SD-OCT system for thickness detection of optical element and detection method thereof

Abstract

The invention belongs to the field of optical element detection, and particularly relates to an SD-OCT system for optical element thickness detection and a detection method thereof, and the detection method comprises the steps: carrying out the mask and Gaussian filtering processing of an OCT image of a to-be-detected optical element; carrying out peak point detection on the OCT image subjected to Gaussian filtering by adopting a standard deviation peak searching algorithm, and identifying a light intensity abrupt change position at a film layer interface; carrying out mode analysis on the detected peak point, and determining the number of film layers; performing connected domain analysis on the peak points after mode analysis, filtering out isolated peak points, and performing interpolation processing on missing positions; and performing curve fitting on the peak points after the connected domain analysis, determining the position of each film layer, and calculating the actual thickness of the optical element to be measured. According to the detection method, the steps of mask, Gaussian filtering, standard deviation peak searching, mode analysis, connected domain filtering and the like are fused, artifacts and noise interference are effectively eliminated, and accurate positioning of the film boundary is achieved.

Description

Technical Field

[0001] This invention belongs to the field of optical component inspection technology, and particularly relates to an SD-OCT system and its inspection method for optical component thickness inspection. Background Technology

[0002] With the rapid development of industrial products towards precision and miniaturization, the thickness of the film structures in optical components such as glass, mobile phone lenses, optical coatings, and eyeglass lenses needs to be controlled with micrometer-level precision. The uniformity and accuracy of this thickness directly determine the product's optical performance, mechanical stability, and reliability. For example, if the thickness deviation of the multilayer coating in a mobile phone lens exceeds the micrometer range, it will lead to image distortion; thickness errors in optical glass may cause optical path differences, affecting the transmission efficiency of the optical system. Therefore, achieving high-precision, high-efficiency, non-destructive testing of optical component films has become one of the core requirements for ensuring product quality in the optical manufacturing industry.

[0003] Optical coherence tomography (OCT) is a three-dimensional imaging technique based on the Michelson interferometry principle. It has significant advantages such as non-contact operation, high resolution (1-15 μm), high speed, and high signal-to-noise ratio. By interfering the scattered light from broadband lasers at different depths within the sample with the reflected light from the reference arm, it can directly obtain the tomographic depth information of the sample. Combined with two-dimensional galvanometer scanning, it can further generate three-dimensional structural images, achieving micron-level thickness detection and interface positioning, and has become a potential technological direction for solving the problem of precision detection of optical components.

[0004] Currently, OCT technology has been widely applied and achieved mature results in biomedical diagnostics (such as retinal imaging). Biological tissues have uniform scattering characteristics, and the gray-level differences between different tissue layers in OCT images are obvious with clear boundaries. Based on traditional algorithms such as filtering, gradient difference, and multi-scale segmentation, automatic segmentation and thickness measurement of the multi-layered structure of the retina can be stably completed, demonstrating good robustness and reliability, laying the foundation for the engineering application of OCT technology.

[0005] Although OCT technology performs excellently in the biomedical field, its application to the inspection of industrial samples such as optical components faces numerous technical bottlenecks due to the differences in the characteristics of industrial samples and inspection requirements, making traditional OCT detection algorithms difficult to apply. 1. Signal interference issues with industrial samples Strong reflection and non-uniform scattering often exist on the surface and inside of optical components (such as glass and optical coatings). On the one hand, the high reflectivity of the material surface will generate strong noise signals, which will mask the true reflection information of the film layer interface. On the other hand, internal non-uniform scattering will cause the signal intensity of OCT images to vary significantly at different locations. Obvious burrs are likely to appear at the layer boundary, making the interface outline blurred. Traditional algorithms are prone to misidentifying noise as false boundaries, or missing detection or missegmentation due to unclear boundaries, which seriously affects the detection accuracy.

[0006] 2. Performance limitations of traditional OCT algorithms Existing OCT thickness detection algorithms (such as those based on A-scan scanning) have significant limitations in their applicable scenarios: Poor adaptability to film morphology: It can only handle uniform films with linear distribution. When the film is curved (such as the arc coating of mobile phone lens) or deformed due to stress, A-scan scanning projects the film thickness information as a one-dimensional signal, which cannot reflect the spatial distribution characteristics of the film, resulting in significant deviations in thickness calculation. Insufficient accuracy and efficiency: Most traditional algorithms are designed for samples with a thickness of millimeters, which is difficult to meet the measurement requirements of micrometer-level accuracy; and some algorithms require manual intervention (such as manually selecting the light intensity peak), resulting in low detection efficiency and making them unsuitable for batch detection scenarios in industrial production lines. Weak anti-interference capability: It lacks a robust processing mechanism for industrial environmental noise, and cannot effectively filter isolated noise points or abnormal signals. In the detection of complex multilayer films (such as the 5-layer tightly coated eyeglass lens), interlayer confusion is prone to occur, resulting in unreliable measurement results. Summary of the Invention

[0007] In view of this, the present invention aims to provide an SD-OCT system and its detection method for optical component thickness detection, so as to solve the technical problem that traditional OCT algorithms cannot be applied to industrial sample detection.

