TW202117276A - System and method for measuring a surface in contoured glass sheets - Google Patents

Abstract

An optical inspection system is provided for an ultraviolet laser and associated optics forming a planar laser sheet directed to a glass sheet. The planar laser sheet intersects a surface of the glass sheet thereby causing the surface of the glass sheet to fluoresce and form a visible wavelength line on the surface. A camera has an image sensor for detecting the visible wavelength line. A control system is configured to receive image data indicative of the visible wavelength line, analyze and triangulate the data to determine a series of coordinates associated with the line, and create a three-dimensional map of the surface of the glass sheet as a function of the series of coordinates. Methods for using an optical inspection system, for gauging a surface using an optical inspection system, and for providing optical reflectance information for a surface using an optical inspection system are also provided.

Description

System and method for measuring the surface of curved glass sheet

Various specific embodiments relate to systems and methods for measuring the surface of curved glass sheets.

Manufacturers of glass sheets are interested in the measurement and evaluation of the surface of the glass sheets, especially glass sheets formed into various curved shapes for use as automobile windshields, backlites and sidelites. The manufacturer may want to determine whether the glass sheet is within the predetermined specifications of the gauge. The manufacturer may also want to measure and evaluate the amount of reflected optical distortion of the formed sheet as perceived by human observers, such as outside observers or operators or passengers in a vehicle, where the glass may be installed as a windshield, rear windshield, or side shield Glass, or the like. For example, the regulatory indicators and reflection distortion thresholds become more and more stringent as vehicle applications increase the use of technologies such as head-up displays. Manufacturers also want to identify small traces or other defects visible on the surface of the formed glass sheet.

In a specific embodiment, an optical detection system is provided with an ultraviolet laser and related optical elements, which form a planer laser sheet directed to a glass sheet. The plane laser cut sheet intersects the surface of the glass sheet to cause the surface of the glass sheet to fluoresce and form visible light wavelength lines on the surface. The camera has an image sensor for detecting wavelength lines of visible light that span at least a part of the width of the film. The control system is configured to (i) receive image data representing the visible light wavelength line from the camera, and (ii) analyze the data from the camera to determine the first and second coordinates associated with the line in a series of symbols , (Iii) triangulate a third coordinate each associated with the first and second coordinates in the coordinate series, and (iv) establish a function of the surface of the glass sheet as a function of the coordinate series Three-dimensional map.

In another specific embodiment, a method of using an optical detection system is provided. A flat laser cutting page is formed from an ultraviolet laser and related optical elements and is directed to a surface of a glass sheet. The surface of the glass sheet is excited at an intersection of the planar laser cutting sheet and the surface to form a visible light wavelength line on the surface of the glass sheet. The visible wavelength line is imaged using a camera. The first and second coordinates related to the visible light wavelength line in a series of landmarks are determined by analyzing the imaging data from the camera. In the coordinate series associated with the visible light wavelength line, a third coordinate associated with each of the first and second coordinates is determined by triangulation. Create a three-dimensional map of the surface of the glass sheet as a function of the coordinate series.

In another specific embodiment, a method of using an optical inspection system to gauge a surface is provided. Using a camera to image a position of a fluorescent line on the surface by triangulation, a data set containing a set of coordinates corresponding to a position on a surface of a glass sheet is calculated, wherein the The fluorescent light is created at an intersection of a flat laser cut sheet from an ultraviolet laser and the surface. Create a three-dimensional map of the surface as a function of a series of data sets. Use the data set to compare a gauge model of the surface to calculate an invariant index. The invariant indicator is output.

In another embodiment, a method of using an optical inspection system to provide information on the optical reflectance of a surface is provided. Using a camera to image a position of a fluorescent ray on the surface by triangulation, a data set containing a set of coordinates corresponding to a position on a surface of a glass sheet is calculated, wherein the fluorescent ray is established in The intersection of the plane laser cut sheet from an ultraviolet laser and the surface. Create a three-dimensional map of the surface as a function of a series of data sets. Denoising the three-dimensional image of the surface. Use the data set from the denoising map to compare an optical reflectance specification designated for the surface of the glass sheet G to calculate an invariant index. The invariant indicator is output.

As required, several detailed specific embodiments of the present disclosure are provided here; however, it should be understood that the specific embodiments disclosed are only embodiments and may be embodied in different and alternative forms. The drawings are not necessarily drawn to scale; some features may be exaggerated or minimized to illustrate the details of specific components. Therefore, the specific structure and function details disclosed here should not be construed as limitations, but merely serve as a representative basis for teaching those familiar with the art to use the present disclosure in various ways.

Facts have proved that any circuits or other electronic devices disclosed here can include any number of microprocessors, integrated circuits, memory devices (for example, FLASH, random access memory (RAM), read-only memory (ROM)) , Electronic Programmable Read-Only Memory (EPROM), Electronically Erasable Programmable Read-Only Memory (EEPROM), or other suitable variants of them) and work together to perform the operations disclosed here (s) software. In addition, any one or more of the electronic devices disclosed herein can be configured to execute a computer program embodied in a non-transitory computer-readable medium and programmed to execute any number of functions disclosed herein.

FIG. 1 illustrates an in-line glass sheet optical inspection system 10. The detection system 10 includes a conveyor 12 that conveys the glass sheet G in a first direction substantially parallel to the first dimension of the glass sheet. In the illustrated embodiment, the curved glass sheet G is a generally rectangular vehicle windshield or rear windshield glass, which has a relatively small first size (and alternatively may be referred to as a height) and a second relatively large size. Size (can be called width alternatively). The glass sheet G has a thickness in the third dimension, which is smaller than the width and height. The glass sheet G is curved with respect to one or more curvature axes that are substantially parallel to the first direction. In other embodiments, the glass sheet G may have other axes of curvature, or be provided as a flat or substantially flat sheet.

The conveyor 12 may be a single conveyor, which individually and exclusively transports the glass sheet G through the inspection system 10 that can be assembled and/or operated as an independent optical inspection system. In other embodiments, the conveyor 12 may be one of a series of conveyors that convey glass sheets through various processing workstations, such as heating and shaping that appear in typical automobiles, buildings, and/or solar glass sheet manufacturing systems. And annealing or tempering workstation. In order to process the glass in this manner, the conveyor of the glass sheet G can be provided by various technologies such as rollers, air cushions, or belt conveyors, positioners, and robotic arms. It should also be understood that each of the plurality of conveyors can be independently controlled to move the glass sheet through different processing workstations at high speed to effectively control the flow and processing of the glass sheet in the system 10.

Alternatively, the detection system 10 may be provided as a separate independent system or device without a conveyor. The inspection system 10 may also be provided with a fixture for the glass panel G, wherein the inspection system 14 is configured to be translationally movable relative to the panel G, for example, an optical system 14 mounted on a conveyor system. The detection system 10 may be provided with a glass panel G and an optical system 14 fixed to each other, wherein the optical system has several optical elements configured to scan the surface of the glass panel G.

The inspection system 10 has an optical system 14 used to identify and measure the surface of the glass sheet, and can be further used to gauge the sheet, identify and measure small defects in the sheet, and/or measure reflection optical distortion. The detection system 14 is described in detail with reference to FIG. 2.

Generally, the optical system 14 includes a laser or other light source with a wavelength selected at least partly based on the optical properties of the glass sheet G. The light from the light source is guided to the glass sheet G by the optical system 14. _{In one embodiment, the light source is selected to have a wavelength λ 1 that} is opaque or substantially non-transmissive to the glass sheet G in the tuning narrow frequency band. The light source is also selected such that the wavelength λ _{1 of} the light source can induce or cause the surface of the glass sheet G to emit light at a wavelength λ ₂ different from the light source. For example, the light source is also selected such that the wavelength λ _{1 of} the light source can induce or cause the surface of the glass sheet G to fluoresce or emit light at a wavelength λ _{2 that is longer than the wavelength λ 1 of the light source.}

The optical system 14 has at least one camera or other detector for detecting light emitted from the glass sheet G. The optical system 14 also includes various optical elements to control and guide the light from the light source to the glass sheet G and from the glass sheet G to the detector.

