CN116245879A - Glass substrate flatness evaluation method and system - Google Patents
Glass substrate flatness evaluation method and system Download PDFInfo
- Publication number
- CN116245879A CN116245879A CN202310505983.3A CN202310505983A CN116245879A CN 116245879 A CN116245879 A CN 116245879A CN 202310505983 A CN202310505983 A CN 202310505983A CN 116245879 A CN116245879 A CN 116245879A
- Authority
- CN
- China
- Prior art keywords
- glass substrate
- image
- surface height
- flatness
- height image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000011521 glass Substances 0.000 title claims abstract description 230
- 239000000758 substrate Substances 0.000 title claims abstract description 226
- 238000011156 evaluation Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000001914 filtration Methods 0.000 claims abstract description 21
- 230000011218 segmentation Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000005452 bending Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 5
- 238000010191 image analysis Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 101150050759 outI gene Proteins 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 11
- 238000004458 analytical method Methods 0.000 abstract description 6
- 238000005259 measurement Methods 0.000 abstract description 6
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000011326 mechanical measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Length Measuring Devices With Unspecified Measuring Means (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for evaluating flatness of a glass substrate, comprising the following steps: s1: collecting a glass substrate surface height image, denoising the image, and obtaining a denoised glass substrate surface height image; dividing the denoised glass substrate surface height image, extracting pixels belonging to the glass substrate area, and dividing the glass substrate from the image; extracting features from the segmented glass substrate region, the features including tortuosity, and sphericity; performing low-pass filtering on the segmented glass substrate region based on discrete Fourier transform to obtain a smoothed glass substrate surface height image, and calculating average value, variance, skewness and kurtosis information of the smoothed glass substrate surface height image; comparing the results obtained in S3 and S4 with the standard glass substrate, and evaluating the flatness of the glass substrate. By adopting the digital image processing and mathematical analysis method, the automatic evaluation of the flatness of the glass substrate is realized, and the measurement efficiency and the evaluation accuracy are improved.
Description
Technical Field
The invention relates to the technical field of glass detection, in particular to a method and a system for evaluating flatness of a glass substrate.
Background
Evaluation of flatness of a glass substrate is an important element in an industrial manufacturing process. The flatness of the glass substrate plays a vital role in the precise manufacturing process, and the glass substrate is used as a high-precision processing master plate in the field of chip processing and the like, and the higher the flatness is, the more precise the processing is, so that the glass substrates with different flatness also correspond to different grades of selling prices. Currently, the commonly used methods for evaluating the flatness of a glass substrate mainly comprise two types of mechanical measurement and optical measurement. However, the conventional evaluation method often requires a professional technician to perform the operation and judgment, and is easily interfered by human factors, so that the evaluation result is unstable and unreliable. Second, because these methods require testing of the actual samples, the number of samples is limited and rapid and efficient evaluation of large batches of samples may not be possible. Meanwhile, the methods require specialized instruments and technicians, so that the evaluation cost is high, and the methods are not suitable for real-time monitoring and control in the mass production process. According to the glass substrate flatness evaluation method, flatness evaluation can be automatically carried out by adopting the digital image processing and mathematical analysis methods, so that manual intervention is reduced, and the measurement efficiency is greatly improved. Meanwhile, a plurality of characteristic factors are comprehensively considered, so that a more comprehensive and accurate evaluation result can be obtained.
Disclosure of Invention
In view of the above, the present invention provides a method and a system for evaluating flatness of a glass substrate, and aims to provide a method for evaluating flatness of a glass substrate based on image processing and mathematical analysis, so as to solve the problems of high cost, complex operation, low precision, etc. existing in the existing methods. By adopting the digital image processing and mathematical analysis method, the automatic evaluation of the flatness of the glass substrate is realized, and the measurement efficiency and the evaluation accuracy are improved. Meanwhile, by comprehensively considering a plurality of characteristic factors, a more comprehensive and accurate evaluation result of the flatness of the glass substrate is obtained.