[0008] To achieve the above objectives, the technical solution created by this invention is implemented as follows: A method for detecting the thickness of optical components based on OCT images, implemented using an SD-OCT system, includes the following steps: S1: Perform masking on the OCT image of the optical element under test to remove non-target areas and retain the target area; S2: Perform Gaussian filtering on the masked OCT image to suppress noise in the OCT image; S3: The standard deviation peak-finding algorithm is used to detect peak points in the OCT image after Gaussian filtering, and the positions of light intensity abrupt changes at the film interface are identified. S4: Perform mode analysis on the detected peak points to determine the number of film layers; S5: Perform connected component analysis on the peak points after mode analysis, filter out isolated peak points, and perform interpolation on missing positions; S6: Perform curve fitting on the peak points after connected component analysis to determine the position of each film layer, and calculate the actual thickness of the optical element under test based on the number of pixels between the film layers and the maximum imaging depth of the SD-OCT system.

[0009] Furthermore, in step S3, the step of using the standard deviation peak-finding algorithm to detect peak points in the Gaussian-filtered OCT image includes: S31: Calculate the mean gray value of each column of pixels in the OCT image. and standard deviation : ; ; in, Indicates the number of valid pixels within the target area. This represents the grayscale value of the i-th valid pixel within each column of pixels; S32: Average of grayscale values ​​for each column of pixels and standard deviation Set an adaptive threshold for the grayscale value of each column of pixels. : ; in, This represents an adjustable empirical coefficient; S33: Traverse each pixel of the OCT image column by column, find pixels that satisfy the following two conditions, and mark them as peak points: Condition 1: The grayscale value of the pixel is greater than the adaptive threshold. ; Condition 2: The gray value of a pixel is greater than the gray values ​​of the pixels directly above and below it.

[0010] Furthermore, in step S4, the step of performing mode analysis on the detected peak points includes: traversing all columns of the OCT image, counting the number of peak points in each column, and taking the mode of all peak points as the number of film layers.

[0011] Furthermore, in step S5, the connected component analysis of the peak points after mode analysis includes: S51: Take each peak point as the center point, set a fixed-size neighborhood for each center point, count the other peak points in the neighborhood except the center point, and take them as neighborhood points; S52: By calculating the proximity of each center point to its neighboring points. : ; in, Indicates the center point. This indicates the number of neighboring points within the neighborhood. Represents a point within a neighborhood. Indicates the center point To neighboring points The distance; S53: Calculate the average proximity of all peak points. And compare it with the proximity of each center point, and then compare all the proximity values. Less than the average The center point is filtered out.

[0012] Furthermore, in step S6, the step of performing curve fitting on the peak points after connected component analysis to determine the location of each film layer includes: S61: Set the number of lines in the fitted curve to the mode of the peak points; S62: Classify the peak points after connected component analysis using density clustering or distance clustering. S63: The peak points of each class are fitted with curves using the polynomial method or spline interpolation method. The position of each curve in the OCT image is the corresponding film layer position.

[0013] Furthermore, in step S6, the actual thickness of the optical element under test is... The calculation formula is: ; in, This indicates the number of pixels spaced between film layers. This indicates the number of pixels in the camera of the spectrometer in the SD-OCT system. This indicates the maximum imaging depth of the SD-OCT system. , Indicates the refractive index of the optical element under test. This indicates the center wavelength of the light source in the SD-OCT system. This indicates the resolution of the spectrometer. , This indicates the operating bandwidth of the spectrometer.

[0014] Furthermore, in step S3, the minimum spacing between two adjacent peak points is set to 3 pixels.

[0015] An SD-OCT system for measuring the thickness of optical components includes: A light source used to emit a broadband spectrum; Fiber optic couplers are used to separate light emitted from a light source into measurement light and reference light. The sample arm and reference arm are used to transmit the measurement light and reference light, respectively; A spectrometer is used to acquire the interference signal between the measurement light and the reference light; The SD-OCT system also includes a computer for processing the interference signals acquired by the spectrometer and performing the aforementioned optical element thickness detection method.