The optical system 14 has at least one computer and/or control unit, which includes at least one processor, which is programmed into executable logic to control the optical system including the light source and the detector, and obtain data from each glass sheet The data of the detector analyzes the data of the glass sheet, and provides information related to the surface shape, reflection optical distortion, or other surface information or defects of the glass sheet. The computer can be integrated with the control system 16 of the detection system 10, as shown in the figure, or can be provided as a separate device for communicating with the control system 16.

The optical system 14 therefore provides a non-contact detection system for quickly acquiring detailed data corresponding to the surface of the glass sheet G and analyzing the acquired surface data to evaluate and report the surface shape of the glass sheet G and the properties related to the optical characteristics of the glass sheet G, In particular, the glass sheet G is transported on the conveyor 12 between or after bending, cooling, or other processing operations.

The detection system 10 includes a programmable control unit 16, which is depicted as a computer in this embodiment. The computer 16 can communicate with or be integrated with the computer of the optical system 14. The computer 16 includes at least one processor, which is programmed to detect the glass sheet as it advances on the conveyor, and controls the motor(s) to control the movement and speed of the conveyor 12.

The conveyor 12 moves the glass sheet G along a path or a direction, which is shown as the y direction here, through the optical system 14. The movement of the conveyor 12 uses one or more motors and supporting rollers or other devices.

The detection system 10 has one or more position sensors 18 to determine the position and timing of the conveyor 12 to analyze the glass sheet G using the optical system 14 when the glass sheet G moves through the system 10. The position sensor(s) 18 may be provided by a digital encoder, an optical encoder or the like. The speed of the conveyor 12 and the speed of the glass sheet G can be selected to allow the light from the light source of the optical system 14 to have sufficient residence time on an area of the surface of the glass sheet G to cause the surface to fluoresce while maintaining the glass sheet G Straight line operation. In one embodiment, the conveyor 12 moves continuously at a speed of 0.1 to 0.2 m/s, or moves at a rate coordinated with the optical system 14 to obtain data based on the specified movement of the glass panel G, for example, in the range of 1 to 4 mm Move within. In a further embodiment, the movement of the conveyor 12 causes the glass panel G to be detected within a time frame of about 10 to 15 seconds. In another embodiment, the conveyor is continuously moved at a speed of 0.01 m/sec or more to acquire data based on the movement of the glass panel G of about 5 mm or less. The information obtained from the glass sheet can be moved based on the change of the glass panel G corresponding to the grid size, and can be less than 5 mm, less than 2 mm, about 1 mm, or another value. In yet another embodiment, the speed of the conveyor can be increased to obtain data based on the movement of the glass panel G by about 5 mm or more, for example, a grid size of 5 mm or more, or 10 mm or more. When the speed of the conveyor decreases, the time to scan the panel G increases, and it may be about several seconds, tens of seconds, or several minutes, for example, two minutes. For a glass panel G with a complex surface profile, the time to scan the panel may also increase, because the use of a smaller grid size may require an increase in resolution. The position sensor 18 can be used as an input to the optical system 14 to determine the timing of data acquisition, for example, as a camera trigger.

The detection system 10 may be provided with an additional sensor, such as a photoelectric sensor or the like, which communicates with the control system 16 to determine the proper position of the glass sheet G on the conveyor 12 or has advanced to the optical system 14. The computer 16 then communicates with the optical system 14 to activate the system 14 and start measuring the surface of the sheet G. In other embodiments, the optical system 14 can continue to operate and begin to acquire and process data related to the glass sheet G in response to the system 14 detector recording an appropriate signal indicating that the light source has begun to interact with the glass sheet G.

Please refer to FIGS. 2 to 4 for a detailed description of the optical system 14 according to various specific embodiments. In the disclosed embodiment, the light source is provided by the laser 50, and may be a diode laser. The laser 50 can be packaged with a diode laser to have a suitable base and lens to provide a collimated beam, and a heat sink can be additionally provided. In other specific embodiments, the laser 50 may be provided by another laser or a combination of lasers, which is configured to provide laser light of a desired wavelength at a desired laser intensity and other beam parameters. In a further specific embodiment, the light source can be provided as an ultraviolet light source, which has related filters and optical elements to provide a light sheet guided to the glass sheet G. In a further specific embodiment, the system 14 may be provided with more than one laser 50.

Based on the intended use of the glass sheet G formed of soda lime silicate glass, the laser 50 is selected to have a wavelength in the ultraviolet range, and is either opaque or non-transmissive to the glass sheet G or substantially non-transmitting to the glass sheet G The specific wavelength has, for example, a laser output transmittance of less than 5% or 2%. In the disclosed embodiment, the laser pulses 50 provided by a solid state laser diode, which is to be output by the harmonization with a central wavelength λ ₁ of light. The center wavelength can be selected to correspond to the opacity or opacity of the glass sheet G, and it can be a non-visible light wavelength, and it can also be selected to induce fluorescence or light emission on the surface of the glass sheet G. The center wavelength may be less than 350 nanometers, and in one embodiment, 266 nanometers in the ultraviolet range are provided.

In one embodiment, the laser has a power output of 150 microjoules at 15 kHz. In a further embodiment, a laser 50 with other power output and repetition rate may be provided, and a laser with a center wavelength of 500-100 microjoules and one kilohertz or more at a center wavelength of 266 nm may be considered. The systems 10 and 14 can be configured to measure the glass sheet G within a specified period of time, for example, about 10 seconds per sheet. However, based on the size of the sheet and the desired resolution of the model, other times can also be considered. Of course, based on the composition of the glass sheet G, other wavelengths can be selected.

Figure 3 is a graph showing the transmittance of soda lime silicate glass versus the wavelength of incident light. It can be seen from this figure that the transmittance of the glass sheet G decreases sharply as the wavelength of the incident light decreases and is similar to an opaque material. The wavelength of the laser 50 is also plotted on the graph, and it is shown that the glass sheet G is opaque at this wavelength. For a glass sheet G with another material composition, a laser 50 that is tuned to another wavelength can be selected for use in the optical detection system 14 so that the selected wavelength does not pass through the glass sheet G and induces luminescence or luminescence in the glass sheet G Fluorescent.

Please refer to FIG. 2 again. Several optical elements 54 are arranged downstream of the laser 50 to interact with the laser beam to shape and guide the laser beam toward the glass sheet G. The optical element 54 may include one or more beam shapers, lenses, mirrors, and the like. In the disclosed embodiment, by converting the collimated Gaussian beam profile of the laser beam into a collimated flat profile beam with a more uniform intensity distribution and no internal laser focus, a beam shaper 55 is provided to increase the laser intensity Therefore, the uniformity of the intensity of the lightning rays on the surface of the glass sheet G is increased. The first lens 56 is provided to form a planar laser sheet from the laser beam 52. In the disclosed embodiment, the first lens 56 is a cylindrical plano-concave lens whose focal length is selected based on the positioning of the laser 50 relative to the conveyor and the glass sheet, or the distance D1 from the laser 50 to the glass sheet G. In order to further focus the plane laser cutting page, a second lens 58 may also be provided. In the disclosed embodiment, the second lens 58 may be a cylindrical plano-concave lens to further focus the plane laser cutting page. An additional focusing lens 57 can be provided to narrow and focus the laser cut page. Although the second lens 58 is illustrated after or downstream of the first lens 56, in other specific embodiments, the positioning of the lenses 56, 57, 58 may be changed or reversed so that the second lens 58 is in front of the first lens 56. Based on the opacity of the material forming the glass sheet G, the lenses 56, 57, and 58 are used to form a focal plane laser cut sheet 60 directed to the first surface 62 of the glass sheet G _{with a harmonic center wavelength λ 1 of 266 nm.} In the illustrated embodiment, the optical elements 56, 57, and 58 are each provided by f=-8mm plano-concave cylindrical lens, f=1000mm plano-convex cylindrical focusing lens, and f=-25mm plano-concave cylindrical lens. . The optical elements 54 cooperate to form a flat laser cut sheet with a beam width of 1 to 2 mm on the surface of the glass sheet G. In another embodiment, the optical elements may have different focal lengths, and the focal length and lens selection may be based in part on the size of the panel G. In other specific embodiments, additional optical elements such as filters, choppers and the like can be installed between the laser 50 and the glass sheet G.