The invention provides a glass substrate flatness evaluation method, which comprises the following steps:
s1: collecting a glass substrate surface height image, denoising the image, and obtaining a denoised glass substrate surface height image;
s2: dividing the denoised glass substrate surface height image, extracting pixels belonging to the glass substrate, and dividing a glass substrate area from the image;
s3: extracting features from the segmented glass substrate region, the features including tortuosity, and sphericity;
s4: performing low-pass filtering on the segmented glass substrate region based on discrete Fourier transform to obtain a smoothed glass substrate surface height image, and calculating average value, variance, skewness and kurtosis information of the smoothed glass substrate surface height image;
s5: comparing the results obtained in S3 and S4 with the standard glass substrate, and evaluating the flatness of the glass substrate.
As a further improvement of the present invention:
optionally, the step S1 of collecting a surface height image of the glass substrate, denoising the image to obtain a denoised surface height image of the glass substrate, and the method includes:
the method comprises the steps of placing a glass substrate needing flatness evaluation on a platform, measuring heights of different positions of the platform by using laser, and recording the heights of each point to form a surface height image, wherein the distance from the surface of the glass substrate to a reference surface is recorded by each pixel of the surface height image, and the reference surface is a plane for placing the glass substrate. And denoising the image by utilizing improved median filtering to obtain a more accurate glass substrate surface height image. The improved median filtering includes:
s11: marking potential noise points:
wherein ,
is->
Window->
Center pixel value of (2); />
and />
Representing the maximum and minimum values within the window, respectively. If->
Or->
Mark->
As potential noise points, otherwise
Is a non-noise point;
s12: reconfirming the potential noise point by using a large window, if the potential noise point is still the noise point, replacing the pixel value by using a median value in the window, otherwise, reserving the pixel value:
wherein ,
is->
Window->
Center pixel value of (2); />
The function expression is->
Median of all pixel values in (a);
s13: and (3) performing S11 and S12 operation on each pixel of the glass substrate surface height image until the whole image is traversed, and obtaining the denoised glass substrate surface height image.
Optionally, in the step S2, the denoising method further includes segmenting the surface height image of the glass substrate, extracting pixels belonging to the glass substrate, and segmenting the glass substrate region from the image, including:
s21: selecting a segmentation threshold t, classifying the denoised glass substrate surface height image in the step S1 into two categories of background and glass substrate, and calculating probability of each category:
wherein ,
l is the maximum value of the surface height image; />
，/>
The pixel number is the pixel number of the image value q, and N is the total pixel number of the image; />
and />
Representing probabilities of pixels belonging to the background and the glass substrate class in the glass surface height image respectively;
s22: calculating the average value of pixels in the two categories of the background and the glass substrate, and calculating the inter-category variance:
the average value calculation method of the pixels of the background and glass substrate class is as follows:
the average value calculation mode of the surface height image of the whole glass substrate is as follows:
and then calculate the inter-class variance:
s23: constructing an optimization target:
wherein ,
a segmentation threshold for the optimization objective to reach a maximum value.
Traversing the values of t and calculating S21 and S22 for each t to obtain
. According to the->
And dividing each pixel on the denoised glass substrate surface height image into a background or glass substrate class, extracting pixels belonging to the glass substrate, and dividing the glass substrate area from the image.
Optionally, features are extracted from the segmented glass substrate region in the step S3, wherein the features include curvature, torsion and sphericity:
wherein A1, A2, A3 respectively represent the bending degree, the torsion degree and the sphericity of the glass substrate,
indicate->
Curvature of individual pixels ∈>
Mean value representing curvature of all pixels, +.>
The number of pixels in the glass substrate region is indicated. The curvature is calculated in the following way: />
, wherein />
,/>
,/>
and />
Respectively represent +.>
The first and second derivatives of the surface height represented by the individual pixels in the x and y directions. />
Indicate->
Surface height represented by individual pixels, +.>
The average value of the surface height is shown. />
Indicating that the fitting sphere is at +.>
The height value of each pixel point, and the fitting sphere is obtained by a spherical model of the fitting surface height.
