Method for detecting internal defects of superimposed optical waveguide

Abstract

The invention relates to a method for detecting internal defects of a laminated optical waveguide, which comprises the following steps: S300, focusing a visual module to an ith longitudinal focusing position in a Z-axis direction, detecting the internal defect of an ith layer of the laminated optical waveguide, and collecting an internal defect graph of the ith layer; s500, the visual module is focused to the (i + 1) th longitudinal focusing position in the Z-axis direction, internal defects of the (i + 1) th layer of the superimposed optical waveguide are detected, and an internal defect graph of the (i + 1) th layer is collected; s600, continuously iteratively executing the step S500 by enabling the updated i to be equal to i + 1 until the visual module is focused to the final position Zm in the Z-axis direction, detecting the internal defect of the mth layer of the superimposed optical waveguide, and collecting an internal defect map of the mth layer; and S700, according to the collected internal defect graphs from the first layer to the mth layer, generating a total graph of the internal defect graphs of the laminated optical waveguide from the first layer to the mth layer by using a longitudinal image synthesis algorithm. The detection accuracy is improved, and the detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of optical detection, and in particular to a method for detecting internal defects of a laminated optical waveguide. Background Art

[0002] This background generally presents the context of the present invention. To the extent described in this background, the work of the presently named inventors and aspects of the specification that may not constitute prior art at the time of filing are neither explicitly nor implicitly admitted to be prior art to the present invention.

[0003] In the field of augmented reality (AR) display technology, reflective waveguides have become a mainstream optical solution due to their high brightness output, excellent outdoor visibility, and excellent color reproduction. The technical principle of reflective waveguides is to transmit light through an array of semi-transparent and semi-reflective mirrors. The optical loss of reflective waveguides is significantly lower than that of diffraction waveguides. The color reproduction of reflective waveguides is closer to the visual effect of real environments.

[0004] The quality and function of reflective optical waveguides directly affect the user experience of AR devices. In order to meet the display requirements of large viewing angles or high-quality color images, reflective optical waveguides cut from stacked multi-layer optical laminated glass are used, also known as laminated optical waveguides. However, when manufacturing laminated optical waveguides, if there are tiny defects inside the laminated glass, it will directly cause black spots, bright spots or distortion in the user's field of view of the AR device, seriously reducing the user experience of the AR device. In the process of stacking multi-layer optical laminated glass, defects such as dust, foreign matter, and overflowed glue may exist between adjacent laminated glass sheets, and individual laminated glass sheets themselves may also have defects such as pitting and scratches. Therefore, high-precision three-dimensional detection of internal defects in laminated optical waveguides is a key technology to ensure the display quality of AR devices.

[0005] However, existing manual microscopic inspection methods rely on manual visual inspection through a microscope during the semi-finished stage of laminated optical waveguides, relying on the operator's subjective judgment of internal defects. Existing manual microscopic inspection methods are time-consuming, labor-intensive, and inefficient, as well as highly subjective, with low accuracy and a high false positive rate. Microscope imaging systems have a shorter working distance than telecentric lenses, and the longitudinal mechanical focusing method has a limited focusing distance. Existing manual microscopic inspection methods are unable to quantify the three-dimensional coordinates of internal defects and lack defect annotation capabilities. These methods are unable to meet the requirements for high-precision inspection of internal defects in reflective laminated optical waveguides with large dimensions (50 mm × 50 mm), large thicknesses (up to 50 mm), and multi-layer structures (30 layers). They are unable to determine the three-dimensional coordinates of internal defects within the laminated optical waveguide, and in particular, are unable to accurately locate the defect in which layer within the laminated glass.

[0006] In view of the above problems, it is necessary to develop a method for detecting internal defects of laminated optical waveguides, which has high detection efficiency, accurate positioning of internal defects, and can provide complete internal defect information. Summary of the Invention

[0007] One purpose of the present invention is to solve the problem of high detection efficiency, accurate positioning of internal defects, and provision of complete internal defect information during internal defect detection of a laminated optical waveguide, thereby achieving efficient and accurate internal defect detection.

[0008] In order to achieve the above-mentioned objectives, a first aspect of the present invention provides a method for detecting internal defects of a laminated optical waveguide, which comprises the following steps: S300: The vision module focuses at an X-axis and Y-axis position in the Z-axis direction to an i-th longitudinal focus position, detects internal defects in the i-th layer of the stacked optical waveguide, and collects an i-th internal defect image; where 1 ≤ i ≤ m; when i=1, the i-th longitudinal focus position is the initial position Z1; and m is the number of steps; S500: The vision module focuses on the i+1th longitudinal focus position in the Z-axis direction at the X-axis and Y-axis positions, detects internal defects in the i+1th layer of the laminated optical waveguide, and collects an internal defect image of the i+1th layer; wherein, for the internal defect image of the i+1th layer, the i+1th longitudinal focus position is a position moved downward along the Z-axis from the i-th longitudinal focus position by a Z-axis step interval ΔZ; S600: Continue to iterate step S500 by setting the updated i=i+1 until the vision module focuses on the final position Z in the Z-axis direction. m , detect the internal defects of the mth layer of the stacked optical waveguide and collect the internal defect map of the mth layer; among them, the final position Z m is a position that moves downward along the Z axis by a cumulative number of m-1 Z-axis step intervals ΔZ from the initial position Z1; S700: generating a total map of the internal defect maps of the first to mth layers of the laminated optical waveguide at the X-axis and Y-axis positions by using a longitudinal image synthesis algorithm based on the collected internal defect maps of the first to mth layers.

[0009] Compared with the prior art, the method of the present invention has the following beneficial technical effects: The present invention uses a large-area camera, a high-precision telecentric lens, and a customized high-collimation backlight source to achieve high-precision detection of internal defects in laminated optical waveguide products. Through the optical calibration method of the altimeter and the high-precision telecentric lens, the longitudinal initial position Z1 can be automatically located. The present invention uses the product of the depth of field L of the lens and the refractive index n of the sample as the Z-axis step interval, L×n, taking into account the change in depth of field caused by the refractive index when scanning inside the sample, which can ensure the clarity continuity of two adjacent images while taking into account efficiency; Through optical calibration and active displacement compensation function of X-axis and Y-axis, the vertical focus position (Z1 to Z m ) The internal defect position offset of the image at the same X-axis and Y-axis position can be controlled at the μm level; The longitudinal image synthesis algorithm included in the present invention fuses multiple longitudinal images into a single overall image based on a clarity algorithm and locates the precise XYZ three-dimensional position of the defect. The longitudinal precision is the product of the depth of field of the high-precision telecentric lens and the refractive index n of the sample. The longitudinal image synthesis algorithm uses a focusing method to restore the three-dimensional information of the object by comparing images of the same scene captured at different longitudinal focus positions. Specifically, the algorithm analyzes the clarity of each pixel at different longitudinal focus positions and selects the longitudinal focus position with the highest clarity as the depth value for that pixel. Ultimately, the sharpest portions of all scanned longitudinal focus positions are synthesized into a single overall image.

[0010] Further areas of applicability will become apparent from the description provided herein.It should be understood that the description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present disclosure.

[0011] The above-described features and advantages and other features and advantages of the presently disclosed systems and methods are readily apparent from the following detailed description, including the claims and exemplary embodiments, when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The following is a brief introduction to the drawings required for use in specific implementations or descriptions of the prior art.

[0013] Obviously, some of the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0014] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, more details, features and advantages of the present invention are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 Schematic diagram of a system for detecting internal defects of a laminated optical waveguide according to the present invention; Figure 2 1 is a schematic diagram of an embodiment of a longitudinal step-scanning step of a method for detecting internal defects of a laminated optical waveguide according to the present invention; Figure 3A schematic diagram of a specific embodiment of a longitudinal step-scan sequence for each region in the X-axis and Y-axis directions; Figure 4 is a schematic diagram of a specific embodiment of a complete internal defect distribution map of a laminated optical waveguide sample to be tested; Figure 5 is a schematic diagram of a specific embodiment of a summary table of internal defect information of a laminated optical waveguide sample to be tested; Figure 6 is Figure 3 A diagram showing the longitudinal variation of a defect obtained by scanning area 1 and a schematic diagram of the defect location identified by the clarity algorithm. DETAILED DESCRIPTION

[0015] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other.

[0016] In the following description, many specific details will be explained with reference to several examples of the present invention illustrated in the accompanying drawings to facilitate a full understanding of the present invention. Whenever possible, the same or similar reference numerals are used in the drawings and the description to represent the same or similar components or steps.

[0017] However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited by the specific embodiments disclosed below.

[0018] In a specific embodiment of the present invention, a method for detecting internal defects of a laminated optical waveguide is provided, comprising the following steps: S300: The vision module focuses at an X-axis and Y-axis position in the Z-axis direction to an i-th longitudinal focus position, detects internal defects in the i-th layer of the stacked optical waveguide, and collects an i-th internal defect image; where 1 ≤ i ≤ m; when i=1, the i-th longitudinal focus position is the initial position Z1; and m is the number of steps; S500: The vision module focuses on the i+1th longitudinal focus position in the Z-axis direction at the X-axis and Y-axis positions, detects internal defects in the i+1th layer of the laminated optical waveguide, and collects an internal defect image of the i+1th layer; wherein, for the internal defect image of the i+1th layer, the i+1th longitudinal focus position is a position moved downward along the Z-axis from the i-th longitudinal focus position by a Z-axis step interval ΔZ; S600: Continue to iterate step S500 by setting the updated i=i+1 until the vision module focuses on the final position Z in the Z-axis direction. m, detect the internal defects of the mth layer of the stacked optical waveguide and collect the internal defect map of the mth layer; among them, the final position Z m is a position that moves downward along the Z axis by a cumulative number of m-1 Z-axis step intervals ΔZ from the initial position Z1; S700: generating a total map of the internal defect maps of the first to mth layers of the laminated optical waveguide at the X-axis and Y-axis positions by using a longitudinal image synthesis algorithm based on the collected internal defect maps of the first to mth layers. Figure 1 FIG. 1 is a schematic diagram of a system for detecting internal defects of a laminated optical waveguide according to the present invention. Figure 1 As shown, the detection system for internal defects of laminated optical waveguides includes: a vision module, a motion control module and an image fusion module. The laminated optical waveguide sample 2 to be tested is placed on a hollow carrier 1.

[0019] The vision module includes a large-area camera 3, a high-precision telecentric lens 4, an altimeter 5, and a custom backlight 6. After optical calibration, the altimeter automatically locates the initial longitudinal position Z1 based on the light signal reflected from the upper surface of the laminated optical waveguide sample under test. Z1 corresponds to the upper surface of the laminated optical waveguide sample 2 under test. The custom backlight 6, which comprises a point light source and a custom focusing lens, offers excellent collimation. Compared to conventional backlights and coaxial light sources, this highly collimated custom backlight offers better results for inspecting the interior of glass.

[0020] The motion control module can control the XYZ axes separately, and is used to control the movement of the vision module on the X, Y, and Z axes.

[0021] The image fusion module can realize functions such as longitudinal image synthesis algorithm, image stitching algorithm and defect information summary.

[0022] like Figure 1 As shown, the laminated optical waveguide product 2 is made of multiple layers of laminated glass. The length and width of the product range from 20 mm to 50 mm, the thickness can reach up to 50 mm, and the number of layers can reach up to 30. During the stacking process, defects such as dust, foreign matter, and glue overflow may exist between adjacent glasses, and the glass itself may also have defects such as pitting and scratches. Figure 2 It is a schematic diagram of an embodiment of the longitudinal step-scanning step of the method for detecting internal defects of a laminated optical waveguide according to the present invention.

[0023] In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, as Figure 2 As shown, the image is scanned downward in the Z-axis direction to collect images at each longitudinal focus position. The longitudinal focus positions are Z1, Z2, etc., and the image is finally collected at the bottom surface of the sample. The focus position is Z m , a total of m internal defect images are obtained; Then, the vertical image synthesis algorithm is used to detect the clarity of each layer of the collected image. The vertical image synthesis algorithm compares the same XY position at different vertical focus positions of the lens (Z1 to Z m ) and uses a focusing method to restore the three-dimensional information of the object. Specifically, the vertical image synthesis algorithm analyzes the clarity or sharpness of each pixel at different vertical focus positions and selects the vertical focus position with the highest clarity as the depth value for that pixel.

[0024] Finally, all the different vertical focus positions (Z1 to Z m ) Only the clear parts of the image are retained to form a complete picture. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, step S700 includes the following steps: S710: Analyze the clarity of the pixels of the collected internal defect images of each layer at each longitudinal focus position; S730: Comparing the clarity of the pixels of the internal defect map of each layer at each longitudinal focus position; S750: selecting a longitudinal focus position corresponding to a pixel of the internal defect image with the highest definition as the depth of the internal defect of the laminated optical waveguide; S770: For the pixel points of the internal defect map of each layer, select the pixel point of the internal defect map with the highest definition as the total pixel point of the internal defect map of the first to mth layers of the laminated optical waveguide; S790: Merging all total pixel points of the internal defect maps of the first to mth layers of the stacked optical waveguide into a total map of the internal defect maps of the first to mth layers of the stacked optical waveguide.

[0025] In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, after step S700, the following steps are further included: S800: The vision module moves from the X-axis and Y-axis position to the next X-axis and Y-axis position by means of the XY motion platform; S850: The vision module continues to execute steps S300-S700 at the next X-axis and Y-axis position, collects a general image of the internal defects of the laminated optical waveguide, and then proceeds to step S800 until a collection of general images of the internal defects of the laminated optical waveguide at all X-axis and Y-axis positions is obtained; S900: performing image stitching on a collection of total images of internal defects of the laminated optical waveguide at all X-axis and Y-axis positions by using an image stitching algorithm to obtain a complete defect distribution map of internal defects of the laminated optical waveguide. Figure 3 This is a specific embodiment of the longitudinal step scanning sequence of the divided regions in the X-axis and Y-axis directions.

[0026] In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, as Figure 3 As shown, the size of the laminated optical waveguide samples to be tested is generally larger than the field of view of the vision module. Therefore, an XY motion platform is used to capture images to obtain an atlas, and then an image stitching algorithm is used to stitch the images together. Specifically, the images are hard-stitched, and a weighted fusion algorithm is used to address the problem of stitching seams.

[0027] Taking 3×4 stitching as an example, Figure 3 The S-shaped scanning sequence of the X-axis and Y-axis directions can increase the scanning speed. Figure 3 In the example, the cross section of the laminated optical waveguide is divided into four rows and three columns of rectangular boxes. Along the X-axis, from left to right, the first row consists of Regions 1, 2, and 3; the second row consists of Regions 6, 4, and 4; the third row consists of Regions 7, 8, and 9; and the fourth row consists of Regions 12, 11, and 10. A total of 3 × 4 = 12 regions need to be scanned, with each region being scanned longitudinally along the Z-axis in the following order: Region 1 -> Region 2 -> Region 3 -> Region 4 -> Region 5 -> Region 6 -> Region 7 -> Region 8 -> Region 9 -> Region 10 -> Region 11 -> Region 12.

[0028] In a specific embodiment of the present invention's method for detecting internal defects in laminated optical waveguides, a large-area camera and a high-magnification, narrow-depth-of-field lens are employed to achieve high-precision defect detection. An altimeter automatically determines the initial longitudinal position based on the light signal reflected from the upper surface of the laminated optical waveguide sample under test. The specific configuration includes: a 65M camera; a 2x telecentric lens; a custom, highly collimated backlight; and a 15 mm × 11 mm field of view.

[0029] In one embodiment of the method for detecting internal defects in a laminated optical waveguide of the present invention, a height gauge is used to measure the light signal reflected from the upper surface of the laminated optical waveguide sample, and the height value C is read. The inverse of this height value C is the initial position Z1. For example, if the reading is 5 mm, the Z axis should be moved -5 mm. In this case, the initial position is the upper surface of the sample to be tested. Figure 4 、 Figure 5 and Figure 6 It is the complete test data and measurement results.

[0030] like Figure 3 As shown in the figure, assuming the sample size is 40×40 mm and the field of view of this solution is 15 mm×11 mm, then the X direction needs to move 3 times and the Y direction needs to move 4 times, and a total of 3×4=12 areas need to be scanned, each time one area is scanned longitudinally along the Z axis.

[0031] First Figure 3The vertical step scanning is performed on area 1 of the sample. Assuming that the sample thickness is 15 mm, the refractive index is 1.5, and the depth of field of the 2X lens is 0.5 mm, then according to the optical path theory, the equivalent depth of field is 0.5 mm×1.5=0.75 mm, and each step of the Z axis is 0.75 mm. The Z axis needs to move 15 mm / 0.75=20 times, so in Figure 3 The Z axis needs to move down 20 times when scanning area 1.

[0032] Then Figure 3 Scan area 2, and move the X axis so that the field of view of the optical device is Figure 3 For area 2, scan from the bottom of the sample, also 20 times.

[0033] This method is used to scan the entire sample longitudinally and obtain a detection atlas of this sample. This embodiment is only an example. When the lens and camera are replaced, the specific number of steps and the step interval will change, but the calculation method remains unchanged. The calculation method of the Z-axis step number is protected.

[0034] The vertical image synthesis algorithm will be Figure 3 The image collected in area 1 is inspected for clarity, defects in each layer are identified and cut out, and a defect distribution map is synthesized; similarly, corresponding defect distribution maps are synthesized in other areas.

[0035] Then, the defect map collection synthesized in each area is spliced ​​by position to obtain the following: Figure 4 Defect distribution map for the entire product is shown.

[0036] Finally available Figure 4 and Figure 5 . Figure 4 It is a specific embodiment of a complete internal defect distribution map of the laminated optical waveguide sample to be tested. Figure 5 This is a specific embodiment of the internal defect information summary table of the laminated optical waveguide sample to be tested. Figure 4 The defect information summary table shown includes the defect number, X coordinate, Y coordinate, Z coordinate, defect diameter, defect length, and defect width. In a specific embodiment of the method for detecting internal defects in a laminated optical waveguide of the present invention, the defect information summary table (not shown) also includes the defect type.

[0037] Figure 6 is Figure 3 A diagram showing the longitudinal variation of a defect obtained by scanning area 1 and a schematic diagram of the defect location identified by the clarity algorithm.

[0038] According to the clarity algorithm and Figure 4The complete defect distribution map shown can locate the exact longitudinal position of the defect, so the 3D information of the defect can be determined from the 2D image, and the detection results can be output through the defect information summary function. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, after step S500, the following steps are further included: S520: Compare the X-axis pixel offset Δx and the Y-axis pixel offset Δy of the internal defect in the (i+1)th layer of the stacked optical waveguide with the internal defect in the i-th layer of the stacked optical waveguide; S540: Multiply the pixel equivalent δ of the visual module of the automatic detection device by the X-axis pixel offset Δx and the Y-axis pixel offset Δy to obtain the X-axis physical compensation D x and Y-axis physical compensation D y ; S560: The visual module of the automatic detection device moves linearly in the reverse direction of the X-axis in the X-axis physical compensation amount D x And move the Y-axis physical compensation amount D in the reverse linear direction of the Y-axis y . Through detailed optical calibration methods, the parallelism between the imaging plane of the high-precision telecentric lens and the carrier of the superimposed optical waveguide product to be measured can be adjusted to no more than 0.5 degrees.

[0039] However, for high-precision telecentric lenses of 2x and above, 0.5 degrees of parallelism will still result in different longitudinal focus positions (Z1 to Z m ) Defects at the same position in the image will result in obvious positional offset, affecting the effect of the longitudinal image synthesis algorithm.

[0040] The X-axis and Y-axis active displacement compensation function ensures that the focus can be adjusted at different vertical focus positions (Z1 to Z m ) Accuracy and stability of image acquisition.

[0041] Even if a higher precision telecentric lens with a factor of 2 or more is replaced, the position offset of defects in the same position of the image at different longitudinal focus positions (Z1 to Zm) can still be controlled within the μm level.

[0042] In a specific embodiment of the method for detecting internal defects in laminated optical waveguides of the present invention, during the inspection process, a characteristic defect is selected from captured images. The pixel offset and offset direction of two adjacent images of the characteristic defect when clearly imaged are first determined. Then, the actual X- and Y-axis offset distances for each Z-axis movement are calculated based on the pixel equivalent δ (i.e., the actual length corresponding to each pixel). This is typically 1-10 μm. The pixel equivalent δ establishes a proportional relationship between pixels and physical dimensions, calibrating the actual physical dimension represented by each pixel. The pixel equivalent δ is determined by the lens magnification and the pixel size of the vision module. For example, the pixel equivalent δ for a 2X lens is 1.6 μm / px. The pixel offset and offset direction of the characteristic defect, as well as the pixel equivalent δ, are input. The actual X- and Y-axis offset distances are calculated. The output is used to guide the motion control module to perform linear motion compensation on the X and Y axes during Z-axis motion capture. This compensation is used to control the X and Y axes to move a corresponding distance in the direction opposite to the defect offset. The vision module performs μm-level linear motion in the X-axis and Y-axis directions during or before the Z-axis motion, and the motion direction is opposite to the offset direction of the defect.

[0043] Beneficial technical effects: This solution solves the problem that defects at the same position in images at different longitudinal focus positions will be offset due to the parallelism problem of the high-precision telecentric lens, affecting the effect of the longitudinal image synthesis algorithm. Compensates for defect position offset caused by lens tilt (parallelism ≥ 0.5 degrees) or high-magnification imaging to ensure pixel alignment during longitudinal image synthesis. The XY-axis active displacement compensation function ensures that the XY-axis offset of the defect position at the same position in images at different longitudinal focus positions is controlled at the micron level, improving the accuracy and stability of image acquisition, thereby improving the accuracy of defect detection. The image offset problem of high-magnification lenses (≥2X) is handled, resulting in the defect position error in the composite image being controlled within 10μm, which meets the requirements for inter-layer defect differentiation in 30-layer glass laminates.

[0044] Supporting high-magnification lenses of 2X and above, the system improves inspection accuracy from 32μm to 4.8μm. To meet customer requirements for different inspection accuracy, interchangeable high-precision telecentric lenses of varying magnifications can be used. For example, if the defect inspection accuracy requirement is 4.8μm, a 2X telecentric lens can be used with a 65M camera with a 3.2μm pixel size and 1.6μm single-pixel accuracy. The measured depth of field is 500μm. For example, if the defect inspection accuracy requirement is 32μm, a 0.5X telecentric lens can be used with a 65M camera with a 3.2μm pixel size and 6.4μm single-pixel accuracy. The measured depth of field is 1500μm.

[0045] In a specific embodiment of the method for detecting internal defects of the laminated optical waveguide of the present invention, the visual module again collects images of each longitudinal position of the sample to verify the position from Z1 to Z m Whether the pixel shift of the same characteristic defect in all images is controlled at the μm level. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, before step S300, the following steps are further included: S100: Using an optical calibration method, the reference plane of the altimeter and the imaging plane of the vision module are aligned in the same plane; S200: The vision module is automatically positioned at the initial position Z1 by using the altimeter. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, as Figure 1 As shown, in step S100, the optical calibration method for the altimeter and the imaging plane of the high-precision telecentric lens to be at the same position is: First, place the chessboard on the hollowed-out carrier and use a high-precision telecentric lens to image the chessboard. Adjust the Z axis to drive the large-target camera and the high-precision telecentric lens to move together. When the center of the chessboard is clearly imaged, that is, when there are no more than 3 transition pixels at the junction of the black and white checkerboard, it indicates that the high-precision telecentric lens is focused on the upper surface of the chessboard. Then adjust the X and Y axes so that the center of the chessboard is at the center of the image captured by the camera. The camera's tilt angle is then adjusted so that the number of transition pixels at the black and white junctions on the four sides of the chessboard is no more than three, indicating that the imaging surface of the camera lens is parallel to the plane of the chessboard, and the parallel angle is no more than 0.5 degrees. The X and Y axes are then moved, while the Z axis is kept stationary, so that the altimeter's light is at the exact center of the chessboard. The altimeter is used to measure the height of the chessboard, and the fixing mechanism of the altimeter is adjusted so that the altimeter reading is 0. This ensures that the imaging surface of the altimeter is consistent with that of the high-precision telecentric lens.

[0046] Next, in S200, the method uses an altimeter to automatically locate the initial position Z1 of the vision module: replace the checkerboard with the laminated optical waveguide sample to be tested, use the altimeter to measure the light signal reflected from the upper surface of the laminated optical waveguide sample to measure the reading C, and adjust the Z axis in the opposite direction by a distance C to make the altimeter read 0. At this point, the imaging plane of the high-precision telecentric lens is located at the upper surface of the laminated optical waveguide sample to be tested. The initial Z axis position Z1 at this time is recorded. This calibration method can automatically locate the longitudinal initial position Z1 using the altimeter, ensuring that the imaging plane of the high-precision telecentric lens is located at the upper surface of the laminated optical waveguide sample to be tested. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, step S100 includes the following steps: S120: Place the chessboard on the hollow vehicle; S140: The vision module focuses on the upper surface of the chessboard in the Z-axis direction; the X-axis and Y-axis are adjusted so that the center of the chessboard is located at the center of the image captured by the vision module; S180: Adjust the tilt angle of the vision module so that the imaging surface of the vision module is parallel to the plane of the chessboard; adjust the X-axis and Y-axis so that the light of the altimeter is in the center of the chessboard, use the altimeter to measure the height of the chessboard, adjust the altimeter reading to zero, so that the reference plane of the altimeter and the imaging plane of the vision module are in the same plane.

[0047] The imaging plane of the high-precision telecentric lens is made coplanar with the altimeter reference plane of the altimeter through optical calibration; specifically, the steps S120, S140 and S180 are included.

[0048] In a specific embodiment of the method for detecting internal defects in a laminated optical waveguide according to the present invention, in step S140, a vision module is used to image a checkerboard. The Z axis is adjusted to move the large-area camera and the vision module together, so that the center of the checkerboard is clearly imaged (i.e., the number of transition pixels at the black-white junction in the center of the checkerboard is ≤ 3). At this point, the vision module focuses on the upper surface of the checkerboard along the Z axis. The X and Y axes are then adjusted so that the exact center of the checkerboard coincides with the exact center of the image captured by the vision module.

[0049] In a specific embodiment of the method for detecting internal defects in a laminated optical waveguide according to the present invention, in step S180, the tilt angle of the vision module is adjusted so that the number of transition pixels at the black and white intersections on all four sides of the checkerboard is ≤3, indicating that the imaging surface of the vision module is parallel to the plane of the checkerboard (i.e., the parallelism error is ≤0.5°). The X and Y axes are then moved, while the Z axis is kept stationary, so that the light from the altimeter is centered on the checkerboard. The altimeter is used to measure the height of the checkerboard. The fixing mechanism for the altimeter is adjusted so that the altimeter reading is 0. This completes the coplanar alignment between the altimeter's reference plane and the imaging surface of the high-precision telecentric lens. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, step S200 includes the following steps: S220: Using a height gauge, measure the light signal reflected from the upper surface of the stacked optical waveguide to obtain a reading C. S250: Adjust the Z axis to move a distance C in the opposite direction to return the altimeter reading to zero. At this time, the imaging plane of the vision module is located at the upper surface of the stacked optical waveguide to be measured, so that the vision module automatically locates the initial position Z1.

[0050] In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, in step S220, the checkerboard is replaced with the laminated optical waveguide sample to be tested, and a height reading C is measured using a height meter to measure the light signal reflected from the upper surface of the laminated optical waveguide sample to be tested.

[0051] In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, in step S250, the Z-axis is adjusted to move in the opposite direction by a distance C so that the height value of the altimeter reads 0, returning to zero. At this time, the imaging plane of the high-precision telecentric lens is located at the upper surface of the laminated optical waveguide sample being measured. The Z-axis position at this time is recorded as the initial position Z1, so that the vision module automatically locates the initial position Z1 (i.e., the first longitudinal focus position).

[0052] Beneficial Technical Effect: The Z1 initial position optical positioning method solves the problem of inaccurate positioning of transparent material upper surfaces, achieving a positioning accuracy of ≤1μm and establishing a reference surface for longitudinal step scanning. Altimeter zeroing replaces traditional focusing, converting optical plane matching into mechanical displacement and outputting the Z1 initial position coordinates. This step directly addresses the existing technology's inability to quickly locate the upper surface of thick samples. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, the vision module is a high-precision telecentric lens.

[0053] Beneficial technical effect: Using the altimeter to automatically locate the longitudinal initial position Z1 can ensure that the imaging plane of the high-precision telecentric lens is exactly on the upper surface of the measured superimposed optical waveguide sample.

[0054] like Figure 2As shown, focusing is achieved via Z-axis motion, with Z1 as the initial position. The Z-axis step interval is the product of the lens's depth of field L and the sample's refractive index n. To meet customer requirements for different inspection accuracy, high-precision telecentric lenses of varying magnifications can be used. For example, if the defect inspection accuracy requirement is 4.8µm, a 2X telecentric lens and a 65M camera with a 3.2µm pixel size and 1.6µm single-pixel accuracy can meet the inspection requirement. The depth of field is measured to be 500µm. For example, if the defect inspection accuracy requirement is 32µm, a 0.5X telecentric lens and a 65M camera with a 3.2µm pixel size and 6.4µm single-pixel accuracy can be used. The depth of field is measured to be 1500µm. To ensure clear imaging of all Z-axis defects within the laminated waveguide sample during longitudinal step scanning and to maintain the continuity of the atlas, according to optical principles, since the defects scanned are within the laminated waveguide, the optical path F = n × H, where n is the refractive index and H is the physical distance traveled by the light. Similarly, the equivalent depth of field for internal measurements can be derived. The product of the lens's depth of field, L, and the sample's refractive index, n, L×n, is the Z-axis stepping interval. This Z-axis stepping interval is the equivalent depth of field. This theory can be verified by actual measurements on a machine and calculations from specific examples. The number of Z-axis steps is h / (n×L), where h is the thickness of the laminated optical waveguide sample being measured. The lens' depth of field, L, is generally provided by the lens supplier but can also be measured. The sample's refractive index, n, is provided by the customer. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, in step S500 , the step interval ΔZ is obtained based on the product of the depth of field L and the refractive index n.

[0055] In a specific embodiment of the method for detecting internal defects in a laminated optical waveguide according to the present invention, to ensure that all internal defects in the laminated optical waveguide sample along the Z-axis are clearly imaged during longitudinal step scanning and to maintain the continuity of the atlas, according to optical principles, since the defects within the laminated optical waveguide are scanned, the optical path F = n × H, where n is the refractive index and H is the physical distance traveled by the light.

[0056] In a specific embodiment of the present invention's method for detecting internal defects in a laminated optical waveguide, to ensure clear imaging of all internal defects along the Z-axis of the laminated optical waveguide sample during longitudinal step scanning and to maintain the continuity of the atlas, an equivalent depth of field (DFO) is derived based on optical principles. The Z-axis step interval ΔZ is determined by multiplying the lens's depth of field L by the sample's refractive index n (DFO). This DFO is the equivalent depth of field. This DFO is determined by multiplying the lens's depth of field L by the sample's refractive index n (DFO = L × n), where L is the inherent depth of field parameter of the telecentric lens (provided by the supplier or measured), and n is the material refractive index of the laminated optical waveguide sample being tested (entered by the user or measured). Assuming the sample thickness H is 15 mm, the refractive index n is 1.5, and the depth of field L of the 2X lens is 0.5 mm, then according to the optical path theory, the equivalent depth of field is L×n = 0.5 mm×1.5=0.75 mm, and each step of the Z axis is a step interval ΔZ=0.75 mm.

[0057] Beneficial Technical Effect: The Z-axis step interval ΔZ is used to guide the subsequent Z-axis stepping movement of the vision module, ensuring accurate focus at each location within the stacked optical waveguide during scanning and capturing clear images. This solves the problem in existing technologies where the Z-axis step interval cannot be accurately determined, resulting in inaccurate focus and unclear image acquisition. By taking into account the depth of field changes caused by the refractive index when scanning within the sample, it can ensure the clarity continuity of two adjacent images while also taking efficiency into account, thereby improving the accuracy and efficiency of defect detection. The physical depth of field L is converted into an equivalent depth of field using the refractive index n. Input: supplier-provided lens depth of field L + customer-provided sample refractive index n; Output: Z-axis step interval ΔZ. Z-axis positioning accuracy reaches 0.75 mm (example value), and the efficiency of longitudinal step scanning in the Z-axis is improved. In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, the number of steps m is proportional to the thickness H of the laminated optical waveguide and inversely proportional to the step interval ΔZ.

[0058] In one specific embodiment of the method for detecting internal defects in a laminated optical waveguide according to the present invention, the number of Z-axis steps, m, equals H / ΔZ, which is the sum of H / (n×L), rounded upwards. h represents the thickness of the laminated optical waveguide sample to be tested. Assuming the sample thickness H is 15 mm, the refractive index n is 1.5, and the depth of field L of the 2X lens is 0.5 mm, then according to optical path theory, the equivalent depth of field is L×n = 0.5 mm×1.5 = 0.75 mm. The stepping interval ΔZ for each Z-axis step is 0.75 mm, and the Z-axis requires 20 movements (15 mm / 0.75 mm). Therefore, the vision module must move downward in the Z-axis 20 times during the step-scan process.

[0059] Beneficial Technical Effects: Calculation of the number of steps m: Automated layering is achieved using the formula H / (L×n), eliminating manual intervention and improving inspection efficiency. In practical examples, only 20 scans were required for a 15 mm thick sample, significantly reducing the time required by traditional methods. Full-layer scanning of a 50 mm thick sample was also achieved.

[0060] In a specific embodiment of the method for detecting internal defects of a laminated optical waveguide of the present invention, after step S700, the following steps are further included: S950: Based on the collected internal defect images of the 1st to mth layers, a defect information table is generated; the defect information table includes: defect serial number, defect location ID, defect depth, defect coordinate Z, defect coordinate X, defect coordinate Y, defect diameter, defect length, defect width and defect type.

[0061] In a specific embodiment of the method for detecting internal defects in a laminated optical waveguide according to the present invention, in step S950, a defect information table is generated based on the collected internal defect images of layers 1 to m. The defect information table includes the defect number, defect location ID, defect depth, defect coordinate Z, defect coordinate X, defect coordinate Y, defect diameter, defect length, defect width, and defect type. Specifically, the defect information table includes the three-dimensional coordinates (X, Y, Z) of all defects and the morphological parameters (length, width, and diameter) of all defects.

[0062] Beneficial Technical Effect: This technology solves the quality traceability issue for laminated optical waveguides, enabling the identification of internal defects, their location on the layer, and their severity. The defect type can also be clearly indicated in the defect information table, for example, whether it is a pit or scratch on the glass itself, or dust, foreign matter, or glue overflow between adjacent laminated glass sheets. The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to the above embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0064] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0065] The foregoing description is merely illustrative and is in no way intended to limit the invention, its application, or uses. The broad teachings of the invention can be implemented in many forms. Therefore, although this disclosure includes specific examples, the scope of the invention should not be limited thereto, as other modifications will become apparent upon a study of the drawings, the specification, and the appended claims.

Claims

1. A method for detecting internal defects of a laminated optical waveguide, characterized in that: The following steps are involved: S300: The vision module focuses at an X-axis and Y-axis position in the Z-axis direction to an i-th longitudinal focus position, detects internal defects in the i-th layer of the stacked optical waveguide, and collects an i-th internal defect image; where 1 ≤ i ≤ m; when i=1, the i-th longitudinal focus position is the initial position Z1; and m is the number of steps; S500: The vision module focuses on the i+1th longitudinal focus position in the Z-axis direction at the X-axis and Y-axis positions, detects internal defects in the i+1th layer of the laminated optical waveguide, and collects an internal defect image of the i+1th layer; wherein, for the internal defect image of the i+1th layer, the i+1th longitudinal focus position is a position moved downward along the Z-axis from the i-th longitudinal focus position by a Z-axis step interval ΔZ; S600: Continue to iterate step S500 by setting the updated i=i+1 until the vision module focuses on the final position Z in the Z-axis direction. m , detect the internal defects of the mth layer of the stacked optical waveguide and collect the internal defect map of the mth layer; among them, the final position Z m is a position that moves downward along the Z axis by a cumulative number of m-1 Z-axis step intervals ΔZ from the initial position Z1; S700: generating a total map of the internal defect maps of the first to mth layers of the laminated optical waveguide at the X-axis and Y-axis positions by using a longitudinal image synthesis algorithm based on the collected internal defect maps of the first to mth layers.

2. The detection method according to claim 1, wherein Step S700 includes the following steps: S710: Analyze the clarity of the pixels of the collected internal defect images of each layer at each longitudinal focus position; S770: For the pixel points of the internal defect map of each layer, select the pixel point of the internal defect map with the highest definition as the total pixel point of the internal defect map of the first to mth layers of the laminated optical waveguide; S790: Merging all total pixel points of the internal defect maps of the first to mth layers of the stacked optical waveguide into a total map of the internal defect maps of the first to mth layers of the stacked optical waveguide.

3. The detection method according to claim 1, wherein After step S700, the following steps are further included: S800: The vision module moves from the X-axis and Y-axis position to the next X-axis and Y-axis position by means of the XY motion platform; S850: The vision module continues to execute steps S300-S700 at the next X-axis and Y-axis position, collects a general image of the internal defects of the laminated optical waveguide, and then proceeds to step S800 until a collection of general images of the internal defects of the laminated optical waveguide at all X-axis and Y-axis positions is obtained; S900: performing image stitching on a collection of total images of internal defects of the laminated optical waveguide at all X-axis and Y-axis positions by using an image stitching algorithm to obtain a complete defect distribution map of internal defects of the laminated optical waveguide.

4. The detection method according to claim 1, wherein After step S500, the following steps are further included: S520: Compare the X-axis pixel offset Δx and the Y-axis pixel offset Δy of the internal defect in the (i+1)th layer of the stacked optical waveguide with the internal defect in the i-th layer of the stacked optical waveguide; S540: Multiply the pixel equivalent δ of the visual module of the automatic detection device by the X-axis pixel offset Δx and the Y-axis pixel offset Δy to obtain the X-axis physical compensation D x and Y-axis physical compensation D y ; S560: The visual module of the automatic detection device moves linearly in the reverse direction of the X-axis in the X-axis physical compensation amount D x And move the Y-axis physical compensation amount D in the reverse linear direction of the Y-axis y .

5. The detection method according to claim 1, wherein Before step S300, the following steps are also included: S100: Using an optical calibration method, the reference plane of the altimeter and the imaging plane of the vision module are aligned in the same plane; S200: The vision module is automatically positioned at the initial position Z1 by using the altimeter.

6. The detection method according to claim 5, characterized in that Step S100 includes the following steps: S140: The vision module focuses on the upper surface of the chessboard in the Z-axis direction; the X-axis and Y-axis are adjusted so that the center of the chessboard is located at the center of the image captured by the vision module; S180: Adjust the tilt angle of the vision module so that the imaging surface of the vision module is parallel to the plane of the chessboard; adjust the X-axis and Y-axis so that the light of the altimeter is in the center of the chessboard, use the altimeter to measure the height of the chessboard, adjust the altimeter reading to zero, so that the reference plane of the altimeter and the imaging plane of the vision module are in the same plane.

7. The detection method according to claim 5, characterized in that Step S200 includes the following steps: S220: Using a height gauge, measure the light signal reflected from the upper surface of the stacked optical waveguide to obtain a reading C. S250: Adjust the Z axis to move a distance C in the opposite direction to return the altimeter reading to zero. At this time, the imaging plane of the vision module is located at the upper surface of the stacked optical waveguide to be measured, so that the vision module automatically locates the initial position Z1.

8. The detection method according to claim 1, wherein In step S500 , the stepping interval ΔZ is obtained based on the product of the depth of field L and the refractive index n.

9. The detection method according to claim 8, characterized in that The number of steps m is proportional to the thickness H of the stacked optical waveguide and inversely proportional to the step interval ΔZ.

10. The detection method according to claim 1, characterized in that After step S700, the following steps are further included: S950: Based on the collected internal defect images of the 1st to mth layers, a defect information table is generated; the defect information table includes: defect serial number, defect location ID, defect depth, defect coordinate Z, defect coordinate X, defect coordinate Y, defect diameter, defect length, defect width and defect type.

Images