Alternatively or in addition, other optical elements such as Powell lenses can be used to provide a more uniform distribution of laser intensity on a flat laser cutting sheet. In addition, although one laser 50 is shown in FIGS. 2 to 4, in other specific embodiments, the system 14 may be provided with more than one laser 50. For example, the system 14 may have two lasers 50 directed to different areas of the glass sheet G, or the light beams may be aligned to form a common sheet on the glass sheet G. The intensity of the laser cut sheet is different on the laser cut sheet from a laser, for example, according to the distribution, and multiple lasers can be used to provide a more uniform intensity on the surface of the glass sheet G, or to provide individual applications. Fluorescent light from multiple cameras.

The glass sheet G has first and second surfaces 62, 64 that form the first and second surfaces of the glass sheet G. The first and second surfaces 62 and 64 are separated from each other by the thickness of the glass sheet G. _{Since the wavelength λ 1 of the} laser cut sheet 60 is opaque or substantially non-transmissive to the glass sheet G, the laser cut sheet 60 interacts with the first surface 62 without traveling through the glass sheet G or to the second surface 64, and therefore only The glass sheet G is excited on the first surface 62. The thickness of the glass sheet G, that is, the distance between the first and second surfaces 62 and 64, can be changed, and in some embodiments, it is greater than 1.2 mm or greater than 1.6 mm, and in a non-limiting embodiment, it falls on In the range of 1.6 to 25 mm.

The flat laser cut sheet 60 can be oriented so that it extends laterally or in the x-direction to the belt. The laser cut sheet 60 interacts with the first surface 62 of the glass sheet G along a path 66 (or the intersection of the laser cut sheet 60 and the first surface 62 of the glass sheet G). For the flat glass sheet G, the path 66 is a straight line. With regard to the curved glass sheet G, when the laser cut sheet 60 interacts with the curved surface, the path 66 may change with the curvature of the glass sheet G along the surface 62.

The laser cut sheet 60 excites the material of the glass sheet G on the first surface 62 and causes the first surface 62 to emit light. The glass sheet G fluoresces along the excited thunder ray 66 to emit a wavelength λ _{2 that is longer than λ 1} . In this embodiment, the radiated light 70 from the glass sheet G has a wavelength λ ₂ in the visible light range or in the near ultraviolet range, and appears as a line 68 along the surface 62 and a line 66 of excitation.

The emitted light 70 is detected by a detector such as a camera 72. The camera can be equipped with a charge coupled device (CCD) image sensor or a complementary metal oxide semiconductor (CMOS) image sensor. In this embodiment and as shown in FIG. 2, the detector 72 is provided with a CMOS sensor and is positioned so that the full width of the glass sheet G, or the glass sheet in the x direction, can be used to capture images. In the illustrated embodiment, the camera is provided by a CMOS sensor camera with 5496x3672 pixels. In one embodiment, it is set in a region of interest with 5496x1836 or 5496x1000 pixels, or in another embodiment, there is CMOS sensor with 5120x5120 pixels. The control of various camera settings can be based on laser parameters, conveyor speed, and other system factors, and these camera settings include lens focal length, aperture, gain, and exposure time. In one embodiment, the camera uses a fixed lens, such as a 16mm or 25mm lens, and is set to an aperture of f2.4 or higher to increase the depth of field, and uses an exposure time of 15-20 milliseconds and 2-15 Gain in decibels. In another embodiment, the camera settings can use another exposure time, for example, between 10 milliseconds and 300 milliseconds or more, and the gain can also be set to another value between 2 and 30 decibels. In other embodiments, the camera 72 may be positioned to only take pictures of selected areas of the glass sheet G. In a further embodiment, the detector can be installed as another light detector or light sensing element. In a further embodiment, additional optical elements such as filters may be provided between the glass sheet G and the detector 72 to further improve the signal-to-noise ratio.

In other embodiments and as shown in FIGS. 4 to 5, the optical system 14 may use more than one camera 72. 4 to 5 are schematic diagrams and do not show various system 14 components, such as optical elements, controllers, and so on. In FIG. 4, a pair of cameras 72A, 72B are located on one side of the laser cut sheet 60. Each camera 72A, 72B photographs the regions 73A, 73B of the glass sheet G, and these regions overlap each other by about 10 cm, for example. Using more than one camera in the system of Figure 3, based on the longer focal length of each camera, the final surface map of the surface of the glass sheet G can reduce the error introduced by optical distortion. In addition, the use of the optical system 14 with two or more cameras can also provide higher-resolution imaging for the glass surface, such as increasing the number of pixels per unit surface area of the glass sheet G, and increasing when the camera entity is closer to the glass sheet G. Accuracy and sensitivity.

In Fig. 5, a pair of cameras 72A and 72B are installed, and the laser cutting page 60 is located between the cameras 72A and 72B. Each camera 72A, 72B photographs the regions 73A, 73B of the glass sheet G, and these regions overlap each other by about 10 cm, for example. Using more than one camera in the system of Figure 3, the final surface map of the surface of the glass sheet G can reduce errors introduced by optical distortion. In addition, the cameras 72A and 72B are positioned, for example, when a camera 72A has a blocked line of sight or a small incident angle or a glancing angle to the left of the surface of the glass sheet G in FIG. 4, the glass sheet G with a high degree of curvature can be photographed. In another specific embodiment, the system 14 may have more than two cameras. For example, it is conceivable that four cameras 72 with two on each side of the laser cutting sheet 60 can be installed, and the areas of all four cameras overlap each other. In other embodiments, there may be more than 4 cameras 72.

Please refer to FIG. 2 again, the laser 50 and the camera 72 communicate with at least one computer 80 in the control unit. The computer 80 can be connected to or integrated with the control system 16 or the detection system 10. The computer 80 has an image processor unit 82 and is also connected to a memory. The computer 80 can receive information from the control system 16 and the position sensor to determine when to send a trigger signal to the camera (s) 72 of FIGS. 2 and 4 to 5. For example, based on the 1-4 mm corresponding movement of the glass sheet G, the camera(s) can obtain data at a frame rate of 80 fps, or 160 fps.

The computer 80 receives image data from the camera 72. The computer 80 uses the data from the image to form a matrix or point cloud. For example, cells in the matrix are associated with positions on the surface 62 of the glass sheet, or an array of points in the point cloud is associated with positions on the surface 62 of the glass sheet. The image data can be provided as a series of grayscale values, where each value is between 0 and 255, if it is based on an 8-bit grayscale. In other embodiments, the image data can be provided as a series of RGB signals each corresponding to the red, green, and blue blocks or color components detected by the camera sensor, and can be based on different color spaces and color models. Provided, or may be provided as another bit value. The following provides further details of data acquisition and pre-cloud and post-cloud data processing with reference to FIGS. 6-13.

The flowchart of FIG. 6 is used to determine the three-dimensional image of the surface of the glass sheet G according to a specific embodiment using a method 200 of an optical detection system. In various specific embodiments, the method 200 can be used with the systems 10 and 14 of FIGS. 1 to 2 and FIGS. 4 to 5, respectively, and according to various specific embodiments, several steps can be rearranged or omitted, or can be added Additional steps.

In step 202, based on the transmittance of the glass sheet, a first wavelength, such as an ultraviolet wavelength, is selected, wherein the first wavelength that is not substantially transmitted through the glass sheet is selected. In step 204, for example, an ultraviolet laser is used to form a plane light-cut sheet at the first wavelength.