[0016] Compared with the prior art, the present invention can achieve the following beneficial effects: (1) Achieved high precision and robustness in thickness detection: This algorithm effectively overcomes interference problems such as strong reflection, non-uniform scattering, and image noise in industrial optical samples by integrating a series of processing steps such as mask processing, Gaussian filtering, standard deviation peak finding, mode analysis, and connected component filtering. Experimental results show that the algorithm can stably control the thickness measurement error of various samples such as glass, mobile phone lenses, and eyeglass lenses to below 2.2%, and can achieve micron-level measurement accuracy for film thicknesses from millimeters to micrometers (such as 18μm), which is significantly better than the traditional A-scan scanning method and row traversal peak finding method. (2) Improved adaptability and automation in complex situations: The algorithm innovatively adopts column-direction standard deviation peak finding and peak number mode analysis, which can automatically identify and eliminate abnormal data columns, effectively avoiding the interlayer overlap and misjudgment problems that are prone to occur in traditional row traversal methods when the film layers are too close or have a curved distribution. Combined with connected component filtering technology, it can intelligently remove isolated noise points, greatly reducing the need for manual intervention, and realizing fully automatic and highly reliable segmentation and positioning of various film layer morphologies such as straight lines and curves; (3) Breakthrough in multi-layer synchronous detection and visualization in a single scan: Unlike the inefficient mode of traditional methods that analyze layer by layer or point by point, this algorithm can simultaneously extract all membrane interfaces in the sample based on a single OCT scan and intuitively display the precise position and morphology of each layer through curve fitting. This not only improves the detection efficiency by several times, but also realizes the visualization of thickness distribution, providing a direct basis for analyzing membrane uniformity, stress deformation, etc.

[0017] (4) Expanding the industrial application boundaries of OCT technology: This study successfully applied the high-resolution advantage of OCT technology to the inspection of weakly scattering, multi-layered and densely packed optical components such as mobile phone lenses and eyeglass lenses, solving the pain points of insufficient resolution or lack of spatial positioning ability of traditional technologies such as ultrasonic and spectral methods in such scenarios, and providing a new and powerful solution for non-destructive testing and quality control of industrial optical products. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 A schematic diagram of the SD-OCT system for optical element thickness detection as described in the embodiments of the present invention; Figure 2 A schematic flowchart of the optical element thickness detection method based on OCT images described in the embodiments of the present invention; Figure 3 A schematic diagram of the original OCT image of the optical element under test as described in the embodiment of the present invention; Figure 4 A schematic diagram of the OCT image after masking and Gaussian filtering as described in the embodiment of the present invention; Figure 5 A schematic diagram of the OCT image after peak point retrieval as described in an embodiment of the present invention; Figure 6 A schematic diagram of the OCT image after mode finding and connected component filtering as described in the embodiment of the present invention; Figure 7 A schematic diagram showing the thickness information and layering results of the three glass panes described in an embodiment of the present invention; Figure 8 A schematic diagram illustrating the thickness information and layering results of a mobile phone lens as described in an embodiment of the present invention; Figure 9 A schematic diagram of the thickness curve of a mobile phone lens as described in an embodiment of the present invention; Figure 10 A schematic diagram illustrating the thickness information and layering results of the spectacle lens as described in an embodiment of the present invention; Figure 11 A schematic diagram comparing the column traversal peak finding results and the row traversal peak finding results described in the embodiments of the present invention; Figure 12 This is a schematic diagram of thickness information obtained by the traditional A-scan scanning method.

[0019] Figure reference numerals: 1. Light source; 2. Fiber optic coupler; 3. Circulator; 4. Measuring arm; 41. Collimating mirror; 42. X-axis reflector; 43. Y-axis reflector; 44. Scanning objective lens; 5. Reference arm; 51. Collimating lens; 52. Focusing lens; 53. Reflector; 6. Spectrometer; 7. Image acquisition system; 8. Computer; 9. Optical element under test. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0022] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0023] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "assembly," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0024] The invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] This invention provides an SD-OCT (Spectral domain OCT) system for measuring the thickness of optical components, and a method for measuring the thickness of optical components using the SD-OCT system. The structure of the SD-OCT system is as follows: Figure 1As shown, the system includes a light source 1, an optical fiber coupler 2, a circulator 3, a measuring arm 4, a reference arm 5, a spectrometer 6, an image acquisition system (IMQA) 7, and a computer 8. The measuring arm 4 includes a collimating mirror 41, an X-axis reflector 42, a Y-axis reflector 43, and a scanning objective lens 44. The X-axis reflector 42 and the Y-axis reflector 43 form a two-dimensional scanning galvanometer. The reference arm 5 includes a collimating lens 51, a focusing lens 52, and a reflector 53. The broadband spectrum emitted by the light source 1 is split into measuring light and reference light by the optical fiber coupler 2. The reference light is collimated into parallel light by the collimating lens 51 and then focused onto the reflector 53 by the focusing lens 52. The reference light is reflected back to the fiber coupler 2 by the reflector 53. After being collimated into parallel light by the collimating lens 41, the light is deflected by the two-dimensional scanning galvanometer and then focused onto the optical element under test 9 by the scanning objective lens 43. The optical path information of the upper and lower surfaces of different regions of the optical film of the optical element under test 9 is brought back to the fiber coupler 2. The sample light and the reference light modulated by the optical film of the optical element under test 9 generate interference signals in the fiber channel. These interference signals are acquired by the spectrometer 6 and sent to the computer 8 by the image acquisition system 7. The computer 8 processes these interference signals acquired by the spectrometer 6 to detect the thickness of the optical element under test 9.