Optionally, in the step S4, low-pass filtering is performed on the segmented glass substrate area based on discrete fourier transform, so as to obtain a smoothed glass substrate surface height image, and calculating average value, variance, skewness and kurtosis information of the smoothed glass substrate surface height image, including:
s41: the surface height data of the glass substrate candidate region obtained in S2 is expressed as a two-dimensional matrix
, wherein />
and />
Representing coordinates in the horizontal and vertical directions;
s42: for a pair of
Performing two-dimensional discrete Fourier transform to obtain frequency domain representation +.>
：
Wherein k is an imaginary unit,
the method comprises the steps of carrying out a first treatment on the surface of the W and M are lengths of the image in horizontal and vertical directions;
s43: filtering out
Is higher frequency part of (2) and is about the remainder->
The frequency information of (a) is inversely changed to obtain the surface height after smoothing:
s44: for a pair of
Analyzing to obtain average value, variance, skewness and kurtosis information:
wherein B1, B2, B3 and B4 respectively represent the average value, variance, skewness and kurtosis information of the surface height after smoothing, and
。
optionally, comparing the results obtained in S3 and S4 with the standard glass substrate in the step S5, and evaluating the flatness of the glass substrate includes:
combining the results obtained in S3 and S4 into a glass substrate overall characteristic:
and compared to the overall characteristics extracted from the standard glass substrate. The standard glass substrate is obtained by manual screening, and the comparison method comprises the following steps:
wherein ,
，/>
the glass substrate to be evaluated is completely consistent with the standard glass substrate, and the flatness of the glass substrate is higher,/-degree>
The glass substrate to be evaluated is completely inconsistent with the standard glass substrate, and the flatness of the glass substrate is low. According to->
The value of the glass substrate is divided into four grades of disqualification, qualification, good and excellent from low to high on the flatness of the glass substrate.
The invention also provides a system for evaluating the flatness of the glass substrate, which comprises the following steps:
and the image acquisition and denoising module: collecting a glass substrate surface height image, and denoising the image;
an image feature segmentation module: dividing the denoised surface height image of the glass substrate, and extracting pixels belonging to the glass substrate;
the image feature extraction module: extracting features from the segmented glass substrate regions;
and an image analysis module: performing low-pass filtering on the segmented glass substrate area based on discrete Fourier transform to obtain a smoothed glass substrate surface height image, and analyzing the image;
and an evaluation module: and comparing the result of the glass substrate to be evaluated with a standard glass substrate, and evaluating the flatness of the glass substrate.
Advantageous effects
The method of the invention adopts digital image processing and mathematical analysis, can automatically segment, extract and evaluate the surface height image of the glass substrate, does not need manual intervention, and improves the measurement efficiency and evaluation accuracy. The traditional glass substrate flatness evaluation method needs manual measurement and recording, is easy to generate errors, requires a large amount of time and labor cost, and is not suitable for quality control in mass production.
The glass substrate flatness evaluation method comprehensively considers a plurality of characteristic factors such as flatness, bending degree, torsion degree, sphericity and the like, and can obtain more comprehensive and accurate evaluation results. However, the traditional evaluation method only considers a certain or a few characteristic factors, and the evaluation result is not comprehensive and accurate enough.
The glass substrate flatness evaluation method adopts discrete Fourier transform to carry out low-pass filtering, can filter high-frequency noise, retains main characteristics of a glass substrate surface height image, adopts mathematical analysis methods such as average value, variance, skewness, kurtosis and the like, analyzes characteristics of the image from multiple angles, and improves evaluation precision and accuracy.
The glass substrate flatness evaluation method disclosed by the invention does not need expensive equipment and tools, is simple and convenient to operate, is suitable for evaluating various glass substrates, and has a great application prospect. However, the conventional glass substrate flatness evaluation method requires specialized equipment and tools, and has high cost.
Drawings
Fig. 1 is a flow chart illustrating a method for evaluating flatness of a glass substrate according to an embodiment of the invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings, without limiting the invention in any way, and any alterations or substitutions based on the teachings of the invention are intended to fall within the scope of the invention.
Example 1: a glass substrate flatness evaluation method, as shown in FIG. 1, comprises the following steps:
s1: and acquiring a glass substrate surface height image, denoising the image, and obtaining the denoised glass substrate surface height image.