In step 206, the optical system 14 is calibrated. Calibrate the camera(s) and the laser(s) separately. A checkerboard chart or other conventional flat calibration surface is provided instead of the glass sheet G, and is located approximately at the middle z-distance of the glass sheet G, or in the middle of the bounding box that defines the camera(s). The bounding box for these cameras is defined as a three-dimensional space that takes into account sufficient vertical depth or depth of field to include the upper and lower boundaries of the glass sheet G, while ignoring any underlying fixture supporting surfaces. In one embodiment, the bounding box has a depth of about 10-15 cm, however, based on the details of the part, other dimensions are conceivable.

The position of the calibration chart is therefore known and provides the basis for the global coordinate system of the bounding box. The camera is calibrated based on a stationary calibration chart, and camera settings including focal length, exposure, aperture, gain, and the like are selected.

The laser is also calibrated. For a laser oriented such that the laser cut sheet extends along the z axis or in the xz plane, the camera(s) detects the position of the laser cut sheet relative to the calibration chart and associates this position with the global coordinate system . For lasers that are not in the x-z plane, additional measurement steps are required to provide the laser cutting page position to the global coordinate system. Choose laser settings including intensity and frequency. Later, referring to the determination of the z-coordinate of the surface in step 214, the camera and laser calibration relative to the calibration map and the global coordinate system will be used.

In step 208, for example, a related light is used to guide the plane light-cut sheet to the glass sheet, and the surface of the glass sheet is excited so that it is along the intersection of the plane light-cut sheet and the surface of the glass sheet facing the light source at a second wavelength, for example, a visible wavelength The path radiates light.

In step 210, one or more cameras are used to measure or detect a series of data representing a series of visible wavelength lines, where each camera takes a picture of the region of interest of the glass sheet G. For a single camera system 14, the camera 72 can detect lines across the panel or in the region of interest on the panel. For the multi-camera system 14, each camera 72 can point to the overlapping individual regions of interest of the glass sheet G and detect each line in the region of interest.

In step 212, each visible light wavelength line is processed and analyzed to determine the first and second coordinates in a coordinate series associated with each line, such as (x, y) coordinates, and the coordinates are stored in a matrix associated with the camera Or point cloud.

The computer 80 uses the image processor unit 82 to process images. Figure 7 provides a representative example image and related schematic overlap regions based on the following image processing.

The computer 80 and the image processor 82 can process the image to reduce noise, for example, by applying a threshold to the image, normalizing the image, using fast Fourier transform to transform the image, and the like. In one embodiment, a median filter for noise removal is used to filter the image, including speckle noise. The need for noise removal steps may increase as the camera gain setting increases.

The computer 80 and the image processor 82 select the line width region in the image based on the input of the indicated line width size and the minimum intensity. By selecting the line width area, the reduced area of the image is used in the processing and determination of coordinates in the subsequent steps. For example, the indicated line width can be selected as the nominal value with tolerance factor. In one embodiment, a nominal line width between 3-10 pixels is selected, and a tolerance between 1-5 pixels is selected. In another embodiment, a nominal line width between 4-6 pixels is selected, and a tolerance between 2-3 pixels is selected. In yet another embodiment, the indicator line width with the minimum width of one pixel is selected. The intensity is also selected as input and can be based in part on the expected intensity of the fluorescent light in the relevant frame rate, and exclude pixels that contain background noise that is unnecessarily included. In various embodiments, the minimum intensity is selected on a gray scale of 5/255, 10/255, 20/255, 30/255 or another value. The area 250 in FIG. 7 is an example of the selected line width area.

When the glass sheet G _{is transparent to the visible light emitted at the second wavelength λ 2} and the light emitted from the surface 62 can pass through and/or reflect from the other surface 64, experiments show that any reflection of the back surface 64 has a very low signal And roughly it can be discarded as noise. For example, the gray scale of the reflection of the radiation 68 on the back surface 64 in the acquired image is about 5/255, while the gray scale of the primary emission line 68 is about 20 to 100/255, so that the reflected signal can be regarded as noise. News. In addition, based on the position of the reflected line in the image and the distance from the primary visible line, the reflected line can be excluded or outside the range of the selected line width area 250, so that it is no longer considered during data processing. In the case where the reflected line is sufficiently close to the radiation 68 so that the use of the line width area 250 cannot be satisfactorily excluded, the reflected line can be included in the data processing; however, the data processing is not greatly affected due to the low signal of the reflected line. Alternatively and in the presence of the second reflection line, the data processing can select a fluorescing line to take into account the camera geometry, the lines are based on the position of the panel, the brightness of the fluorescent light higher than the reflection line, and other factors. The data processing can further receive input on the material of the panel G. For example, it can be expected that the tin-side float glass image has reflection lines, while the air-side float glass image may not have reflection lines.

The computer 80 and the image processor 82 determine or calculate the points on the model line based on the image data and the gray level of the pixels in the visible fluorescent light. It can be seen from the exemplary image that the light emitted from the surface 62 extending across a plurality of pixels in the y direction in the region 250 can be obtained. Similarly, the computer 80 calculates the points used for the model line as a mark or (x, y) data set series based on the grayscale values of the pixels. For example, the points calculated by the model line 100 are overlaid on FIG. 4.

In one embodiment, the computer and the image processor can average the groups of adjacent pixels to find each point and the (x, y) data set as the estimated center of the model line, and it can be each of the image in the x direction The pixel provides an (x, y) data set. In other embodiments, the computer and the image processor can perform a weighted average of pixels to find each point and the (x, y) data set as the estimated center of the line. Computers and image processors use the distribution functions of the pixel values to calculate the points and (x, y) data sets, such as Gaussian distribution, average or median width, or other mathematical functions. Computers and image processors can alternatively use known commercially available image library processing tools or software to use pixel data to determine the values used in these data sets.

Then, the computer 80 inputs the (x, y) data set series in the relevant unit of the matrix or in the point cloud. The data set may include a series of marks from the model line as determined from the image. Each data set may include (x, y) values of the position of the surface 62 of the glass sheet G in the x-y plane, and they collectively define the model line 100 and are connected to the global coordinate system.

The matrix can also be filled into a line scan image, in which each row in the matrix is filled with coordinates from the model line of each glass sheet G image, and subsequent rows are filled with subsequent images. Similarly, a cloud can be formed from the sequential images of the glass sheet G. In one embodiment, the matrix may have a unit and a coordinate set associated with each pixel in the x direction from the image. In other embodiments, the matrix may have a unit and a set of coordinates associated with multiple adjacent pixels, causing the computer to perform an averaging step or the like before entering the data. In some embodiments, the matrix or point cloud can be thinned out, for example, by deleting the collection data set numbered as a multiple of n.

In step 214, the position of the laser, the position of the camera, and the first and second coordinates are used to triangulate the third coordinate, such as the (z) coordinate, and store it in a matrix or point cloud. Then, the computer 80 calculates the z value of each (x, y) coordinate set, which is related to the z-position of the coordinate set on the surface 62 of the glass sheet G. The calculation of the z value uses the (x, y) values from the coordinate set of each unit and the positioning of the laser and the camera from the detection system 14 in the triangulation calculation described below when describing FIGS. 8A to 8B. The computer inputs the (z) value into the relevant cell in the matrix with corresponding (x, y) coordinates to complete the map of the surface 62. When processing the image in step 212, the computer and the imaging program can calculate the (z) value of each coordinate set, or after the glass sheet has been fully imaged, the (z) value of all data sets in the matrix or point cloud can be calculated.

8A to 8B are views illustrating that the coordinate series of each line calculated from the image shown in FIG. 4 is related to the third associated with the first and second (x, y) coordinates in the data set D (z) Triangulation of coordinates. The laser 50 and the camera 72 can be fixed to each other, and the laser beam and the generated planar laser cutting sheet are also fixed, so that the glass sheet G translates or moves in the direction y relative to the optical system 14 and passes through the planar laser cutting sheet. Since the position of the laser cutting page and the camera have been calibrated and connected to the global coordinate system, the z coordinate can be calculated.