[0026] Key performance indicators of the SD-OCT system include: the center wavelength of light source 1 is 840nm, the bandwidth of spectrometer 6 is 80nm, the theoretical axial resolution in air is 3.88μm, the pixel size of the 2D camera in spectrometer 6 is 2048×2048, the scanning speed is 70kHz, and it has excellent tomographic capabilities.

[0027] like Figure 2 As shown, the method for detecting the thickness of optical components specifically includes the following steps: S1: Perform masking on the OCT image of the optical element under test to remove non-target areas and retain the target area.

[0028] In actual scanned images, the top and side edge regions of the OCT image of the optical element under test may contain artifacts, such as device bezels, mounting brackets, or top markings. Figure 3 As shown, these structures produce strong edges similar to bright lines, but they are not the target bright lines. This can interfere with the detection results. The optical components that actually need to be detected are usually not in these non-target areas. Therefore, it is necessary to mask the non-target areas of the OCT image to make them completely black, eliminate the interference from the non-target areas, and make the detection focus on the target area to prevent it from affecting the detection results.

[0029] S2: Apply Gaussian filtering to the OCT image after masking to suppress noise in the OCT image.

[0030] Gaussian filtering is a widely used method in image processing to suppress noise while preserving image edges and details. Its most common application is eliminating Gaussian noise and salt-and-pepper noise in images. By weighted averaging of the pixel neighborhood, isolated, abnormally strong noise points are merged into their surrounding areas, making their intensity approximate that of surrounding pixels, thus smoothing the noise. Noise reduction processing is required on the masked OCT image before finding peak points.

[0031] The key to thickness measurement lies in accurately identifying the positions of the upper and lower surfaces of the film. The core advantage of Gaussian filtering is that it eliminates false edges, preventing noise points from being misidentified as edge points; it connects broken edges, smooths adjacent pixels, and helps connect true boundaries broken by noise. This makes edges more continuous, positioning more accurate, and reduces measurement jump errors.

[0032] Since both masking and Gaussian filtering are existing technologies, the specific processes will not be described in detail in this invention. The OCT image after masking and Gaussian filtering is as follows: Figure 4 As shown.

[0033] S3: The standard deviation peak-finding algorithm is used to detect peak points in the OCT image after Gaussian filtering, and the positions of light intensity abrupt changes at the film interface are identified.

[0034] In thickness detection, the presence of a coating causes a sudden change in light intensity at the interface. The standard deviation peak finding algorithm automatically detects the position of these characteristic peaks, thereby accurately identifying the position corresponding to the coating interface, and finally converting the grayscale information in the OCT image into a quantifiable thickness value.

[0035] The standard deviation peak-finding algorithm does not perform a global analysis of the entire OCT image, but rather processes each column of pixels (i.e., each A-scan signal) of the OCT image independently. This is because noise and signal intensity may vary at different locations (different columns) in the image, and column-by-column processing enables adaptive peak detection.

[0036] The steps for peak point detection in OCT images using the standard deviation peak-finding algorithm include: S31: Calculate the mean gray value of each column of pixels in the OCT image. and standard deviation : ; ; in, Indicates the number of valid pixels within the target area. This represents the grayscale value of the i-th valid pixel within each column of pixels.

[0037] The average grayscale value of a column of pixels The average grayscale value of all pixels in this column represents the overall signal level of this column.

[0038] Standard deviation of a column of pixel grayscale values This represents the dispersion of all pixel grayscale values ​​in this column relative to the mean. Standard deviation The larger the value, the greater the fluctuation in the data in that column, and the more likely there is a significant peak.

[0039] S32: Average of grayscale values ​​for each column of pixels and standard deviation Set an adaptive threshold for the grayscale value of each column of pixels. : ; in, This represents an adjustable empirical coefficient, typically ranging from 1 to 3.

[0040] S33: Traverse each pixel of the OCT image column by column, find the pixel that satisfies the following two peak conditions, and mark it as the peak point.

[0041] Peak condition 1: The grayscale value of a pixel is greater than the adaptive threshold. .

