[0028] Further preferably, after using the region growing algorithm to perform segmentation processing on the image, the self-interference fringes are divided into black and white parts. Using connected region analysis, that is, scanning the image, according to the eight-connected analysis, count the number of all connected regions marked as "1" in the image, which is the number of bright fringes in the interference fringe pattern. The parallelism is calculated by counting the number of interference fringes within the measurement aperture:

[0029] ,

[0030] In the formula, is the angle conversion value, with the unit of seconds, , is the thickness of the workpiece to be measured, is the refractive index of the sample to be measured, is the wavelength of the light source, m is the number of interference fringes, and D is the measurement aperture.

[0031] The beneficial effects of the present invention are as follows: By placing the sample to be measured on the workbench and adjusting the optical path of the interferometer to collect the self-interference fringe pattern of the sample to be measured, and through processing the self-interference fringe pattern, using the region growing algorithm to perform image segmentation on the self-interference fringe pattern of the parallel plate, and then using the connected region algorithm to count the number of bright fringes after image segmentation, thereby realizing the measurement of the parallelism of the sample to be measured. This method is applicable to the measurement of parallel plates with good parallelism and has high measurement accuracy.

[0032] Other features and advantages of the present invention will be described in the subsequent specification, and will be partially obvious from the specification, or understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification, claims, and drawings.

[0033] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides preferred embodiments in conjunction with the accompanying drawings and detailed descriptions are as follows. Description of the Drawings

[0034] The present invention will be further described below in conjunction with the drawings and embodiments.

[0035] Figure 1 It is a structural diagram of a measuring device for the parallelism measurement method of parallel plates of the present invention;

[0036] Figure 2 It is a schematic flowchart of the parallelism measurement method of parallel plates of the present invention;

[0037] Figure 3 It is a schematic flowchart of a method for segmenting an image using a region growing algorithm;

[0038] Figure 4 It is a schematic flowchart of a method based on the region growing rule of 8-neighborhood.

[0039] Reference Signs:

[0040] 1. Laser; 2. Beam expander; 3. Aperture; 4. Beam splitter prism; 5. Collimating lens; 6. Sample to be measured; 7. Imaging lens; 8. CCD; 9. Computer. Detailed Embodiments

[0041] Now, the present invention will be further described in detail with reference to the drawings. These drawings are all simplified schematic diagrams, only showing the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention. In the description of the present invention, it should be understood that unless otherwise specified, the meaning of "a plurality of" is two or more.

[0042] See Figures 1 to 2 , a method for measuring the parallelism of parallel plates in a specific embodiment of the present invention, includes the following steps:

[0043] S1. Collect self-interference fringes of the parallel plate through an interferometer;

[0044] S2. Pretreat the collected self-interference fringes and use median filtering to filter out the noise in the fringe pattern;

[0045] S3. Segment the image using a region growing algorithm;

[0046] S4. Use the connected region algorithm to count the number of bright fringes to calculate the parallelism.

[0047] Specifically, the self-interference fringes are obtained by the measurement device detecting the sample 6 to be measured. The measurement device at least includes a laser 1, a beam expander 2, a diaphragm 3, a beam splitting prism 4, a collimating lens 5, an imaging lens 7, a CCD 8, and a computer 9. Among them, the laser 1, the beam expander 2, the diaphragm 3, the beam splitting prism 4, and the collimating lens 5 are all on the same horizontal plane. The sample 6 to be measured is suitable to be placed on the workbench. If there is an inclination angle between the front and back surfaces of the sample 6 to be measured placed on the workbench, self-interference fringes will be formed on the CCD 8.

[0048] During measurement, the light emitted by the laser 1 is converged on the diaphragm 3 by the beam expander 2. After the light passes through the diaphragm 3 and reaches the beam splitting prism 4, the beam splitting prism 4 splits the light into two beams. One beam is used as the reference light, and the other beam is reflected by the sample 6 to be measured and then coincides with the reference light to form interference fringes. When performing parallelism detection, the part to be measured is placed on the workbench, and the self-interference fringes formed by the light reflected from the front and back surfaces of the part to be measured are used for detection.