The side view of FIG. 8A shows the optical system 14 of the calibration diagram C of the overlying glass sheet G, and also shows the bounding frame B. As shown in FIG. The reference of the laser cut page 60 is relative to the calibration chart C and the global coordinate system. Similarly, each pixel in the camera refers to the calibration chart C and the global coordinate system, so that each pixel has a correlation vector in the (x, y, z) space. The (x, y) coordinates for the data set D can be determined using the above step 212 and bound to the global coordinate system based on the schematic diagram shown in FIG. 8B. Then, referring to the schematic diagram of Figure 8A and using triangulation technology, use the laser to cut the intersection of the page 60 and the relevant pixel vector, compare the intersection of the pixel and the vector with the calibration map C, and calculate the data set D in the global coordinate system. The z coordinate. Therefore, the z value of each (x, y) coordinate along the line in the coordinate series can be calculated, and the control system 80 then inputs this z coordinate into the matrix unit or point cloud, where ( x, y) value.

In step 216, subsequently, for each of the series of visible light wavelength lines of the glass sheet G in the camera, a three-dimensional map of the glass sheet surface as a function of the coordinate series is established from the matrix or point cloud.

Perform the steps instructed by area 218 for the data from each camera for each camera. For a single camera system, step 216 generates the final matrix or point cloud of the surface of the glass panel G.

For the multi-camera system 214, an additional step 220 is provided to combine the matrices or point clouds from different cameras into a single matrix or point cloud representing the glass sheet G. FIG. 9 schematically illustrates the subroutine used in step 220 to align multiple matrices or point clouds, for example, using the dual camera system shown in FIG. 4. As for the dual-camera system or a system with many cameras as shown in FIG. 5, the subroutines for aligning multiple matrices or point clouds are similar, and those skilled in the art are based on the method 220 described herein and the embodiment of FIG. 9 Will understand.

FIG. 9 schematically illustrates step 220 of the method 200. The computer 80 loads two matrices or point clouds M1 and M2 obtained by scanning the glass sheet G from the two cameras 72A and 72B, respectively. In other embodiments, using techniques similar to those described herein, more than two matrices or point clouds can be combined. It can be seen from FIG. 9A that when the depths of field of the cameras 72A and 72B overlap each other, the matrices M1 and M2 overlap. In addition, however, the system has been calibrated by at least one point in the calibration map provided in each camera area 73A, 73B, and it is preferable that these areas overlap to a certain degree. When the matrices M1 and M2 overlap each other, the overlap section There may be a small degree of deviation between the points. In order to form the final matrix representing the surface of the glass sheet G, the computer processes the matrices M1 and M2 to resolve the deviation. In one embodiment, the resolved deviation can be as small as about one thousandth of an inch. In addition, the deviation can be different in the overlap area.

In FIG. 9B, the computer movement matrix or one or both of the point clouds M1 and M2. In one embodiment, the computer 80 uses a rigid transformation to move one of the matrices so that it translates and/or rotates relative to the other matrix so that the two overlapping regions are aligned with each other as shown and reduce errors or misalignment.

In FIG. 9C, the computer 80 then builds the final matrix or point cloud M3 of the glass sheet G. In one embodiment, the computer removes points in the overlapping area from one or the other matrix. In another embodiment, and as shown, the computer uses the line L associated with the common calibration point of the matrix and uses the points from a matrix to fill the final matrix on one side of the line L as the sub-matrix M1*, and uses Points from another matrix fill the final matrix on the other side of line L as sub-matrix M2*.

When the laser light does not pass through the glass sheet G, only the surface 62 facing the detection system emits light. Similarly, a matrix with a coordinate series provides a three-dimensional high-resolution mathematical model of the surface 62 of the glass sheet G through the (x, y, z) coordinate series. In one embodiment, the matrix is provided with a mathematical model for the glass sheet G with more than one million coordinate sets. For example, the model may have a point density of 1,000,000 coordinate sets per square meter of the sheet surface 62, or about one coordinate set per square millimeter of the sheet surface 62.

In the embodiment mentioned in the description of FIGS. 1 to 2, the glass sheet G moves relative to the optical system 14 on the conveyor. In this scenario, the laser 50 and the camera 72 are relatively fixed to each other, and the glass sheet G passes underneath. In another embodiment, the glass sheet G can be fixed, and the laser beam emitted from the laser 50 can be scanned across the first surface 62 of the glass sheet G with a one-axis or two-axis mirror galvanometer or the like. In this scenario, in order to determine the angles associated with the distances D1, D2 associated with the positions along the lines 66, 68 of the glass sheet, the computer 80 also receives input representing the steering angle of the beam provided by the galvanometer.

In a further embodiment, the optical detection system 14 can be used to detect the second surface 64 of the glass sheet G and establish a corresponding three-dimensional matrix or point cloud representing the surface 64 of the glass sheet G.

The matrix or point cloud of the glass sheet G provides a high-resolution three-dimensional image of the surface of the glass sheet G. The coordinates in the matrix can be used to compare the mathematical model of the glass sheet G to determine whether the shape of the glass sheet is within the specifications of the shape, such as curvature. In addition, the coordinates in the matrix can be used to compare the mathematical model of the glass sheet G to determine whether the surface of the glass sheet G is within the specifications or standards of the optical reflectance of the surface.

The matrix and the generated model can be used with the system 10 to provide the measurement of the glass sheet G, as a pass/fail detection system to correct the deviation of the process or prevent the system 10 from deviating from the specifications, to establish the optical reflectivity of the glass sheet G model.

The computer(s) 16, 80 can be programmed to display information related to the three-dimensional map from the matrix in a graphical (for example, color-coded image) and/or statistical form. In various embodiments, statistical data of the glass sheet or predefined area of the glass sheet can be derived and reported, including z-distance, standard deviation, and other surface shaping or optical reflectance indicators.

The detection system 14 may also include a glass sheet part identifier that can be provided by the camera 72 or another component, to recognize the glass sheet as one of a set of known part shapes stored in the memory of the computer 80, wherein each known part There are related shape standards and optical reflectance standards for comparing maps in the matrix. The systems 10 and 14 can be programmed by the user to display various optical or shape marks of the glass sheet G detected by the device 14 in a graphical and/or numerical manner, for example, through a user interface and a display screen 20, including those most consistent with industry standards. Relevant marks, or other marks considered in the industry to be related to the quality of optical reflectance of the glass sheet that is analyzed, shaped, and made into glass. The system 10 can also be programmed to display the location of small defects identified by the device 14.

The selected data output by the disclosed in-line optical inspection system 10, 14 can also be provided as input for the control logic of the relevant glass sheet heating, bending and tempering system (or automobile windshield manufacturing system) to allow The controller(s) of one or more workstations of the glass sheet system modify the working parameters based on the optical data developed from the previously processed glass sheet.

The flowchart of FIG. 10 illustrates a method 300 of using the matrix or point cloud on the surface of the glass sheet as determined by the computer 80 using the optical system 14 to gauge a component such as the glass sheet G. In various embodiments, the steps of method 300 may be omitted or rearranged, or additional steps may be provided.

In step 302, the matrix or point cloud on the surface of the glass sheet G determined by the above-mentioned optical system and method 200 is input into the processor unit by the computer 80. The matrix may represent the entire first surface of the glass sheet G, or may only include a data set of selected surface areas of the glass sheet G.

In step 304, the computer 80 refers to a suitable measurement model of the glass sheet G. The use of computer-aided design (CAD) models and/or data can provide the measurement model, or other mathematical models, or the reproduction of the size or shape. The computer can determine the correct measurement model from one of several models of glass sheets G of various shapes and/or sizes stored in the memory. Then, the computer 80 inputs the selected measurement model to the processor.