[0042] The grayscale value of this pixel must be higher than the calculated adaptive threshold. This ensures that the pixel is a significant signal point.

[0043] Peak condition 2: The gray value of a pixel is greater than the gray values ​​of the pixels directly above and below it.

[0044] The gray value of this pixel must be greater than the gray values ​​of its directly above and below neighboring pixels to ensure that the pixel is a peak, not a rising or falling edge.

[0045] This invention sets different thresholds for each column of pixels, and identifies pixels that meet the peak condition column by column by setting dynamic thresholds, so as to achieve adaptive peak detection.

[0046] To improve reliability, the standard deviation peak finding algorithm can also incorporate other constraints: Minimum peak height: The above adaptive threshold can be used directly. .

[0047] Minimum peak spacing: Prevents a wide bright line (spanning multiple pixels) from being identified as multiple adjacent peaks. Setting the minimum spacing between two adjacent peaks to 3 pixels ensures that even if the film interface is slightly wide, it will only be identified as a single peak.

[0048] Saliency: This is the gray level difference between the current peak point and its neighboring peak points. It can be further required to determine how much a peak needs to stand out from the valleys on both sides to avoid the broad shoulders being falsely detected.

[0049] If a column of pixels has no signal at all (only noise or a black background), its standard deviation It will approach 0. When the standard deviation of a certain column of pixels is detected... When the value is less than a very small value (e.g., 0.001), the standard deviation peak finding algorithm can skip finding the peak of the pixel in that column, assuming that there is no valid peak value in that column, thus avoiding false detection.

[0050] The standard deviation peak finding algorithm can be implemented using dual-mode detection: Mode 1 (Automatic Mode): This mode uses the mature `findpeaks` function (available in scientific computing libraries such as MATLAB and Python SciPy) and provides parameters such as adaptive threshold and minimum spacing for fast and automatic peak finding. In most cases, this mode can successfully detect peaks.

[0051] Mode 2 (Manual Mode): If the findpeaks function fails to find peaks that meet strict conditions in certain columns (possibly due to poor signal strength), it falls back to a more basic manual loop traversal method. This involves directly applying the logic from step S33 above, performing peak detection pixel by pixel. This ensures that the standard deviation peak-finding algorithm is usable even in extreme cases. The OCT image after peak detection is as follows: Figure 5 As shown.

[0052] S4: Perform mode analysis on the detected peak points to determine the number of film layers.

[0053] The mode plays a crucial role in assessing data reliability and determining process stability, improving efficiency when processing multi-point measurement results. It is independent of extreme values, automatically avoiding erroneous data that deviates from the overall picture. It iterates through each column, identifies the peak positions, and records these peaks, marking their x and y coordinates with small red dots. The mode of each column's peaks is then calculated; the mode represents the number of film layers. Columns that do not meet the mode requirement are marked as outliers, and their peaks are not recorded. This process ensures that each column contains valid peaks, eliminating spurious peaks caused by noise.

[0054] ; in, Indicates a column, Represents the mode sequence, This indicates the peak point; only the peak points of each column that satisfies the mode column will be recorded.

[0055] S5: Perform connected component analysis on the peak points after mode analysis, filter out isolated peak points, and perform interpolation on missing positions.

[0056] After finding many peak points, some isolated discrete points will appear because the extracted thickness information shows that the protruding burrs between the two film layers are quite obvious. At this time, it is necessary to use the connected component method to traverse all connected regions and filter out the isolated discrete points.

[0057] The steps for performing connected component analysis on the peak points after mode analysis include: S51: Take each peak point as the center point, set a fixed-size neighborhood for each center point, count the other peak points in the neighborhood except the center point, and take them as neighborhood points.

[0058] S52: By calculating the proximity of each center point to its neighboring points. : ; in, Indicates the center point. This indicates the number of neighboring points within the neighborhood. Represents a point within a neighborhood. Indicates the center point To neighboring points The distance.

[0059] Proximity This measure is used to assess the density or clustering of a peak point with its surrounding peak points. High density indicates that there are many peak points around this peak point, closely clustered together, which likely represents a true film boundary. Low density indicates that the peak point is surrounded by empty space, with few or no peak points, which is likely an isolated spurious peak point caused by random noise.

[0060] S53: Calculate the average proximity of all peak points. And compare it with the proximity of each center point, and then compare all the proximity values. Less than the average The center point is filtered out.

[0061] Through the above connected component analysis, isolated peak points are removed. The locations of the removed peak points form missing locations. At this time, interpolation processing (such as linear interpolation or spline interpolation) is required for the missing locations to complete the peak point coordinates that should be in the missing locations, thereby forming a complete and smooth curve and preventing the missing locations from affecting the data during curve fitting.