The method comprises the steps of placing a glass substrate needing flatness evaluation on a platform, measuring heights of different positions of the platform by using laser, and recording the heights of each point to form a surface height image, wherein the distance from the surface of the glass substrate to a reference surface is recorded by each pixel of the surface height image, and the reference surface is a plane for placing the glass substrate. And denoising the image by utilizing improved median filtering to obtain a more accurate glass substrate surface height image. The improved median filtering includes:
s11: marking potential noise points:
wherein ,
is->
Window->
Center pixel value of (2); />
and />
Respectively representing a maximum value and a minimum value in the window; if->
Or->
Mark->
As potential noise points, otherwise
Is a non-noise point;
s12: reconfirming the potential noise point by using a large window, if the potential noise point is still the noise point, replacing the pixel value by using a median value in the window, otherwise, reserving the pixel value:
wherein ,
is->
Window->
Center pixel value of (2); />
The function expression is->
Median of all pixel values in (a);
s13: and (3) performing S11 and S12 operation on each pixel of the glass substrate surface height image until the whole image is traversed, and obtaining the denoised glass substrate surface height image.
The image of the surface height of the glass substrate is basic data for evaluating the flatness of the glass substrate, and if a large amount of noise is contained in the image, the accuracy and reliability of the subsequent processing and evaluation results are seriously affected. Therefore, denoising the surface height image of the glass substrate can ensure the data quality and reduce errors and interference. Meanwhile, unnecessary details and interference information in the image can be reduced by removing noise, and the accuracy of feature extraction is improved. If the image contains a lot of noise, the accuracy of feature extraction will be greatly affected, thereby affecting the subsequent evaluation result.
S2: and dividing the denoised surface height image of the glass substrate, extracting pixels belonging to the glass substrate, and dividing the glass substrate area from the image.
S21: selecting a segmentation threshold t, classifying the denoised glass substrate surface height image in the step S1 into two categories of background and glass substrate, and calculating probability of each category:
wherein ,
l is the maximum value of the surface height image; />
，/>
The pixel number is the pixel number of the image value q, and N is the total pixel number of the image; />
and />
Representing probabilities of pixels belonging to the background and the glass substrate class in the glass surface height image respectively;
s22: calculating the average value of pixels in the two categories of the background and the glass substrate, and calculating the inter-category variance:
the average value calculation method of the pixels of the background and glass substrate class is as follows:
the average value calculation mode of the surface height image of the whole glass substrate is as follows:
and then calculate the inter-class variance:
s23: constructing an optimization target:
wherein ,
a segmentation threshold for the optimization objective to reach a maximum value. />
Traversing the values of t and calculating S21 and S22 for each t to obtain
. According to the->
And dividing each pixel on the denoised glass substrate surface height image into a background or glass substrate class, extracting pixels belonging to the glass substrate, and dividing the glass substrate area from the image.
The glass substrate can be separated from the image by an image segmentation technology, so that characteristic information of the surface of the glass substrate, such as bending degree, torsion degree, sphericity and the like, can be extracted. Such characteristic information is very important for evaluating flatness of the glass substrate, and can provide powerful support for subsequent processing. There may be a lot of background interference in the glass substrate surface height image, which if not segmented, will greatly interfere with subsequent processing and evaluation results. These disturbances and noise can be removed by segmentation, improving the accuracy and reliability of the assessment.
S3: features including curvature, torsion and sphericity are extracted from the segmented glass substrate regions.
Wherein A1, A2, A3 respectively represent the bending degree, the torsion degree and the sphericity of the glass substrate,
indicate->
Curvature of individual pixels ∈>
Mean value representing curvature of all pixels, +.>
The number of pixels in the glass substrate region is indicated. The curvature is calculated in the following way: />
, wherein />
,/>
,/>
and />
Respectively represent +.>
The first and second derivatives of the surface height represented by the individual pixels in the x and y directions. />
Indicate->
Surface height represented by individual pixels, +.>
The average value of the surface height is shown. />
Indicating that the fitting sphere is at +.>
The height value of each pixel point, and the fitting sphere is obtained by a spherical model of the fitting surface height.
The basic form of the spherical model is:
wherein (a, b, c) is the center of sphere coordinate, r is the radius of the sphere,
is the coordinates of a point on the sphere.
Substituting the abscissa and the ordinate of the pixel point and the surface height into a spherical model, and solving spherical model parameters by using a least square method, wherein the least square method is as follows:
wherein 
The abscissa and the ordinate of the pixel point are the surface height.