In step 306, the computer 80 determines the glass sheet invariant index data to be compared with the measurement model. In one embodiment, the computer can determine the z-distance of the data set or the surface of the glass sheet G compared with the measurement model. The computer can calculate the normal vector distance from the surface of each data set or glass sheet G to the gauge model. Alternatively, the computer can calculate the vertical distance or z-distance from the surface of each data set or glass sheet G to the gauge model.

The computer can be configured to measure the entire glass sheet G with reference to the matrix or point cloud. In other embodiments, the computer may only measure a selected area (s) or part (s) of the glass sheet G or have additional measurement points in a specified area, such as the peripheral area, or it may be used for, for example, a head-up display or The area where cameras or other sensors are used for optical purposes. In one embodiment, the computer uses the selected area of the glass sheet G to simulate the contact gauge method.

In a further embodiment, and as shown by the optional block 308 shown in dashed lines, the computer can perform calculations or sub-programs on the neighborhood of several points to provide post-processed surface data points or data sets for use To determine the invariant index. In one embodiment, the computer performs interpolation, averaging, or another mathematical function, such as thresholding or the like, on adjacent points or data sets to calculate the post-processed data set of the neighborhood . This can provide improved measurement and invariant indicators by eliminating outlier datasets caused by surface dust, etc.

In other embodiments, the computer can determine the standard deviation of the z-distance, or another invariant index between the surface and the glass sheet G, such as calculating the radius of curvature, or other shape measurement indexes. These invariant indexes can be input into the measurement matrix or point cloud of the glass sheet G.

In step 310, the computer 80 outputs information related to the surface measurement, including any invariant indicators, such as a measurement matrix. The computer can further analyze the information in the measurement matrix to determine whether one or more indicators exceed the critical value. The computer can additionally provide information related to the measurement to the user, for example, through a color-coded map of the glass sheet or through numerical values or other forms of expression, such as a table. The computer can also provide information from the regulation to the process steps of the glass sheet G as a part of the control feedback loop.

Figures 11A and 11B illustrate a representative embodiment where the output from the method 300 is displayed to the user. FIG. 11A illustrates a simplified point map of the glass sheet G or the area of the glass sheet G, and provides a representative embodiment of the output of the system. The measurement point has been correlated with the listed invariant indicators, such as the differential value based on the normal, vertical, or other distance and the measurement value. If the differential value exceeds the specified threshold or is outside the allowable tolerance, a flag can be added to make it easy for users to understand. In the illustrated embodiment, the values correspond to the vertical distance from the normal line of the measurement model and are in millimeters, but other units are conceivable. In addition, the underlined value falls outside the threshold or margin and a flag is added to the user.

FIG. 11B illustrates a map of the area of the glass sheet G or the glass sheet G, in which different shadows correspond to different differential value ranges based on the normal, vertical, or other distance from the measured value, and provide another representative implementation of the system output example. If the differential value exceeds the specified threshold or is outside the allowable tolerance, a flag can be added to make it easy for users to understand. Based on the intended use and requirements of the film, the tolerances or thresholds of different panel areas can be set to different values. In the illustrated embodiment, the differentials are in millimeters, but other units are also conceivable.

The method 300 provides non-contact measurement for surfaces and glass sheets G, which allows for on-line monitoring and inspection of components, and allows for quick and convenient measurement of a series of different components, or the use of multiple measurement models. In addition, the non-contact measurement via the method 300 is provided to reduce the time and expense related to the measurement of parts, because the use of CAD data can easily create, change or update the measurement model, and does not require the use of special contact measurement of parts Precision measurement and measurement tool.

The flowchart of FIG. 12 illustrates a method 350, as determined by the computer 80 using the optical system 14, which uses a matrix or point cloud on the surface of the glass sheet to determine and model the optical reflectance of a component such as the glass sheet G. In various embodiments, the steps in the method 300 may be omitted or rearranged, or additional steps may be provided.

In step 352, the surface of the glass sheet G is input into the processor unit by the computer 80 as the matrix or point cloud determined by the optical system and method 200 described above. The matrix may represent the entire first surface of the glass sheet G, or may only include a data set of selected surface areas of the glass sheet G.

In step 354, the computer 80 performs a post-processing operation(s) on the matrix or point cloud. In one embodiment, the computer 80 modifies or removes the noise matrix or point cloud to remove certain points or artifacts from the point cloud. For example, the computer 80 can remove or modify the points in the point cloud that are adjacent to the edge of the panel G or within a specified distance of, for example, 1 to 2 mm, to remove points with edge effects or deviations in the measurement position, such as caused by beam steering effects. , And/or calculate the line center based on the width of thunder ray or visible light combined with the deviation that the panel edge is not perpendicular to the line or sheet. In addition, the fluorescent visible light may appear immediately outside the panel G and may be caused by, for example, the grinding of the panel or other parameters to cause the light to spread, thereby generating additional artifacts in the point cloud or matrix. In one embodiment, these inaccurate or false shadow points form a curve with the sign of curvature opposite to that of the adjacent panel G, or between the inaccurate or false shadow points and the adjacent panel G to form a knee of the curve, resulting in The computer 80 may use inflection points or inflection points as boundaries to cut off or delete these points from the point cloud or matrix. Although the description of this denoising step involves denoising the point cloud created when measuring the surface of the glass sheet G using visible fluorescence from the laser cutting page, the denoising step can also be applied to denoising edge effects Other sights and measurement systems.

The computer 80 also executes the noise removal algorithm on the point cloud or matrix in step 354. According to one embodiment, noise removal can be provided by averaging the normal vectors in the neighborhood of the data set, and then the average normal vector is used to update the data set to create a post-processed data set that matches the normal change. In a non-limiting embodiment, the point cloud composed of data sets has a triangular mesh with each data set as a vertex. In other embodiments, other mesh shapes can be used. Calculate the normal vector of each mesh triangle. The divergent normal vector can represent the noise in the point cloud, while the convergent normal vector can represent the smoother surface of the point cloud. The point cloud is smoothed by averaging or otherwise mathematically combining the normal vectors of the neighborhood or group of mesh triangles, and these neighborhoods can be defined as meshes that share common vertices or common edge(s) . Each average normal vector is used to adjust the coordinates of the relevant vertices to establish the post-processed vertices or data sets in the post-processed denoising point cloud or matrix. In other embodiments, other denoising algorithms may be provided and may be based on mathematical algorithms related to surface shaping techniques, such as the description of the following document: Author Gabriel Taubin published "A" at SIGGRAPH '95 Proceedings of the 22nd annual conference signal processing approach to fair surface design"; "Mesh Smoothing via Mean and Median Filtering Applied to Face Normals" published by Hirokazu Yagou and others in GMP '02 Proceedings of the Geometric Modeling and Processing — Theory and Applications (GMP'02); Shachar Fleishman et al. published "Bilateral mesh denoising" in SIGGRAPH '03 ACM SIGGRAPH 2003 Papers; and Thouis R. Jones et al. published "Non-iterative, feature-preserving mesh smoothing" in SIGGRAPH '03 ACM SIGGRAPH 2003 Papers.

According to another non-limiting embodiment, the computer 80 executes a noise removal algorithm on the point cloud or matrix, which uses the moving least squares method to smooth and interpolate the data. According to the present disclosure and for a given data point in the point cloud, a computer 80 is used to fit many adjacent points in the point cloud to a function in a least squares manner. In an embodiment, the function may be a polynomial. The determination of the post-processed data points is by moving the original given point in the point cloud to the polynomial surface of the determination. The computer 80 iterates the denoising of the point cloud to create a post-processed denoising point cloud or matrix. The embodiment of the moving least squares method and its surface application are in the book "Surfaces generated by moving least squares methods" by the authors Lancaster, Peter, and Kes Salkauskas in Mathematics of computation 37.155 (1981):141-158; and the authors Alexa, Marc, etc People can find it in the book "Computing and rendering point set surfaces" of IEEE Transactions on visualization and computer graphics 9.1 (2003): 3-15.