[0062] The OCT image after filtering out isolated peaks by finding the mode and connected components is as follows: Figure 6 As shown.

[0063] S6: Perform curve fitting on the peak points after connected component analysis to determine the position of each film layer, and calculate the actual thickness of the optical element under test based on the number of pixels between the film layers and the maximum imaging depth of the SD-OCT system.

[0064] The steps for determining the location of each film layer by curve fitting of the peak points after connected component analysis include: S61: Set the number of lines in the fitted curve to the mode of the peak points.

[0065] The mode value obtained through mode analysis directly corresponds to the number of film layers. The number of film layers is the number of curves that need to be fitted. In other words, the number of film layers is equal to the number of fitted curves.

[0066] S62: Classify the peak points after connected component analysis using density clustering or distance clustering.

[0067] After connected component filtering, the remaining peak points are the valid points of the real membrane interface. Since the peak points of the same membrane interface have similar spatial distribution trends in the OCT image, the coordinates of the peak points (the horizontal and vertical coordinates of the image) can be used to group spatially adjacent peak points with consistent distribution trends into one class through density clustering (DBSCAN, etc.) or distance clustering methods.

[0068] S63: The peak points of each class are fitted with curves using the polynomial method or spline interpolation method. The position of each curve in the OCT image is the corresponding film layer position.

[0069] Each point set represents incomplete, discrete location information of a membrane layer. A continuous boundary line needs to be obtained from it. A fitting algorithm is used to independently fit a curve to each point set. Commonly used fitting algorithms include polynomial and spline interpolation algorithms.

[0070] After fitting, the position of each membrane layer is no longer an isolated point, but a continuous curve defined by a function. The area that this curve passes through in the OCT image is the position of a certain membrane layer interface.

[0071] Actual thickness of the optical element under test The calculation formula is: ; in, This indicates the number of pixels spaced between film layers. This indicates the number of pixels in the camera of the spectrometer. This indicates the maximum imaging depth of the SD-OCT system. , Indicates the refractive index of the optical element under test. Indicates the center wavelength of the light source. This indicates the resolution of the spectrometer. , This indicates the operating bandwidth of the spectrometer.

[0072] The detection method provided by this invention can locate the film layer and position of the optical element under test in a single scan. To verify the effectiveness of this invention, experiments were conducted using glass, mobile phone lenses, and eyeglass lenses, respectively.

[0073] Experiment 1: Three Glasses like Figure 7 As shown in the figure, (a), (c), and (e) are the original images of the thickness information of the three glass pieces. The background is mostly black with some bright lines in the middle, which are the positions of the film layers. The distance between the bright lines is the thickness of the film layer. In addition to the required bright lines, there are also some noise points and protruding burrs in the figure. An algorithm is needed to eliminate the influence of noise and burrs and find the number and position of the film layers.

[0074] The thickness information and layering results of the first piece of glass are as follows: Figure 7 As shown in (a) and (b), the vertical pixel count in (a) and (b) is 2048, and the distance between the two lines is 479.6528 pixels; the thickness information and layering results of the second glass are as follows. Figure 7 As shown in (c) and (d), the vertical pixel count in (c) and (d) is 1024, and the distance between the two lines is 752.118 pixels; the thickness information and layering results of the third glass are as follows. Figure 7 As shown in (e) and (f), the vertical pixel count in (e) and (f) is 2048, and the distance between the two lines is 994.166 pixels. By calculating the number of pixels between the two lines, the actual thickness of the three glass pieces can be calculated. The SD-OCT system can generate multiple thickness information images of the same glass piece in a single scan; the pixel count between film layers is the average of the multiple images. Table 1 below shows the glass type, refractive index, theoretical value, measured value, and error analysis. The refractive index is obtained using the Cauchy dispersion formula, the error value is the difference between the theoretical and measured values ​​of the sample, and the error percentage is the ratio of the error value to the theoretical value multiplied by 100%.

[0075] Table 1. Test results and error analysis of the three pieces of glass.

[0076] As shown in Table 1, the detection accuracy for the position and thickness of the glass film is high, with an error of less than 1%, which is at the micrometer level. Therefore, the position and thickness of the glass film can be accurately located.

[0077] Experiment 2: Mobile Phone Lens The thickness information collected by the spectrometer on the glass shows two bright lines that are far apart and approximately parallel, making it relatively easy to locate the film layer and fit it with colored lines to calculate the film thickness. Next, we will test this with a more complex mobile phone lens.

[0078] The thickness information and layering results of the mobile phone lens are as follows: Figure 8 As shown in (a) and (b) in the figure.