S4: and performing low-pass filtering on the segmented glass substrate region based on discrete Fourier transform to obtain a smoothed glass substrate surface height image, and calculating the average value, variance, skewness and kurtosis information of the smoothed glass substrate surface height image.
S41: the surface height data of the glass substrate candidate region obtained in S2 is expressed as a two-dimensional matrix
, wherein />
and />
Representing coordinates in the horizontal and vertical directions;
s42: for a pair of
Performing two-dimensional discrete Fourier transform to obtain frequency domain representation +.>
：
Wherein k is an imaginary unit,
the method comprises the steps of carrying out a first treatment on the surface of the W and M are lengths of the image in horizontal and vertical directions;
s43: filtering out
Is higher frequency part of (2) and is about the remainder->
The frequency information in the frequency information is inversely changed to obtainSmooth surface height:
s44: for a pair of
Analyzing to obtain average value, variance, skewness and kurtosis information: />
Wherein B1, B2, B3 and B4 respectively represent the average value, variance, skewness and kurtosis information of the surface height after smoothing, and
。
the average value, variance, skewness and kurtosis can be used for describing the statistical characteristics of the surface height of the glass substrate after the surface height image of the glass substrate is smoothed, so that the flatness of the glass substrate is evaluated. The average value reflects the average height value of the pixels in the image and can be used to evaluate the overall flatness of the glass substrate surface. The variance reflects the degree of dispersion of the pixel height values in the image and can be used to evaluate the local flatness of the glass substrate surface. The degree of deflection, which reflects the degree of deflection of the pixel height distribution in the image, can be used to evaluate the non-uniformity of the glass substrate surface height. Kurtosis reflects the degree of kurtosis of the height distribution of pixels in an image and can be used to evaluate the sharpness of the surface height of a glass substrate.
The importance of calculating these statistical features after low pass filtering is that they can provide a comprehensive description of the surface height distribution of the glass substrate, evaluating its flatness from multiple angles. At the same time, low pass filtering can reduce high frequency noise in the image, making these statistical features more reliable and accurate.
S5: comparing the results obtained in S3 and S4 with the standard glass substrate, and evaluating the flatness of the glass substrate.
Combining the results obtained in S3 and S4 into a glass substrate overall characteristic:
and compared to the overall characteristics extracted from the standard glass substrate. The standard glass substrate is obtained by manual screening, and the comparison method comprises the following steps:
wherein ,
，/>
the glass substrate to be evaluated is completely consistent with the standard glass substrate, and the flatness of the glass substrate is higher,/-degree>
The glass substrate to be evaluated is completely inconsistent with the standard glass substrate, and the flatness of the glass substrate is low. According to->
The value of the glass substrate is divided into four grades of disqualification, qualification, good and excellent from low to high on the flatness of the glass substrate.
Example 2: the invention also discloses a method and a system for evaluating the flatness of the glass substrate, wherein the method comprises the following five modules:
and the image acquisition and denoising module: collecting a glass substrate surface height image, and denoising the image;
an image feature segmentation module: dividing the denoised surface height image of the glass substrate, and extracting pixels belonging to the glass substrate;
the image feature extraction module: extracting features from the segmented glass substrate regions;
and an image analysis module: performing low-pass filtering on the segmented glass substrate area based on discrete Fourier transform to obtain a smoothed glass substrate surface height image, and analyzing the image;
and an evaluation module: and comparing the result of the glass substrate to be evaluated with a standard glass substrate, and evaluating the flatness of the glass substrate.
It should be noted that, the foregoing reference numerals of the embodiments of the present invention are merely for describing the embodiments, and do not represent the advantages and disadvantages of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (7)
1. A method for evaluating flatness of a glass substrate, comprising the steps of:
s1: collecting a glass substrate surface height image, denoising the image, and obtaining a denoised glass substrate surface height image;
s2: dividing the denoised glass substrate surface height image, extracting pixels belonging to the glass substrate, and dividing a glass substrate area from the image;
s3: extracting features from the segmented glass substrate region, the features including tortuosity, and sphericity;
s4: performing low-pass filtering on the segmented glass substrate region based on discrete Fourier transform to obtain a smoothed glass substrate surface height image, and calculating average value, variance, skewness and kurtosis information of the smoothed glass substrate surface height image;
s5: comparing the results obtained in S3 and S4 with the standard glass substrate, and evaluating the flatness of the glass substrate.