In step 356, the computer 80 determines or calculates one or more invariant indicators of the surface or one or more selected areas of the surface from the post-processed data set of the denoising matrix or point cloud. The invariant index may include horizontal curvature, vertical curvature, radius of curvature, principal curvature, Gaussian curvature, average curvature, derivative or rate of change of one or more of these curvatures, refractive power or optical power index, and the like.

In step 358, the computer can determine whether the invariant index representing the optical reflectance of the surface is within the optical reflectance specifications designated for the surface of the glass sheet G or the area of the surface. The computer can compare one or more of the invariant indexes with the corresponding design indexes of the glass sheet G. For example, the optical reflectance specification may be a standard or a thresholded invariant metric of the glass sheet G. In one embodiment, the specification includes, for example, a calculated curvature from a computer-aided engineering (CAE) model or an index for other model data of the glass sheet G. The computer can compare invariant indicators with critical values, or create a color-coded map or other visual output to indicate whether the glass sheet or area of the glass sheet is within a predetermined specification, or compare the surface with a predetermined specification. In a further embodiment, the computer may use more than one invariant mathematical function to provide optical reflectance scores or other readings to compare with specifications, and the specifications are weighted or otherwise added to different specifications when it is related to optical reflectance. Invariant and surface area.

In step 360, and in some specific embodiments, the computer 80 determines or calculates the variable index of the glass sheet G. In one embodiment, the computer 80 creates a simulated reflection grid, zebra board, or other image for the simulated visual performance of optical reflectivity and any distortion. The computer algorithm used to create the simulated reflection image can use a technique such as ray tracing, which uses the post-processed vertices in the denoising matrix and the normal vector calculated from each post-processed vertex. For example, referring to the usage of ray tracing, the rays from the vertices processed later in the denoising point cloud hit a virtual camera at an angle at one pixel to the normal vector, and since the incident angle is equal to the reflection angle, it comes from the same back on the point cloud. The rays of the processed vertices are therefore known and the intersection with the virtual grid can be determined. Using the pixels on the virtual camera and the points on the virtual grid plate can construct a reflection image such as a reflective grid plate or a zebra plate because they correspond one-to-one. Using the actual reflection image of the grid or zebra board that is the same as the algorithm used in the incident angle, camera settings, etc., the computer can be tuned or calibrated to simulate the algorithm of the reflection grid or zebra board. In yet another embodiment, the computer can combine, for example, a series of simulated grids or simulated zebra boards obtained at different angles of incidence to provide a video or flip-flop of the zebra board for visual display to the user. Change the position to simulate scanning the zebra board.

In step 362, the computer 80 outputs information related to the simulated optical reflectance and distortion of the surface, including any variable or invariant index, invariant data, or the visual performance or mapping of the simulated zebra board, or other calculations or simulations. In one embodiment, the computer 80 displays information on a display screen or other user interface. Figure 13 illustrates a representative embodiment of a simulated reflection optical image of a grid plate constructed using the methods 200, 350 as described herein. The computer can additionally provide this information to the process steps of the glass sheet G as part of the control feedback loop.

The method 350 provides non-contact detection and determination of the optical reflectivity and any distortion of the surface of the glass sheet G, which allows online monitoring and inspection of components, and allows a series of different components to be detected quickly and conveniently. In addition, non-contact optical reflection detection via method 350 is provided to reduce the time and cost associated with detecting components, and in addition, non-subjective indicators can be provided to determine whether the component has passed the specification.

In other embodiments, and depending on the manufacturing technology used to form the glass sheet, the first and second surfaces 62, 64 may emit light of different wavelengths in response to activation from the laser 50. In one embodiment, the glass sheet G is formed using a float glass process and has a higher tin concentration on one surface than on the other surface. In this scenario, a surface with a higher tin concentration fluoresces at a different intensity and/or wavelength from the other surface, and the detection system can be further used to identify the piece _{based on the different intensity and/or wavelength λ 2 of the emitted light} One side and the other. For example, in the case of a glass sheet with a higher tin concentration on the first surface than on the second surface, the first surface can emit fluorescence at a different intensity and/or wavelength than the second surface. In addition, the first surface may be higher than the second surface. High intensity and/or shorter wavelength fluoresce. In addition, the control unit and the computer 80 can modify the intensity of the laser 50 based on the surface of the glass sheet facing the laser, and/or adjust camera settings, such as gain or image processing settings. For example, for a glass sheet with a higher tin concentration on the first surface than on the second surface, compared to the second surface, the laser intensity on the first surface can be reduced, for example, to prevent oversaturation of the camera sensor. Alternatively or additionally, the gain of the second surface may be increased, or the second surface may require additional image processing steps for noise removal.

In addition, for example, in order to use a larger glass sheet G, reduce scanning time, or increase measurement accuracy, the system 10 may be provided with more than one optical detection system 14.

The system 14 with the light source of the non-transmissive glass sheet G can provide the measurement of the surface of the glass sheet G, and avoid the use of other systems that use visible light to query the combined scattering and reflectance of the glass sheet and from both the front and back surfaces 62 and 64 The problem. Similarly, by making the light emitted from the glass sheet G in the visible spectrum, the measurement can be simplified so that no ultraviolet sensor is required.

In a further specific embodiment, the optical system can be used to form a three-dimensional surface map of the object instead of the glass sheet G. In a limited embodiment, an optical system in which a laser emits light at another wavelength, such as a visible light wavelength, can be used to scan an object with a diffuse surface. The optical system uses one or more cameras(s) as described above to determine the three-dimensional surface map of the diffuse surface.

Although various embodiments of the present disclosure are described above, it is not intended that these specific embodiments describe all possible forms of the present invention. On the contrary, the vocabulary used in the patent specification is a description rather than a limitation vocabulary, and it should be understood that various changes can be made without departing from the spirit and scope of the present disclosure. In addition, the features of various implementation specific embodiments may be combined to form other specific embodiments.

10: (Online glass sheet optics) detection system/system 12: conveyor 14: detection system/optical system/single/dual/multi-camera system 16: programmable control unit 18: position sensor 20: user interface and display Screen 50: Laser 52: Laser beam 54: Optical element 55: Beam shaper 56: First lens 57: Additional focusing lens 58: Second lens 60: (planar) laser cut page 62: First surface 64: Second surface 66: path/lightning ray 68: original radiation (line) 70: radiation 72, 72A, 72B: camera 73A, 73B: area 80: computer/control system 82: image processor unit 100: model line 200: Method 202-220: Step 250: Area 300: Method 302-310: Step 350: Method 352-362: Step B: Bounding Box C: Calibration Chart D: Data Set D1, D2: Distance G: Glass Sheet/Glass Panel L: line M1, M2: two matrices or point clouds M1*, M2*: sub-matrix M3: final matrix or point cloud λ ₁ , λ ₂ : wavelength

Fig. 1 schematically illustrates a specific embodiment of a glass sheet detection system according to a specific embodiment;

Fig. 2 schematically illustrates an optical system used with the detection system of Fig. 1 according to a specific embodiment;

The graph of FIG. 3 illustrates the optical transmittance of the glass sheet used in the systems of FIGS. 1 and 2;

Fig. 4 schematically shows another optical system used with the detection system of Fig. 1;

Fig. 5 schematically shows another optical system used with the detection system of Fig. 1;

The flowchart of FIG. 6 illustrates a method of measuring the surface of a glass sheet using the systems of FIG. 1, FIG. 2, FIG. 4, and FIG. 5 according to a specific embodiment;

Fig. 7 shows a part of the visible spectrum image obtained on the surface of a glass sheet using the system of Fig. 2

8A and 8B are schematic diagrams showing the optical detection system of FIGS. 1 to 2 for use in triangulating the third coordinate;

Fig. 9 schematically illustrates a data alignment step sequence of a multi-camera system used with the method of Fig. 7 according to a specific embodiment;

The flowchart of FIG. 10 illustrates the method of using the surface map as determined from the method of FIG. 6 to measure the surface;

Figures 11A and 11B are representative system outputs displayed to the user when the method of Figure 10 is used to determine;

The flowchart of FIG. 12 illustrates the method of modeling and determining the optical reflectivity and distortion of the surface using the surface map as determined from the method of FIG. 6; and

Fig. 13 is a representative system output displayed to the user when the method of Fig. 12 is used to determine.