[0079] The thickness curve of a mobile phone lens is as follows Figure 9 As shown in (a) above, the comparison results between the thickness curve of the mobile phone lens and the theoretical curve are as follows: Figure 9 As shown in (b). The horizontal axis of (a) and (b) represents the number of pixels in the horizontal direction, and the vertical axis represents the difference in pixels between the two fitted curves in the vertical direction. The coating layer of a mobile phone lens is a smooth film, and the difference between the coating layer curve and the theoretical curve is very small. The thickness curve of the coating layer of a mobile phone lens can be accurately plotted. Therefore, this detection method is also applicable to the detection of mobile phone lenses.

[0080] Experiment 3: Eyeglass Lenses The thickness of eyeglass lenses is very small (micrometer level). The thickness information and layering results of the eyeglass lenses are as follows: Figure 10 As shown in (a) and (b) in the figure. Through Figure 10 As can be seen from (b) in the figure, this detection method can accurately locate the position of the lens coating, realize the layering and visualization of the lens coating.

[0081] Table 2 below shows the lens type, refractive index, theoretical value, measured value, and error analysis.

[0082] Table 2. Test results and error analysis of spectacle lenses

[0083] In Table 2, the first column (1-2) represents the film thickness between the first and second layers, (2-3) represents the film thickness between the second and third layers, (3-4) represents the film thickness between the third and fourth layers, and (4-5) represents the film thickness between the fourth and fifth layers. As shown in Table 2, this testing method is also applicable to the testing of spectacle lenses. The measurement error between the theoretical and actual values ​​is very small, below 2.2%, thus allowing for accurate location and thickness determination of the film layers in spectacle lenses.

[0084] It should be noted that, assuming the remaining steps of the method remain unchanged, when finding the peak point, if we iterate through each row of pixels and find the peak point by calculating the average brightness of each row of pixels and then finding the row with the highest brightness, the fitted result will be as follows when processing two closely spaced bright lines: Figure 11 As shown, Figure 11 Image (a) in the image is the correctly identified image. Figure 11(b) in the figure shows the error identification. As can be seen, because the distance between the two film layers is too close, and the two film layers are roughly distributed diagonally from the upper left to the lower right, there is a section of overlap between the lower right end of the upper bright line and the upper left end of the lower bright line in the row direction. If the peak is searched by traversing each row, the overlapping part will be incorrect, which will have a great impact on the fitted curve results and the measurement results, making it difficult to find the correct film layer position and calculate the correct thickness value.

[0085] The edge tracking method involves setting a threshold to convert the denoised OCT image into a binary image, and then selecting the first pixel with an intensity of 1 on the right side of the brightness as the starting point for edge tracking. This method is prone to errors when the distance between the two membranes is too close, or when there are burrs connecting the two membranes. Figure 11 The case shown in (b) is as follows.

[0086] Traditional methods for detecting thickness using OCT systems employ A-scan scanning to project the sample's thickness information, such as... Figure 12 As shown, this method finds several points with the highest light intensity, thus converting the thickness information into one-dimensional intensity values. The pixel difference is obtained by subtracting the x-coordinates of two light intensity peaks, and the thickness value is then calculated using the system's imaging depth and the sample's refractive index. Compared to algorithmic curve fitting, this method has a larger error, often on the order of several micrometers, and cannot intuitively display the film's position, thus failing to achieve thickness visualization. Furthermore, this method requires manual calculation, resulting in low efficiency. When the sample film is subjected to stress, or when the film itself is curved rather than linear, this method cannot determine the degree of curvature and therefore cannot assess the impact of stress on the film.

[0087] This invention combines masking, Gaussian filtering, standard deviation peak finding, mode analysis, and connected component filtering to construct a complete automated detection process. This effectively suppresses noise interference caused by strong reflection and uneven scattering, achieving high-precision positioning of the film boundary. Compared to ultrasonic methods (millimeter-level resolution) and spectral reflectance methods (lacking spatial resolution), the OCT image-based detection method provides thickness distribution with micrometer-level accuracy within a millimeter-level range, making it more suitable for detecting weakly scattering materials. Compared to row traversal methods, this method avoids overlap errors when film layers are close together. Row traversal methods are prone to measurement deviations due to overlapping portions when bright lines are obliquely distributed. Compared to traditional OCT thickness detection methods, this method can visualize the thickness, more intuitively displaying the film layer's position, and can handle cases where the film layer is curved, making it more accurate and efficient. This method suppresses noise points by introducing Gaussian filtering, eliminates interference from non-target areas through masking, accurately identifies the film layer's position and number through standard deviation peak finding and mode analysis, further eliminates outliers through connected component filtering, and finally obtains the thickness value through curve fitting. Experimental results show that this method can simultaneously extract the thickness of all film layers in typical samples such as glass, mobile phone lenses, and eyeglass lenses in a single scan, significantly improving detection efficiency, achieving micron-level measurement accuracy, and controlling the error to below 2.2%, which is superior to methods such as ultrasound and spectral reflectance. This method not only expands the application scope of OCT technology in the detection of weak scattering and multilayer optical material thickness, but also confirms the core advantages of OCT in industrial non-destructive testing.

[0088] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0089] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting the thickness of optical components based on OCT images, implemented using an SD-OCT system, characterized in that, Includes the following steps: S1: Perform masking on the OCT image of the optical element under test to remove non-target areas and retain the target area; S2: Perform Gaussian filtering on the masked OCT image to suppress noise in the OCT image; S3: The standard deviation peak-finding algorithm is used to detect peak points in the OCT image after Gaussian filtering, and the positions of light intensity abrupt changes at the film interface are identified. S4: Perform mode analysis on the detected peak points to determine the number of film layers; S5: Perform connected component analysis on the peak points after mode analysis, filter out isolated peak points, and perform interpolation on missing positions; S6: Perform curve fitting on the peak points after connected component analysis to determine the position of each film layer, and calculate the actual thickness of the optical element under test based on the number of pixels between the film layers and the maximum imaging depth of the SD-OCT system.

2. The method for detecting the thickness of optical components based on OCT images according to claim 1, characterized in that, In step S3, the step of using the standard deviation peak-finding algorithm to detect peak points in the Gaussian-filtered OCT image includes: S31: Calculate the mean gray value of each column of pixels in the OCT image. and standard deviation : ； ； in, Indicates the number of valid pixels within the target area. This represents the grayscale value of the i-th valid pixel within each column of pixels; S32: Average of grayscale values ​​for each column of pixels and standard deviation Set an adaptive threshold for the grayscale value of each column of pixels. : ； in, This represents an adjustable empirical coefficient; S33: Traverse each pixel of the OCT image column by column, find pixels that satisfy the following two conditions, and mark them as peak points: Condition 1: The grayscale value of the pixel is greater than the adaptive threshold. ; Condition 2: The gray value of a pixel is greater than the gray values ​​of the pixels directly above and below it.

3. The method for detecting the thickness of optical components based on OCT images according to claim 1, characterized in that, In step S4, the step of performing mode analysis on the detected peak points includes: traversing all columns of the OCT image, counting the number of peak points in each column, and taking the mode of all peak points as the number of film layers.

4. The method for detecting the thickness of optical components based on OCT images according to claim 1, characterized in that, In step S5, the steps of performing connected component analysis on the peak points after mode analysis include: S51: Take each peak point as the center point, set a fixed-size neighborhood for each center point, count the other peak points in the neighborhood except the center point, and take them as neighborhood points; S52: By calculating the proximity of each center point to its neighboring points. : ； in, Indicates the center point. This indicates the number of neighboring points within the neighborhood. Represents a point within a neighborhood. Indicates the center point To neighboring points The distance; S53: Calculate the average proximity of all peak points. And compare it with the proximity of each center point, and then compare all the proximity values. Less than the average The center point is filtered out.

5. The method for detecting the thickness of optical components based on OCT images according to claim 1, characterized in that, In step S6, the step of curve fitting to the peak points after connected component analysis to determine the location of each film layer includes: S61: Set the number of lines in the fitted curve to the mode of the peak points; S62: Classify the peak points after connected component analysis using density clustering or distance clustering. S63: The peak points of each class are fitted with curves using the polynomial method or spline interpolation method. The position of each curve in the OCT image is the corresponding film layer position.

6. The method for detecting the thickness of optical components based on OCT images according to claim 1, characterized in that, In step S6, the actual thickness of the optical element under test is... The calculation formula is: ； in, This indicates the number of pixels between film layers. This indicates the number of pixels in the camera of the spectrometer in the SD-OCT system. This indicates the maximum imaging depth of the SD-OCT system. , Indicates the refractive index of the optical element under test. This indicates the center wavelength of the light source in the SD-OCT system. This indicates the resolution of the spectrometer. , This indicates the operating bandwidth of the spectrometer.

7. The method for detecting the thickness of optical components based on OCT images according to claim 1, characterized in that, In step S3, the minimum spacing between two adjacent peak points is set to 3 pixels.

8. An SD-OCT system for measuring the thickness of optical components, comprising: A light source used to emit a broadband spectrum; Fiber optic couplers are used to separate light emitted from a light source into measurement light and reference light. The sample arm and reference arm are used to transmit the measurement light and reference light, respectively; A spectrometer is used to acquire the interference signal between the measurement light and the reference light; The SD-OCT system is characterized by further including a computer for processing the interference signal acquired by the spectrometer and executing the optical element thickness detection method according to any one of claims 1 to 6.

Images