2. The method for evaluating the flatness of a glass substrate according to claim 1, wherein the step S1 of collecting the surface height image of the glass substrate and denoising the image to obtain the denoised surface height image of the glass substrate comprises:
shooting a surface height image of a glass substrate to be subjected to flatness evaluation, wherein the distance from the surface of the glass substrate to a reference surface is recorded in each pixel of the surface height image, and the reference surface is a plane for placing the glass substrate; denoising the image by utilizing improved median filtering to obtain a more accurate glass substrate surface height image; the improved median filtering includes:
s11: marking potential noise points:
wherein ,
is->
Window->
Center pixel value of (2); />
and />
Respectively representing a maximum value and a minimum value in the window; if->
Or->
Mark->
Is a potential noise point, otherwise->
Is a non-noise point;
s12: reconfirming the potential noise point by using a large window, if the potential noise point is still the noise point, replacing the pixel value by using a median value in the window, otherwise, reserving the pixel value:
wherein ,
is->
Window->
Center pixel value of (2); />
The function expression is->
Median of all pixel values in (a);
s13: and (3) performing S11 and S12 operation on each pixel of the glass substrate surface height image until the whole image is traversed, and obtaining the denoised glass substrate surface height image.
3. The method according to claim 2, wherein the step S2 of dividing the denoised glass substrate surface height image, extracting pixels belonging to the glass substrate, and dividing the glass substrate region from the image comprises:
s21: selecting a segmentation threshold t, classifying the denoised glass substrate surface height image in the step S1 into two categories of background and glass substrate, and calculating probability of each category:
wherein ,
l is the maximum value of the surface height image; />
，/>
The pixel number is the pixel number of the image value q, and N is the total pixel number of the image; />
and />
Representing probabilities of pixels belonging to the background and the glass substrate class in the glass surface height image respectively;
s22: calculating the average value of pixels in the two categories of the background and the glass substrate, and calculating the inter-category variance:
the average value calculation method of the pixels of the background and glass substrate class is as follows:
the average value calculation mode of the surface height image of the whole glass substrate is as follows:
and then calculate the inter-class variance:
s23: constructing an optimization target:
wherein ,
a segmentation threshold for the optimization objective to reach a maximum;
traversing the values of t and calculating S21 and S22 for each t to obtain
According to the->
And dividing each pixel on the denoised glass substrate surface height image into a background or glass substrate class, extracting pixels belonging to the glass substrate, and dividing the glass substrate area from the image.
4. The method according to claim 3, wherein the step S3 of extracting features including curvature, torsion and sphericity from the divided glass substrate region comprises:
wherein A1, A2, A3 respectively represent the bending degree, the torsion degree and the sphericity of the glass substrate,
indicate->
Curvature of individual pixels ∈>
Mean value representing curvature of all pixels, +.>
The number of pixels in the glass substrate area is represented; the curvature is calculated in the following way: />
, wherein />
,/>
,/>
and />
Respectively represent +.>
First and second derivatives of the surface height represented by the individual pixels in the x and y directions; />
Indicate->
Surface height represented by individual pixels, +.>
Mean value representing surface height; />
Indicating that the fitting sphere is at +.>
The height value of each pixel point, and the fitting sphere is obtained by a spherical model of the fitting surface height.