10: (Online glass sheet optics) detection system/system

12: Conveyor

14: Inspection system/optical system/single/dual/multi-camera system

16: Programmable control unit

18: position sensor

20: User interface and display screen

G: Glass sheet/glass panel

λ 1, λ 2: wavelength

Claims (30)

An optical detection system, which includes: An ultraviolet laser and related optical elements, which form a flat laser cut sheet directed to a glass sheet, wherein the flat laser cut sheet intersects a surface of the glass sheet, thereby causing the surface of the glass sheet to emit light Fluorescent and form a visible light wavelength line on the surface; A camera having an image sensor for detecting the visible light wavelength line spanning at least a part of a width of the glass sheet; and A control system, which is configured to (i) receive image data representing the visible light wavelength line from the camera, (ii) analyze the data from the camera to determine the first line in a standard series associated with the line One and second coordinates, (iii) triangulate a third coordinate associated with each of the first and second coordinates in the coordinate series, and (iv) establish the surface of the glass sheet as the coordinate A three-dimensional plot of the function of the series. Such as the optical inspection system of claim 1, wherein the laser, the flat laser cutting sheet and the camera are relatively fixed to each other. Such as the optical inspection system of claim 2, which further includes: a conveyor configured to make at least one of the glass sheet and the laser relatively translate. Such as the optical detection system of claim 3, wherein the control system is further configured to receive a series of data representing a series of visible light wavelength lines from the camera when measuring across the width of the glass sheet, each The line corresponds to a different position along the surface of the glass sheet; and Among them, the control system is further configured: analyze each of the series of visible light wavelength lines to determine the first and second coordinates in a coordinate series related to each line, and the triangulation is each with the coordinate series A third coordinate is associated with the first and second coordinates in each of them, and a three-dimensional map of the surface of the glass sheet is established from each of the coordinate series. Such as the optical inspection system of claim 1, wherein the control system is further configured to use a common calibration chart to calibrate the image sensor and the laser before receiving data. The optical detection system of claim 5, further comprising: another camera having another image sensor for detecting the visible light wavelength line across another part of the width of the glass sheet, Wherein, the camera and the other camera have a field of view overlapping on the surface of the glass sheet and on the calibration chart; and Wherein, the control system is further configured to: (i) receive data representing the visible light wavelength line from the other camera, (ii) analyze the data from the other camera to determine whether it is associated with the line The first and second coordinates in another coordinate series, (iii) triangulate a third coordinate each associated with the first and second coordinates in the other coordinate series, and (iv) create the glass sheet Another three-dimensional map of the surface as a function of the other coordinate series. Such as the optical detection system of claim 6, wherein the control system is configured to use the map and the other map to form a combined map. Such as the optical detection system of claim 7, wherein the control system is further configured to use a rigid transformation to move the figure relative to the other figure to align the figure and the other figure, and then from the figure and The coordinate series are removed from the overlapping field of view of one of the other images to form the combined image. Such as the optical detection system of claim 1, wherein a wavelength of the ultraviolet laser is selected to be non-transmissive through the glass sheet. Such as the optical detection system of claim 9, wherein the wavelength is selected to be between 260 and 270 nm. Such as the optical detection system of claim 1, wherein the control system is further configured to use a predetermined line width area to analyze the data to determine the first and second coordinates associated with the line in the coordinate series, wherein , The first and second coordinates in the coordinate series fall within the line width area. For example, the optical detection system of claim 11, wherein the line width area is a pixel function and/or a grayscale threshold. The optical inspection system of claim 1, wherein the surface of the glass sheet is a first surface; Wherein, the glass sheet has a second surface opposite to the first surface; and Wherein, the control system is further configured to adjust the intensity of the laser in response to that one of the first and second surfaces contains a higher tin concentration than the other of the first and second surfaces. Such as the optical inspection system of claim 13, wherein the control system is further configured to make the laser operate at a lower intensity when the first surface contains a higher tin concentration. Such as the optical inspection system of claim 1, wherein the image sensor of the camera is one of a CCD sensor and a CMOS sensor. Such as the optical detection system of claim 1, wherein the control system is further configured to use the three-dimensional image of the surface to determine a simulated optical reflectance of the surface. Such as the optical inspection system of claim 1, wherein the control system is further configured to use the three-dimensional map of the surface to measure the glass sheet. Such as the optical detection system of claim 1, wherein a single camera position is used to triangulate the third coordinate in the coordinate series. A method of using an optical detection system, which includes: Form a flat laser cut sheet from an ultraviolet laser and related optical parts and direct it to a surface of a glass sheet; Exciting the surface of the glass sheet at an intersection of the planar laser cutting sheet and the surface to form a visible light wavelength line on the surface of the glass sheet; Use a camera to take pictures of the visible light wavelength line; By analyzing the imaging data from the camera, determine the first and second coordinates associated with the visible light wavelength line in a symbol series; In the coordinate series associated with the visible light wavelength line, a third coordinate associated with each of the first and second coordinates is determined by triangulation; and Create a three-dimensional map of the surface of the glass sheet as a function of the coordinate series. Such as the method of claim 19, which further includes: Move the glass sheet relative to the laser, the flat laser cutting sheet and the camera; and When the glass sheet moves relative to the plane laser cutting page, use the camera to take pictures of a series of visible light wavelength lines; Among them, the three-dimensional map of the surface is established as a function of the coordinate series in the series of wavelength lines. A method of using an optical inspection system to gauge a surface, the method includes: By triangulating a position of a fluorescent light imaged on a surface of a glass sheet using a camera, a data set including a set of coordinates corresponding to the position on the surface of the glass sheet is calculated A position where the fluorescent light is established at an intersection of a plane laser cut sheet from an ultraviolet laser and the surface; Create a three-dimensional map of the surface as a function of a series of the data set; Use the data set to compare a measurement model of the surface to calculate an invariant index; and Output the invariant index. Such as the method of claim 21, wherein the invariant index is one of a normal distance and a vertical distance. Such as the method of claim 21, which further includes: using the data set to compare the measurement model to calculate a series of invariant indicators. Such as the method of claim 21, which further includes: using one of a table and a graph to display the series of invariant indicators. A method for providing optical reflectance information of a surface using an optical detection system, the method comprising: By triangulating a position of a fluorescent light imaged on a surface of a glass sheet using a camera, a data set including a set of coordinates corresponding to the position on the surface of the glass sheet is calculated A position where the fluorescent light is established at an intersection of a plane laser cut sheet from an ultraviolet laser and the surface; Create a three-dimensional map of the surface as a function of a series of the data set; De-noise the three-dimensional image of the surface; Use the data set from the denoising three-dimensional map to compare with an optical reflectance specification designated for the surface of the glass sheet to calculate an invariant index; and Output the invariant index. Such as the method of claim 25, wherein the three-dimensional image is denoised by the following methods: gridding the series of data sets, averaging the normal vectors of the grid neighborhood, and adjusting a position of the data sets To create a series of post-processed data sets. Such as the method of claim 25, wherein the three-dimensional image is de-noised by the following method: applying a moving least square algorithm. The method of claim 25, wherein the three-dimensional image is denoised by using one of an inflection point and a tortuous point to remove points adjacent to an edge of the three-dimensional image. The method of claim 25, wherein the invariant index includes at least one of a horizontal curvature and a vertical curvature. The method of claim 25, further comprising: constructing a simulated reflection optical image of one of a grid plate and a zebra plate by ray tracing the data set from the denoising map; and The simulated image is output.

Images