[0049] If the front and back surfaces of the part to be measured are strictly parallel, an interference field with uniform brightness will be obtained, and no interference fringes can be seen on the CCD 8; if there is a certain small inclination angle between the front and back surfaces of the part to be measured, interference fringes can be seen on the CCD 8. Then, by counting the number of self-interference fringes within the measurement range, the parallelism measurement can be realized.

[0050] In addition, after the interferometer acquires the self-interference fringes of the parallel plate, median filtering can be used to filter out most of the noise in the interference fringe image. Median filtering takes each pixel point in the interference fringe image as the center, selects a 3×3 filtering window, sorts the gray values of the 9 pixel points in this window, and takes the median value as the gray value of this pixel point.

[0051] Then, referring to Figure 3 , when using the region growing algorithm to perform segmentation processing on the image, the following steps are adopted to extract the region of interest:

[0052] S31. Select appropriate initial seed points;

[0053] S32. Traverse the entire image to find the pixel points with the maximum and minimum gray values;

[0054] S33. Use the pixel points with the maximum and minimum gray values as the initial seed points for region growing to accurately capture the main part of the fringes;

[0055] S34. After obtaining the initial seed points, judge whether to merge the surrounding pixels into the current growing region according to the difference in gray values between pixels.

[0056] Among them, to judge whether to merge the surrounding pixels into the current growing region, the following calculation formula is used for judgment:

[0057]

[0058] In the formula, is the gray value of the pixel to be judged, is the pixel value of the initial seed point, is the set gray threshold.

[0059] When selecting the threshold, through experiments on multiple interference fringe images, observing the region growing effect, and gradually adjusting the threshold size. When the set threshold can completely separate the interference fringes, then the threshold at this time is the optimal value.

[0060] It should be noted that referring to Figure 4 , when performing region growing, the eight-neighborhood growing method is selected. The eight-neighborhood refers to the eight adjacent pixel points in the up, down, left, right, and diagonal directions of the current pixel point. The region growing rule based on the 8-neighborhood is as follows:

[0061] S1’ Determine the seed point, traverse the entire image, find the first pixel point that has not been grown, take this pixel point as the center, and judge whether the pixel values of the 8 neighborhoods of this pixel point meet the growth condition (the difference between the pixel value and the pixel value of the seed point is less than the set threshold). If it meets, mark this point as "1";

[0062] S2’ Take the newly grown point as the new seed point and repeat S1’;

[0063] S3’ Until the entire image is traversed and the growth stops.

[0064] After using the region growing algorithm to segment the image, the self-interference fringes are divided into black and white parts. Using connected region analysis, that is, scanning the image, according to the eight-connected analysis, count the number of all connected regions marked as "1" in the image, which is the number of bright fringes in the interference fringe image. The calculation of the parallelism is mainly obtained by counting the number of interference fringes within the measurement aperture:

[0065] ,

[0066] In the formula, is the angle conversion value, and the unit is seconds, 、 is the thickness of the part to be measured, is the refractive index of the sample 6 to be measured, is the wavelength of the light source, m is the number of interference fringes, and D is the measurement aperture.

[0067] Therefore, in summary, the test sample 6 is placed on the workbench through the above measurement steps, the optical path of the interferometer is adjusted, the self-interference fringe pattern of the test sample 6 is collected, and the self-interference fringe pattern is preprocessed. In the preprocessing, first, most of the noise in the interference fringe pattern can be filtered out by median filtering. Secondly, the self-interference fringe pattern is segmented by the region growing algorithm to separate the bright and dark fringes (by traversing the entire image, finding the pixel points with the maximum and minimum gray values, and using them as the initial seed points for region growing; through experiments on multiple interference fringe patterns, observing the region growing effect, gradually adjusting the threshold size, when the set threshold can completely separate the interference fringes, setting this value as the threshold, and then traversing the entire image, marking the grown seed points as "1" until all pixel points are traversed and the growth stops). Finally, the number of bright fringes is counted by using connected component analysis (the image after region growing is a binary image, and then the connected regions are divided using the 8-neighborhood connectivity rule, traversing all pixel points on the image, and counting the number of connected regions). The parallelism can be calculated from the number of fringes. This method is applicable to the measurement of parallel plates with good parallelism and has high measurement accuracy.

[0068] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0069] The above is based on the inspiration of the ideal embodiment of the present invention. Through the above description, relevant staff can completely make various changes and modifications without departing from the technical idea of this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for measuring the parallelism of a parallel plate, characterized in that: The following steps are involved: S1, collect parallel plate self-interference fringes through interferometer; S2, preprocessing the collected self-interference fringes and filtering out the noise in the fringe pattern by using median filtering; S3, using region growing algorithm to segment the image; S4, using the connected area algorithm to count the number of bright fringes to calculate the parallelism; After the image is segmented using the region growing algorithm, the self-interference fringes are divided into black and white parts. The image is scanned using the connected region analysis. According to the eight-connected analysis, the number of all connected regions marked as "1" in the image is counted, which is the number of bright fringes in the interference fringe diagram. The parallelism is calculated by counting the number of interference fringes within the measurement aperture: ， In the formula, is the angle conversion value, the unit is seconds, , is the thickness of the test piece, is the refractive index of the sample to be tested (6), is the wavelength of the light source, m is the number of interference fringes, and D is the measurement aperture.

2. The parallelism measurement method of parallel plates according to claim 1, characterized in that: The self-interference fringes are obtained by detecting a sample to be measured (6) via a measuring device, wherein the measuring device comprises at least a laser (1), a beam expander (2), an aperture (3), a beam splitter (4), a collimator (5), an imaging lens (7), a CCD (8) and a computer (9).

3. The parallelism measurement method of parallel plates according to claim 2, characterized in that: The laser (1), the beam expander (2), the aperture (3), the beam splitter (4) and the collimator (5) are all located on the same horizontal plane. The sample to be tested (6) is suitable for being placed on a workbench. If there is an inclination angle between the front and rear surfaces of the sample to be tested (6) placed on the workbench, the self-interference fringes are formed on the CCD.

4. The parallelism measurement method of parallel plates according to claim 3, characterized in that: When using median filtering to filter out noise in the fringe pattern, the median filtering takes each pixel in the self-interference fringe pattern as the center, selects a 3×3 filtering window, sorts the grayscale values ​​of the 9 pixels in the window, and takes the middle value as the grayscale value of the pixel.

5. The parallelism measurement method of parallel plates according to claim 4, characterized in that: When using the region growing algorithm to segment an image, the following steps are used to extract the region of interest: S31, selecting a suitable initial seed point; S32, traverse the entire image and find the pixel points with the maximum and minimum grayscale values; S33, taking the pixel points with the maximum and minimum grayscale values ​​as initial seed points for region growing, so as to accurately capture the main part of the stripes; S34: After obtaining the initial seed point, determine whether to merge the surrounding pixels into the current growth area based on the difference in grayscale values ​​between pixels.

6. The method for measuring the parallelism of parallel plates according to claim 5, characterized in that: To determine whether to merge surrounding pixels into the current growth area, the following calculation formula is used: In the formula, is the gray value of the pixel to be judged, is the initial seed point pixel value, is the set grayscale threshold.

7. The parallelism measurement method of parallel plates according to claim 6, characterized in that: When selecting the threshold, multiple interference fringe images are tested, the regional growth effect is observed, and the threshold value is gradually adjusted. When the set threshold can completely separate the interference fringes, the threshold at this time is the optimal value.

8. The parallelism measurement method of parallel plates according to claim 7, characterized in that: When performing region growing, the eight-neighborhood growing method is selected. The eight-neighborhood refers to the eight adjacent pixels above, below, left, right, and diagonally of the current pixel. The region growing rules based on the eight-neighborhood are as follows: S1' determines the seed point, traverses the entire image, finds the first pixel point that has not grown, takes the pixel point as the center, and determines whether the pixel values ​​of the eight neighborhoods of the pixel point meet the growth condition. The difference between the pixel value and the seed point pixel value is less than the set threshold. If it meets the condition, the point is marked as "1"; S2' uses the newly grown point as a new seed point and repeats S1'; S3' continues until the entire image is traversed and the growth is terminated.

Images