5. The method according to claim 4, wherein the step S4 of performing low-pass filtering on the segmented glass substrate region based on discrete fourier transform to obtain a smoothed glass substrate surface height image, and calculating the average value, variance, skewness, kurtosis information thereof comprises:
s41: the surface height data of the glass substrate candidate region obtained in S2 is expressed as a two-dimensional matrix
, wherein />
and />
Representing coordinates in the horizontal and vertical directions;
s42: for a pair of
Performing two-dimensional discrete Fourier transform to obtain frequency domain representation +.>
：
Wherein k is an imaginary unit,
the method comprises the steps of carrying out a first treatment on the surface of the W and M are lengths of the image in horizontal and vertical directions; />
Representing coordinates in the horizontal and vertical directions in the frequency domain; e is a natural constant; />
Is the circumference ratio; />
S43: filtering out
Is higher frequency part of (2) and is about the remainder->
The frequency information of (a) is inversely changed to obtain the surface height after smoothing:
s44: for a pair of
Analyzing to obtain average value, variance, skewness and kurtosis information:
wherein B1, B2, B3 and B4 respectively represent the average value, variance, skewness and kurtosis information of the surface height after smoothing, and
。
6. the method for evaluating flatness of a glass substrate according to claim 5, wherein comparing the results obtained in S3 and S4 with a standard glass substrate in step S5, the method comprising:
combining the results obtained in S3 and S4 into a glass substrate overall characteristic:
and comparing with the overall features extracted from the standard glass substrate; the comparison method comprises the following steps:
wherein ,
，/>
the glass substrate to be evaluated is completely consistent with the standard glass substrate, and the flatness of the glass substrate is higher,/-degree>
The glass substrate to be evaluated is completely inconsistent with the standard glass substrate, and the flatness of the glass substrate is low.
7. A glass substrate flatness evaluation system, comprising:
and the image acquisition and denoising module: collecting a glass substrate surface height image, and denoising the image;
an image feature segmentation module: dividing the denoised glass substrate surface height image, and extracting pixels belonging to the glass substrate area;
the image feature extraction module: extracting features from the segmented glass substrate regions;
and an image analysis module: performing low-pass filtering on the segmented glass substrate area based on discrete Fourier transform to obtain a smoothed glass substrate surface height image, and analyzing the image;
and an evaluation module: comparing the result of the glass substrate to be evaluated with a standard glass substrate, and evaluating the flatness of the glass substrate;
to realize a glass substrate flatness evaluation method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310505983.3A CN116245879B (en) | 2023-05-08 | 2023-05-08 | Glass substrate flatness evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310505983.3A CN116245879B (en) | 2023-05-08 | 2023-05-08 | Glass substrate flatness evaluation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116245879A true CN116245879A (en) | 2023-06-09 |
| CN116245879B CN116245879B (en) | 2023-08-01 |
Family
ID=86624614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310505983.3A Active CN116245879B (en) | 2023-05-08 | 2023-05-08 | Glass substrate flatness evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116245879B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116523906A (en) * | 2023-06-28 | 2023-08-01 | 长沙韶光芯材科技有限公司 | Method and system for detecting optical performance of glass substrate |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446911A (en) * | 2016-09-13 | 2017-02-22 | 李志刚 | Hand recognition method based on image edge line curvature and distance features |
| US20170069089A1 (en) * | 2011-10-12 | 2017-03-09 | Seno Medical Instruments, Inc. | System And Method For Acquiring Optoacoustic Data And Producing Parametric Maps Using Subband Acoustic Compensation |
| CN106803237A (en) * | 2016-12-14 | 2017-06-06 | 银江股份有限公司 | A kind of improvement self-adaptive weighted average image de-noising method based on extreme learning machine |
| CN111145161A (en) * | 2019-12-28 | 2020-05-12 | 北京工业大学 | Method for processing and identifying pavement crack digital image |
| CN113324498A (en) * | 2021-05-06 | 2021-08-31 | 西安理工大学 | Multi-parameter high-precision measurement system and method for flatness of ultrathin glass substrate |
| CN113674204A (en) * | 2021-07-16 | 2021-11-19 | 杭州未名信科科技有限公司 | Wood board deformation detection method and system based on deep learning and 3D point cloud data |
| CN113689478A (en) * | 2021-09-03 | 2021-11-23 | 凌云光技术股份有限公司 | Alignment method, device and system of measuring equipment |
| CN115239728A (en) * | 2022-09-23 | 2022-10-25 | 江苏海舟安防科技有限公司 | Fire-fighting equipment identification method |
| CN115311309A (en) * | 2022-09-05 | 2022-11-08 | 中科微影(浙江)医疗科技有限公司 | Method and system for identifying and extracting focus of nuclear magnetic resonance image |
-
2023
- 2023-05-08 CN CN202310505983.3A patent/CN116245879B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170069089A1 (en) * | 2011-10-12 | 2017-03-09 | Seno Medical Instruments, Inc. | System And Method For Acquiring Optoacoustic Data And Producing Parametric Maps Using Subband Acoustic Compensation |
| CN106446911A (en) * | 2016-09-13 | 2017-02-22 | 李志刚 | Hand recognition method based on image edge line curvature and distance features |
| CN106803237A (en) * | 2016-12-14 | 2017-06-06 | 银江股份有限公司 | A kind of improvement self-adaptive weighted average image de-noising method based on extreme learning machine |
| CN111145161A (en) * | 2019-12-28 | 2020-05-12 | 北京工业大学 | Method for processing and identifying pavement crack digital image |
| CN113324498A (en) * | 2021-05-06 | 2021-08-31 | 西安理工大学 | Multi-parameter high-precision measurement system and method for flatness of ultrathin glass substrate |
| CN113674204A (en) * | 2021-07-16 | 2021-11-19 | 杭州未名信科科技有限公司 | Wood board deformation detection method and system based on deep learning and 3D point cloud data |
| CN113689478A (en) * | 2021-09-03 | 2021-11-23 | 凌云光技术股份有限公司 | Alignment method, device and system of measuring equipment |
| CN115311309A (en) * | 2022-09-05 | 2022-11-08 | 中科微影(浙江)医疗科技有限公司 | Method and system for identifying and extracting focus of nuclear magnetic resonance image |
| CN115239728A (en) * | 2022-09-23 | 2022-10-25 | 江苏海舟安防科技有限公司 | Fire-fighting equipment identification method |
Non-Patent Citations (2)
| Title |
|---|
| ZHIGANG JIANG 等: "Lgt-net: Indoor panoramic room layout estimation with geometry-aware transformer network", 《CVPR2022》 * |
| 王伶 等: "建筑玻璃平整度技术要求及检测方法综述", 《绿色建筑》, vol. 14, no. 2 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116523906A (en) * | 2023-06-28 | 2023-08-01 | 长沙韶光芯材科技有限公司 | Method and system for detecting optical performance of glass substrate |
| CN116523906B (en) * | 2023-06-28 | 2023-09-12 | 长沙韶光芯材科技有限公司 | Method and system for detecting optical performance of glass substrate |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116245879B (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022042579A1 (en) | Lcd screen defect detection method and apparatus | |
| CN109632808B (en) | Edge defect detection method and device, electronic equipment and storage medium | |
| CN116245879B (en) | Glass substrate flatness evaluation method and system | |
| CN115861291A (en) | Patch circuit board production defect detection method based on machine vision | |
| CN116645367B (en) | Steel plate cutting quality detection method for high-end manufacturing | |
| JP5718781B2 (en) | Image classification apparatus and image classification method | |
| CN108257125B (en) | Depth image quality non-reference evaluation method based on natural scene statistics | |
| CN115018835B (en) | Automobile starter gear detection method | |
| CN112991287A (en) | Automatic indentation measurement method based on full convolution neural network | |
| CN113870202A (en) | Far-end chip defect detection system based on deep learning technology | |
| US20110164129A1 (en) | Method and a system for creating a reference image using unknown quality patterns | |
| CN112836726B (en) | Pointer instrument indication reading method and device based on video information | |
| CN112686896B (en) | Glass defect detection method based on frequency domain and space combination of segmentation network | |
| CN114419140A (en) | Positioning algorithm for light spot center of track laser measuring device | |
| CN113705564A (en) | Pointer type instrument identification reading method | |
| CN117576100A (en) | Surface defect grading detection and evaluation method for FPC connector | |
| CN116883987A (en) | Pointer instrument reading identification method for unmanned inspection of transformer substation | |
| CN116485749A (en) | Self-encoder-based method for identifying dirt in lens module | |
| CN113191997B (en) | Clamp spring measuring method based on machine vision | |
| CN115761606A (en) | Box electric energy meter identification method and device based on image processing | |
| CN115239947A (en) | Wheat stripe rust severity evaluation method and device based on unsupervised learning | |
| CN114842041A (en) | Embedded-based steel rail profile detection light bar extraction method | |
| CN114742832A (en) | Welding defect detection method for MiniLED thin plate | |
| CN114549403A (en) | Multi-monomer intelligent precise geometric circle center detection method for side section of mechanical part | |
| CN114387232A (en) | Wafer center positioning, wafer gap positioning and wafer positioning calibration method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |