JP6670241B2 - Method for determining the shape of a substantially cylindrical specular surface - Google Patents

Description

Related technology cross-reference

  This application claims priority under 35 U.S.C. 119 of U.S. Provisional Patent Application No. 61/918262, filed December 19, 2013, which is hereby incorporated by reference in its entirety. Is incorporated herein by reference.

  The present invention relates generally to a method for determining a shape, and more particularly, to a method for determining a shape of a substantially cylindrical specular reflective surface.

  Generally, a glass ribbon that can be separated into glass plates is formed using a glass manufacturing apparatus. In some applications, there may be a desire to determine the shape of a glass ribbon, glass plate, or other glass element.

  A brief summary of the disclosure is provided below to provide a basic understanding of some exemplary aspects described in the detailed description.

  In a first aspect of the present disclosure, a method for determining a shape of a substantially cylindrical specular surface comprises the steps of: (I) obtaining calibration data; and target data for a target structure having a spatial frequency pattern. (II). The method further comprises a step (III) of capturing a reflection image of the target structure on the specular reflection surface, and a step (IV) of obtaining reflection data from the reflection image. The method further includes a step (V) of determining a correspondence between the target data and the reflection data, and a step (VI) of determining a shape of the specular reflection surface using the correspondence and the calibration data.

  In one embodiment of the first aspect, step (V) comprises a fast Fourier transform.

  In another embodiment of the first aspect, the spatial frequency pattern comprises a function of spatial position.

  In yet another embodiment of the first aspect, the specularly reflective surface extends substantially along a plane and the structure of interest is substantially perpendicular to the plane. For example, the spatial frequency pattern has a periodic pattern in a direction substantially parallel to the generatrix of the specular reflection surface. In another embodiment, the spatial frequency pattern includes frequencies that vary monotonically in a direction substantially perpendicular to the generatrix of the specular surface.

  In still another embodiment of the first aspect, the specularly reflective surface includes a main surface of the blank.

  In still another embodiment of the first aspect, the shape approximates the cross-sectional profile of the specular reflection surface.

  In yet another embodiment of the first aspect, the method further comprises determining a plurality of shapes of the specular surface, each shape approximating a cross-sectional profile of the specular surface. I have. For example, the method further comprises approximating a surface profile of the specular surface based on the plurality of shapes.

  The first aspect can be provided alone or in combination with one or any combination of the embodiments of the first aspect described above.

  In a second aspect of the present disclosure, a method of determining a shape of a glass ribbon drawn from a large amount of molten glass includes the steps of obtaining calibration data (I) and subject data relating to a subject structure having a spatial frequency pattern. And obtaining (II). The method further comprises the step of capturing a reflection image of the target structure on the glass ribbon (III) and the step of obtaining reflection data from the reflection image (IV). The method further includes determining a correspondence between the target data and the reflection data (V), and determining a shape of the glass ribbon using the correspondence and the calibration data (VI).

  In one embodiment of the second aspect, step (V) comprises a fast Fourier transform.

  In another embodiment of the second aspect, the glass ribbon extends substantially along a plane and the structure of interest is substantially perpendicular to the plane. For example, the spatial frequency pattern has a periodic pattern in a direction substantially parallel to the generatrix of the specular reflection surface. In another embodiment, the spatial frequency pattern includes frequencies that vary monotonically in a direction substantially perpendicular to the generatrix of the specular surface.

  In yet another embodiment of the second aspect, the glass ribbon moves continuously in the stretching direction.

  In yet another embodiment of the second aspect, the shape is used to control upstream parameters of the glass forming method.

  In yet another embodiment of the second aspect, the shape is used to control downstream process parameters.

  In yet another embodiment of the second aspect, the shape is used to control the upstream parameters and the downstream process parameters of the glass forming method.

  In yet another embodiment of the second aspect, the shape is used to determine an attribute of the glass ribbon, and the quality of the glass ribbon is classified based on the attribute.

  The second aspect can be provided alone or in combination with one or any combination of the embodiments of the second aspect described above.

  These and other aspects will be better understood by reading the following detailed description with reference to the accompanying drawings.

Illustration showing an exemplary cylindrical surface Figure showing another exemplary cylindrical surface FIG. 7 illustrates yet another exemplary cylindrical surface. FIG. 3 is a top view illustrating an exemplary configuration in which the subject structure is substantially perpendicular to the specular surface Side view of the configuration of FIG. 4 is a perspective view of the configuration of FIG. FIG. 4 illustrates an exemplary method comprising obtaining target data of a target structure having a spatial frequency pattern, capturing a reflection image of the target structure on a specular reflection surface, and obtaining reflection data from the reflection image. FIG. 4 shows an exemplary shape of a specular surface, which can be determined by the method described above. FIG. 6 shows another exemplary shape of a specular surface, which can be determined by the method described above. Exemplary flowchart showing steps of the method described above Schematic diagram of an exemplary apparatus for manufacturing a glass ribbon FIG. 15 is an enlarged cross-sectional perspective view, taken along line 2-2, of the apparatus of FIG. 14, illustrating an exemplary method in which the subject structure is substantially perpendicular to the glass ribbon. FIG. 4 illustrates an exemplary method comprising obtaining spatial frequency encoded target data of a target structure, capturing a reflection image of the target structure on the glass ribbon, and obtaining reflection data from the reflection image. FIG. 4 illustrates an example method including one or more target structures. Diagram showing an exemplary black and white spatial frequency pattern Illustration showing an exemplary grayscale spatial frequency pattern

  Hereinafter, examples will be described in more detail with reference to the accompanying drawings showing exemplary embodiments. Throughout the drawings, the same reference numbers are used for identical or similar parts wherever possible. However, aspects may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

  Aspects of the present disclosure include a method for determining a shape, specifically a method for determining a shape of a substantially cylindrical specular reflective surface. A specular reflection surface can exhibit the property of reflecting an incident light beam at the same angle with respect to the surface normal. For example, the angle of incidence is equal to the angle of reflection. Further, the incident beam, reflected beam, and surface normal can all be in the same plane. Using the principle of deflectorometry, or more specifically, reflectometry, the shape of a specular surface can be determined from distortion, or the reflection of a specular surface can be determined from specular reflection. . For example, when considering a structure with a known geometric shape, the distorted reflection of the structure on the specular reflection surface is used to estimate the geometric characteristics of the specular reflection surface that caused the distorted reflection. can do. The reflection of the structure can be distorted for a variety of reasons, including due to surface curvature, defects, abnormalities, or irregularities. Analyzing the reflections, for example, by determining the correspondence between the features of the structure of known geometry and the corresponding reflections of the features of the structure of known geometry on the specular surface, The shape of the resulting surface can be back calculated or restored. This shape can be used for any number of applications, controls, or calculations, for example, to simulate or approximate a three-dimensional profile of a real specular surface.

  1 to 3 are diagrams showing examples of a cylindrical surface. FIG. 1 is a diagram illustrating an exemplary cylindrical surface 10 a defined by a series of parallel lines 11 passing through a curved surface 12. FIG. 2 illustrates another exemplary cylindrical surface 10b, which can be defined by moving a straight line known as the generating line 13 along a curve or path known as the quasi-line 14. It is. In yet another example, FIG. 3 shows a cylindrical surface 10c, which can be defined by projecting the start line 15a in direction 16 such that the start line 15a is parallel to the end line 15b. FIG. Using the methods provided herein, the shape of a substantially cylindrical specular reflective surface can be determined. For example, a surface shape that satisfies or slightly deviates from the mathematical or theoretical characterization of the cylindrical surface can be determined. In one embodiment, the substantially cylindrical specularly reflective surface may include a major surface of a blank, such as a blank of material, or a blank of material separated from the blank of material. For example, a substantially cylindrical specular reflective surface can include a major surface of a glass sheet, such as a glass ribbon or a glass sheet split from a glass ribbon. In another embodiment, the substantially cylindrical specular reflective surface may include an outer surface of an optical fiber or other object.

  If the object has a substantially cylindrical specular surface, the method may be used to form not only the shape of the substantially cylindrical specular surface, but also the shape of the object having the substantially cylindrical specular surface. Can be determined. In the following, for purposes of explanation, reference to a substantially cylindrical specular surface is to be understood that such a surface may exist as a free surface or as a surface of an object. As described above, the methods described herein can be used to determine the shape of such surfaces and / or objects having such surfaces.

  The method includes obtaining calibration data. Calibration data is an image that contains data from which data can be coded directly or indirectly into a computer, observe data using a detector, measure data using a sensor, or extract calibration data. It can be obtained in a variety of ways, including capturing. Examples of calibration data include coordinates or other information representing any property or properties of a system, component, or structure used in or by the method. For example, the calibration data may include the spatial location of components of the system, such as a camera, lens, or focus, information about a specular surface, the structure of interest and its associated features, or any other parameters, initial states, or related thereto. Can contain data. In another example, the calibration data can include reference points or coordinates that can be used to determine and define the spatial location or relationship between various system components, structures, and variables. For example, calibration data can be transformed from three-dimensional coordinates in real space to two-dimensional coordinates by a transformation matrix or other mathematical calculations. In yet another embodiment, the calibration data can be manipulated, combined, analyzed, or processed to further analyze, manipulate, and / or calculate the calibration data.

  The method further comprises obtaining target data for the target structure having the spatial frequency pattern. The subject data is an image containing data that can be directly or indirectly coded into a computer, observe the data using a detector, measure the data using a sensor, or extract the subject data. It can be obtained in a variety of ways, including capturing. Examples of target data include coordinates representing a spatial location or other reference characteristic associated with the target structure and / or its associated features, and any other information associated with the target structure. For example, the subject data can include reference points that can be used to determine and define any number of relationships or properties between the subject structure and / or its associated features and various system components, structures, and variables. . Further, these reference points can be transformed from three-dimensional coordinates in real space to two-dimensional coordinates by a transformation matrix or other mathematical calculations. In yet another embodiment, the subject data can be manipulated, combined, analyzed, or processed to further analyze, manipulate, and / or calculate the subject data.

  As shown in FIGS. 4 to 6, the specular reflection surface 20 can extend substantially along the plane 21, and the target structure 31 can be substantially perpendicular to the plane 21. The target structure 31 can have any one or more arrangements, shapes, structures, or sizes, including any of a plurality of features or characteristics associated with the structure. The target structure can be composed of any of a variety of materials. In one embodiment, the target structure can be composed of one or more materials that have desirable properties for use in various environments. In yet another embodiment, the light source can provide independent or sub-illumination of the target structure. For example, the target structure may emit light or diffusely reflect light from a dedicated light source or ambient light. In yet another embodiment, the subject structure may be dynamic, for example, in that the structure may have features or characteristics that can be changed, manipulated, or controlled automatically or manually at any time.

  As further shown in the embodiment of FIGS. 4-6, the target structure 31 is spaced 54 distances from the edge 24 of the specular reflection surface and 56 distances from the surface 26 of the specular reflection surface 20. Have a spatial frequency pattern 36 that can be substantially perpendicular to The spatial frequency pattern 36 of the target structure 31 can be located at a height position 57 along the height 58 of the specular reflection surface 20. The spatial frequency pattern 36 of the target structure 31 can be at any position on the target structure and can extend at various angles and / or directions relative to the specular reflection surface 20.

  In yet another embodiment, the target structure 31 can have a spatial frequency pattern in various ways. For example, the target structure 31 can be encoded with the spatial frequency pattern 36. In another embodiment, the spatial frequency pattern 36 is printed on the target structure 31, the target structure is manufactured to include the spatial frequency pattern as a feature of the target structure, and / or the spatial frequency pattern is applied to the target structure. Can be bonded or affixed. In another embodiment, spatial frequency pattern 36 may constitute a function of spatial position. For example, the spatial frequency pattern 36 can be defined as a function of spatial position. In yet another embodiment, shown in FIG. 6, the spatial frequency pattern 36 may have a periodic pattern in a direction 22 that is substantially parallel to the bus of the specular surface. In yet another embodiment, shown in FIG. 6, the spatial frequency pattern 36 may include frequencies that monotonically change in a direction 23 that is substantially perpendicular to the generatrix of the specular surface. In FIG. 6, the generatrix of the specular reflection surface exists on the plane 21 and is shown as an arrow indicating the direction 22.

  For example, as shown in FIGS. 15 and 16, the spatial frequency patterns 600, 601 can include a periodic pattern that can be constant in a direction 605 and a frequency that can be varied in a direction 610. 15 and 16, a direction 605 corresponds to a direction substantially parallel to the generatrix of the specular reflection surface, and a direction 610 corresponds to a direction substantially perpendicular to the generatrix of the specular reflection surface. Yes, it is. Further, as shown in FIGS. 15 and 16, the spatial frequency patterns 600 and 601 may be black and white spatial frequency patterns 600 or grayscale spatial frequency patterns 601. In yet another embodiment, the spatial frequency pattern can have any color, curve, or feature.

  As shown in FIG. 7, the method further includes capturing a reflection image 50 of the target structure 31 on the specular reflection surface 20. The reflected image 50 can be captured using an image capture device 51, such as a camera or other image or video recording device. After capture, the reflected image 50 can be analyzed, or transferred to a computer 52 to extract, process, and / or analyze the data contained in the image.

  As further shown in FIG. 7, the method further comprises the step of obtaining reflection data 55 from the reflection image 50. Reflection data 55 can be obtained in a variety of ways, including extracting, processing, and / or analyzing reflection image 50 to obtain reflection data 55. Examples of the reflection data 55 include a spatial position or other reference characteristic associated with the reflection image 50 of the target structure, and / or coordinates representing the reflection features associated therewith, and any other associated with the reflection image 50. There is information. For example, the reflection data 55 can include information on the reflection 37 including the spatial frequency pattern 36 of the target structure 31. In another embodiment, the reflection data 55 determines and relates any number of properties or reflections 50 of the target structure and / or its associated reflection features to various system components, structures, and variables. A reference point that can be used to define may be included. Further, these reference points can be transformed from three-dimensional coordinates in real space to two-dimensional coordinates by a transformation matrix or other mathematical calculations. In yet another embodiment, the reflection data 55 can be manipulated, combined, analyzed, or processed to further analyze, manipulate, and / or calculate the reflection data.

  As described above, the target structure 31 can have a spatial frequency pattern 36 that can include any features or characteristics of the target structure 31. Thus, the reflections of the target structure 31 having the spatial frequency pattern 36 can include anything corresponding to the reflection of a feature or characteristic of the target structure 31, including the reflection 37 of the spatial frequency pattern. In one embodiment, the spatial frequency pattern 36 of the target structure 31 can be printed on the target structure 31 as described above. Thus, the reflection 37 of the spatial frequency pattern 36 of the target structure 31 may be a corresponding reflection of the target structure, including the reflection 37 of the printed spatial frequency pattern 36. Therefore, the reflection data 55 obtained from the reflection image 50 can include any information on the corresponding reflection 37 of the spatial frequency pattern 36 on the specular reflection surface 20.

  The method further includes determining a correspondence between the target data 41 and the reflection data 55. The correspondence relationship may include, for example, a comparison, a correlation, or any other one or more relationships between all or a portion of the target data 41 and all or a portion of the reflection data 55. For example, the target data 41 can be analyzed. In another embodiment, the reflection data 55 can be analyzed. In yet another embodiment, the target data 41 and the reflection data 55 can be analyzed. The correspondence may be determined in any of a variety of ways, including computation by computer or manual processing, mathematical calculations, or other methods. In one embodiment, the correspondence includes determining a distortion of the reflection 37 of the spatial frequency pattern 36 from the reflection data 55 of the reflection image 50 compared to the spatial frequency pattern 36 from the target data 41 of the target structure 31. be able to.

  In one embodiment, the correspondence may include a fast Fourier transform. In yet another embodiment, the spatial frequency pattern of the target structure 31 is determined for which target data 41 can be obtained and for the reflection 37 of the spatial frequency pattern captured in the reflected image 50 from which the reflected data 55 can be obtained. 36 can be subjected to a fast Fourier transform. For example, the dominant frequency in the reflection image corresponding to the target data and the reflection data can be determined. For example, starting from a known data point of the reflection data, another plurality of data points can be calculated. In another embodiment, a dominant frequency can be calculated using a fast Fourier transform, and can also include a correction of the most recent to the highest harmonic. In yet another embodiment, where the structure of interest has a spatial frequency pattern that exhibits a substantially linear dependence on frequency from column coordinates, deviations from the linear dependence in the reflected image may indicate distortion or motion of the specular surface 20. Can be shown.

  The method further includes determining the shape of the specular reflection surface 20 using the correspondence and the calibration data. In one embodiment, all or part of the correspondence may be used. In another embodiment, all or a portion of the calibration data may be used. In yet another embodiment, all or some of the correspondences can be used and all or some of the calibration data can be used. For example, this step may include performing a shape recovery algorithm. The shape recovery algorithm can determine the shape of the specular reflection surface 20 using arbitrary data. For example, the shape of the specular reflecting surface 20 may be restored, recovered, back calculated, or based on the correspondence and calibration data, a contour or profile of the specular reflecting surface that will generate a captured reflected image 50 of the target structure 31. Can be determined by estimating

  As shown in FIG. 8, in one embodiment, the cross-sectional profile 70 of the specular reflection surface 20 can be approximated by the shape. The cross-sectional profile 70 may be, for example, a cross section of the specular reflection surface in a plane 75 intersecting with the specular reflection surface 20. In another embodiment, if the object having the specular surface is thin and has a thickness substantially less than the length and width, the cross-sectional profile 70 may be a line or curve lying in a plane 75 that intersects the specular surface 20. 71 can be approximated. In yet another embodiment, the method can further include determining a plurality of shapes 72 of the specular surface 20. For example, the cross-sectional profile 70 or the curve 71 of the specular reflection surface 20 can be approximated by each shape 73 of the plurality of shapes 72.

  In yet another embodiment, shown in FIG. 9, the method may further include approximating a surface profile 74 of the specular surface 20 based on the plurality of shapes 72. The surface profile 74 can be determined, for example, by arranging the plurality of shapes 72 spatially ordered based on the relationship between them. In one embodiment, digitally assembling the plurality of shapes 72 can generate a rendered image that can approximate the entire surface profile 77 of the specular surface 20. For example, when the specular reflection surface includes the main surface of the material plate, the actual shape of a part or all of the material plate can be approximated or simulated depending on the shape.

  Any of the steps of the method may be performed at any of the same or different time frequencies. For example, as shown in FIG. 10, a step 501 of acquiring calibration data, a step 502 of acquiring target data, a step 503 of capturing a reflection image, a step 503 of acquiring reflection data, a step 505 of determining a correspondence, and a step 505 of determining correspondence. The method step 500 comprising the step 506 of determining the shape of the specular surface using the correspondence and the calibration data can be performed at any same or different time frequency. In one embodiment, any of the steps can be performed at least once per second. In another embodiment, any of the steps can be repeated at a rate such that the repetition period approaches zero. For example, if the change in the shape of the cylindrical surface between iterations is not significant, any of the steps can be performed at a substantially continuous rate in time. In yet another embodiment, any of the steps can be performed at a rate defined by any number of variables. Further, each step can be performed once. In one embodiment, one or more steps can be performed once, while other steps can be performed multiple times.

  Step 501 of acquiring calibration data, step 502 of acquiring target data, step 503 of capturing a reflection image, step 504 of acquiring reflection data, step 505 of determining correspondence, and using the correspondence and calibration data, Any of the method steps 500 shown in FIG. 10, including the step 506 of determining the shape of the specular surface, may use various computer, numerical, mathematical, linear, non-linear, scientific, digital, electronic, or other techniques. Can be. Further, any configuration, estimation, operation, or calculation may be performed, together or alone, on any of the method steps described herein.

  For example, captured or acquired images can be analyzed using image analysis, and data contained in the images can be extracted from the images. In another embodiment, a specific region of the target structure, a specific region of the specular reflection surface, and / or a region of interest that can represent a reflection image of the target structure on the specular reflection surface can be defined. The region of interest can be defined by the user and can be determined directly or indirectly by programming the computer or automatically using software routines or other procedures. In yet another embodiment, derivative convolution can be used to emphasize changes in the direction normal to the nominal features of the target structure. Derivative convolution can indicate, for example, the rate of change of the value of a data point between data points. By this processing, for example, a point having a maximum absolute value representing the maximum change with respect to the value of the data point perpendicular to the feature of the target structure is specified. In yet another example, data points may be filtered to remove or consider outliers points that are too far from the general orientation or directional tendency of features of the target structure. In yet another embodiment, sub-pixel interpolation can be used to determine the data point having the largest absolute value of the derivative. From this, using at least two points on each side of the data point, a polynomial can be fitted to the data point and the actual peak position can be determined. For example, this interpolation can be performed for each acquired data point that can be associated with a feature of the target structure, or a reflection corresponding to the feature on a specular surface. In yet another embodiment, an integrated methodology can be used and integration points can be defined. Since multiple shapes of the specular surface can result in the same reflection, the integration point can be used to establish a starting point for integration across the specular surface. In yet another embodiment, initial conditions for the differential equation recovery method can be specified. In yet another embodiment, three-dimensional point processing is used to define three-dimensional coordinates corresponding to data points of calibration data, target data, or reflection data, and to define a position of a target structure and a corresponding reflection. Can be converted to two-dimensional coordinates. In another example, data filtering can be performed, where calibration data, target data, or reflection data is processed to remove any outliers. In one embodiment, the filtering process includes fitting a polynomial line to data points that can be associated with corresponding reflections, for example, on features of the target structure and / or specular surfaces. In another embodiment, all data points outside a predetermined distance from the fitted line are identified as outliers. Outliers can be removed from the data set or retained in the data set. In yet another embodiment, the process of fitting a line, identifying outliers, and removing or retaining outliers from a data set uses the same or different polynomial fitting and / or the same or different outlier rejection limits. And can be repeated any number of times.

  Another aspect of the present disclosure includes a method for determining the shape of a glass ribbon 103 drawn from a large quantity of molten glass 121, as shown in FIG. After being manufactured, the glass ribbon 103 can be separated into glass plates 104 that can be used for various applications. For example, the glass plate 104 manufactured from the glass ribbon 103 can be used for display purposes and the like. In a specific example, the glass plate 104 can be used to manufacture a liquid crystal display (LCD), an electrophoretic display (EPD), an organic light emitting diode display (OLED), a plasma display panel (PDP), or other display device. it can.

  Glass ribbons can be manufactured by various devices, such as slot draw, float, down draw, fusion down draw, or up draw, for manufacturing glass ribbons in accordance with the present disclosure. Each apparatus can have a melting vessel configured to melt the batch material into a quantity of molten glass. Each apparatus further comprises at least a first conditioning station located downstream of the melting vessel and a second conditioning station located downstream of the first conditioning station.

  FIG. 11 is a schematic diagram of only one exemplary apparatus for manufacturing a glass ribbon in accordance with the present disclosure, the apparatus being used to melt-draw a glass ribbon 103 that is later processed into a glass plate 104. Is provided. The fusion draw apparatus 101 can have a melting vessel 105 configured to receive a batch material 107 from a raw material storage tank 109. The batch material 107 can be introduced by a batch supply device 111 using a motor 113 as a power source. An optional controller 115 for driving the motor 113 can be configured to introduce a desired amount of the batch material 107 into the melting vessel 105 as shown by arrow 117. The glass metal probe 119 is used to measure the level of the glass melt 121 in the standpipe 123, and the measurement information can be communicated to the controller 115 via the communication line 125.

  The fusion draw device 101 also has a first conditioning station, such as a fining vessel 127 (eg, a fining tube), located downstream of the melting vessel 105 and connected to the melting vessel 105 via a first connection pipe 129. be able to. In some embodiments, the glass melt can be gravity fed from the first melting vessel 105 to the fining vessel 127 via a connecting tube 129. For example, gravity can act on the glass melt to force the internal passage of the first connection pipe 129 from the melting vessel 105 to the fining vessel 127. In the fining container 127, air bubbles can be removed from the glass melt by various methods.

  The fusion draw device can also have a second conditioning station, such as a stirring vessel 131 (eg, a stirring chamber), which can be located downstream of the fining vessel 127. Using the stirring vessel 131, a uniform glass melt composition can be provided, and without the stirring vessel, non-uniform streaks that can be present in the clarified glass melt exiting the fining vessel are suppressed or Can be removed. As shown, the clarification vessel 127 can be connected to the stirring vessel 131 via a second connection pipe 135. In some embodiments, the glass melt can be gravity fed from the fining vessel 127 to the stirring vessel 131 via the second connecting pipe 135. For example, gravity can act on the glass melt to force the internal passage of the second connection pipe 135 from the melt fining vessel 127 to the stirring vessel 131.

  The fusion draw device can further include another conditioning station, such as a supply vessel 133 (eg, a bowl), which can be located downstream of the stirring vessel 131. The supply container 133 can adjust the glass supplied to the forming apparatus. For example, the supply container 133 can function as an accumulator and / or a flow regulator for adjusting the glass melt and constantly supplying a constant amount of the glass melt to the forming apparatus. As shown, the stirring vessel 131 can be connected to the supply vessel 133 via the third connection pipe 137. In some embodiments, the glass melt can be gravity fed from the stirring vessel 131 to the supply vessel 133 via the third connection pipe 137. For example, gravity acts on the glass melt to force the internal passage of the third connection pipe 137 from the stirring vessel 131 to the supply vessel 133.

  Further, as shown, a downcomer 139 can be provided to supply the glass melt 121 from the supply container 133 to the injection port 141 of the forming container 143. As shown, the melting vessel 105, the fining vessel 127, the stirring vessel 131, the supply vessel 133, and the forming vessel 143 are examples of a glass melt adjustment station that can be continuously arranged along the fusion draw apparatus 101. .

  The melting vessel 105 is generally made of a refractory material such as a refractory (eg, ceramic) brick. Fusion draw apparatus 101 can generally further include components comprising platinum or a platinum-containing metal, such as platinum-rhodium, platinum-iridium, and combinations thereof, wherein the components are molybdenum, palladium. , Rhenium, tantalum, titanium, tungsten, ruthenium, osmium, zirconium, and alloys thereof, and / or refractory metals such as zirconium dioxide. The components containing platinum include a first connection pipe 129, a fining vessel 127 (for example, a fining pipe), a second connection pipe 135, a standing pipe 123, a stirring vessel 131 (for example, a stirring chamber), and a third connection pipe. It may include one or more of a tube 137, a supply container 133 (eg, a bowl), a downcomer 139, and an inlet 141. The forming container 143 is also made of a refractory material and is designed to form the glass ribbon 103.

  FIG. 12 is a perspective sectional view taken along line 2-2 of the fusion draw device 101 of FIG. As shown, the molding container 143 has a molding wedge 201 having a pair of downwardly inclined molding surface portions 207, 209 extending between opposing ends. The pair of downwardly inclined forming surface portions 207 and 209 join along the extending direction 211 to form a root portion 213. A stretching surface 215 extends through the root 213 and the glass ribbon 103 can be stretched along the stretching surface 215 in a stretching direction 211, eg, a downstream direction. As shown, the extension surface 215 can bisect the root 213, but the extension surface 215 can also extend in other directions relative to the root 213.

  As shown in FIG. 11, the fusion draw apparatus 101 can include a system 300 that performs a method for determining the shape of a glass ribbon 103 drawn from a large quantity of molten glass 121. The method can also be performed to determine the shape of other objects having specular reflection properties, including optical fibers and other glass elements. A method for determining the shape of the glass ribbon 103 drawn from a large amount of the molten glass 121 will be described below. In one embodiment, the glass ribbon 103 can be continuously moved in the stretching direction 211. In another embodiment, the shape of the glass ribbon can be used to control parameters 301 upstream of the glass forming apparatus 101. In yet another embodiment, the shape of the glass ribbon can be used to control parameters of the downstream process 302. In yet another embodiment, the shape of the glass ribbon can be used to control the upstream parameters 301 and the parameters of the downstream process 302 of the glass forming apparatus 101. In yet another embodiment, the shape of the glass ribbon can be used to determine attributes of the glass ribbon, and the quality of the glass ribbon can be classified based on the attributes.

  For example, attributes can include morphological abnormalities, such as inclusions, scratches, or any other defects or irregularities that can occur in the glass ribbon during the molding process. Such anomalies can cause the glass ribbon to deviate from the required specification properties or parameters, and can cause the glass ribbon or glass sheet to be rejected or specified for another use. In another embodiment, the attribute may be indicative of the movement of the glass ribbon or a change in the shape or composition of the glass ribbon. By monitoring these attributes at different locations on the glass ribbon and at different times throughout the forming and / or processing steps, the forming and / or processing steps can be controlled and various glass forming And / or process parameters can be adjusted or matched. These attributes can be monitored periodically, repeatedly, or continuously, and can be used, for example, to generate various output information such as plots, graphs, charts, databases, or numerical data. be able to. In another example, attributes can be associated with a particular glass sheet cut from a glass ribbon. If the properties deviate from the required specifications, the specific glass sheet can then be discarded and further processed if necessary, or specified for a specific application or based on a specific location based on attributes. Allocated to In yet another embodiment, the attributes may be used to determine operating conditions corresponding to stable production, where the quality of the glass ribbon and / or the quality of the glass sheet is a desired quality or characteristic. In yet another embodiment, operating conditions corresponding to undesired production in which the quality of the glass ribbon and / or the quality of the glass sheet is different from the quality of the glass ribbon or glass sheet exhibiting the desired quality or properties may be determined. . In yet another embodiment, attributes can be used to notify a computer or user when a particular component, system, or feature of a glass forming apparatus is functioning properly or improperly. For example, based on certain attributes of the glass ribbon, determined from the shape calculated by the methods disclosed herein, when certain elements of the system require replacement or repair, or for producing molten glass. Various inputs can be adjusted to determine, for example, cases where the quality of the glass ribbon and / or glass sheet can be improved. Further, correlations between attributes can be determined. Such a correlation can be determined over a period of time and can include any of a variety of many parameters relating to the glass forming process, glass ribbon, and / or glass sheet provided by the method or provided by other controls. it can. In yet another embodiment, the shape of the glass ribbon and / or glass sheet can be used to ascertain changes in glass sheet properties such as glass forming, glass ribbon properties, and mechanical stress of the glass sheet. The shape can be monitored and / or analyzed, for example, to improve quality, efficiency, or other features, parameters, or aspects associated with the methods described herein.

  The method has a method of obtaining calibration data. As described above, calibration data is data that can directly or indirectly code data into a computer, observe data using a detector, measure data using a sensor, or extract calibration data. Can be obtained in a variety of ways, including capturing images containing Examples of calibration data include coordinates or other information representing any property or properties of a system, component, or structure used in or by the method. For example, the calibration data may include the spatial location of components of the system such as a camera, lens, or focus, information about the glass ribbon, the target structure and its features, or any other parameters, initial conditions, or data associated therewith. Can be included. In another example, the calibration data may include reference points or coordinates that can be used to determine and define locations or relationships between various system components, structures, and variables. For example, calibration data can be transformed from three-dimensional coordinates in real space to two-dimensional coordinates by a transformation matrix or other mathematical calculations. In yet another embodiment, the calibration data can be manipulated, combined, analyzed, or processed to further analyze, manipulate, and / or calculate the calibration data.

  The method further comprises obtaining target data for the target structure having the spatial frequency pattern. As described above, the target data may be data that can be directly or indirectly coded into a computer, observe the data with a detector, measure the data with a sensor, or extract the target data. Can be obtained in a variety of ways, including capturing images containing Examples of target data include coordinates representing a spatial location or other reference characteristic associated with the target structure and / or its associated features, and any other information associated with the target structure. For example, the subject data can include reference points that can be used to determine and define any number of relationships or properties between the subject structure and / or its associated features and various system components, structures, and variables. . Further, these reference points can be transformed from three-dimensional coordinates in real space to two-dimensional coordinates by a transformation matrix or other mathematical calculations. In yet another embodiment, the subject data can be manipulated, combined, analyzed, or processed to further analyze, manipulate, and / or calculate the subject data. In one embodiment, the target structure is an existing structure in the glass forming apparatus 101 that can serve other functions related to glass forming and processing in addition to serving as the target structure. Good. In another embodiment, the target structure may be a dedicated structure introduced into the glass forming apparatus 101 solely to function as the target structure in the methods described herein.

  As shown in FIG. 12, the glass ribbon 103 can extend substantially along the plane 215, and the target structure 331 can be substantially perpendicular to the plane 215. The target structure 331 can have any one or more arrangements, shapes, structures, or sizes, including any of a plurality of features or characteristics associated with the structure. The target structure can be composed of any of a variety of materials used in various environments. For example, in the glass forming apparatus 101, the target structure can be made of an appropriate material that can withstand a high-temperature environment. In yet another embodiment, the light source can provide independent or sub-illumination of the target structure. For example, the target structure may emit light or diffusely reflect light from a dedicated light source or ambient light. For example, the target structure 331 can reside within the fusion draw device 101 and provide a viewport of a light source that illuminates the target structure, including windows or other openings. The window or other opening is an existing window or opening present in the fusion draw device, or a dedicated window included solely to provide a viewport of the light source illuminating the target structure. Or it may be an opening. In yet another embodiment, the subject structure may be dynamic, for example, in that the structure may have features or characteristics that can be changed, manipulated, or controlled automatically or manually at any time.

  In another embodiment, shown in FIG. 12, the target structure 331 is spaced 354 from the edge 324 of the glass ribbon and 356 from the surface 326 of the glass ribbon 103 and substantially with respect to the glass ribbon 103. There can be a spatial frequency pattern 336 that can be vertical. The spatial frequency pattern 336 of the target structure 331 can be located at a height position 357 along the height 358 of the glass ribbon 103. The spatial frequency pattern 336 of the target structure 331 can be at any position on the target structure, and can extend at various angles and / or directions relative to the glass ribbon 103. In another embodiment, spatial frequency pattern 336 may constitute a function of spatial position. For example, the spatial frequency pattern 336 can be defined as a function of spatial position. In yet another embodiment, shown in FIG. 12, the spatial frequency pattern 336 may have a periodic pattern in a direction 322 that is substantially parallel to the generatrix of the glass ribbon. In yet another embodiment, shown in FIG. 12, the spatial frequency pattern 336 may include a frequency that monotonically changes in a direction 323 that is substantially perpendicular to the glass ribbon generatrix. In FIG. 12, the generatrix of the glass ribbon lies in plane 215 and is shown as an arrow representing direction 211.

  As shown in FIG. 13, the method further includes capturing a reflection image 350 of the target structure 331 on the glass ribbon 103. As described above, the reflected image 350 can be captured using an image capture device 351 such as a camera or other image or video recording device. After capture, the reflected image 350 can be analyzed or transferred to a computer 352 to extract, process, and / or analyze the data contained in the image.

  As shown in FIG. 14, one or more image capture devices 351 can be used to capture one or more reflected images 350 of one or more target structures 331. In another embodiment, shown in FIG. 14, one or more reflected images 350 can be captured at various locations on the glass ribbon 103. In yet another example, the reflection image 350 can include any or all reflections of features of the target structure, as well as any or all reflections of the target structure. For example, an image capturing device 351 such as a camera is disposed on one side of the glass ribbon 103, and the position of the image capturing device capturing an image and the reflection image 350 of the target structure of the glass ribbon on the opposite side of the glass ribbon are displayed. Can be captured by the image capturing device 351. For example, the image capture device can capture the reflected image over about half the width of the glass ribbon 103. In another embodiment, a second image capture device 351 such as a second camera is positioned at the same or similar height as the first image capture device on the opposite side of the glass ribbon, and a second image capture device 351 is provided. The device can also capture the position of the image capture device capturing the image and the reflected image of the target structure opposite the glass ribbon. Similarly, this image capture device can capture the reflected image over approximately half the width of the glass ribbon. The first image capturing device and the second image capturing device can capture a reflection image of the target structure over the entire width of the glass ribbon, for example. In yet another embodiment, the first image capture device and the second image capture device can be configured to capture a reflected image that includes the overlapping region of the glass ribbon. The overlap region can be used, for example, for calibration or other configuration calculations where multiple data points corresponding to the same spatial location of the glass ribbon are beneficial.

  In yet another example, a property or aspect of the reflected image 350 can be captured based on the position or angle of the image capture device or devices relative to the glass ribbon 103. In yet another embodiment, the image capture device may not be able to be located at a position that ideally captures the reflected image due to obstacles and constraints. The position and / or angle of the image capture device can be manually or automatically adjusted to accommodate such obstacles or constraints, and the image capture device can be removed and access to the fusion draw device 101 to inspect and clean the device. For example, the image capture device 351 can be mounted on an adjustable mechanism so that it can be repaired. In yet another embodiment, the same or different image capture devices may be positioned and used in or by the glass ribbon 103, the target structure 331, and the glass forming apparatus 101 or processing step, Images of other components can be captured. In yet another embodiment, the image capture device 351 is positioned so that the glass ribbon 103, target structure 331, or other component is visible through the aforementioned existing or dedicated viewport window in the fusion draw device 101. be able to. In addition, the image capture device provides light or illumination to enhance the reflective properties of the glass ribbon 103, as well as to illuminate the target structure and the glass ribbon with light to enhance the quality of the image capture. It can be placed in close proximity.

  As further shown in FIG. 13, the method includes obtaining reflection data 355 from the reflection image 350. As described above, the reflection data 355 can be obtained in a variety of ways, including extracting, processing, and / or analyzing the reflection image 350 to obtain the reflection data 355. Examples of reflection data 355 include spatial position or other reference characteristics associated with the reflection image 350 of the target structure, and / or coordinates representing the reflection features associated therewith, and any other associated with the reflection image 350. There is information. For example, the reflection data 355 can include information about the reflection 337 including the spatial frequency pattern 336 of the target structure 331. In another embodiment, the reflection data 355 determines and defines the relationship between any number of properties or reflections 350 of the target structure and / or features associated therewith with various system components, structures, and variables. Reference points that can be used for Further, these reference points can be transformed from three-dimensional coordinates in real space to two-dimensional coordinates by a transformation matrix or other mathematical calculations. In yet another embodiment, the reflection data 355 can be manipulated, combined, analyzed, or processed to further analyze, manipulate, and / or calculate the reflection data.

  As described above, the target structure 331 can have a spatial frequency pattern 336, which can include any features or characteristics of the target structure 331. Accordingly, the reflections of the target structure 331 that include the spatial frequency pattern 336 can include anything that corresponds to the reflection of a feature or characteristic of the target structure 331, including the spatial frequency pattern reflection 337. In one embodiment, the spatial frequency pattern 336 of the target structure 331 can be encoded into the target structure 331, as described above. In another example, the spatial frequency pattern 336 of the target structure 331 can be printed on the target structure 331. Thus, the reflection 337 of the spatial frequency pattern 336 of the target structure 331 may be a corresponding reflection of the target structure, including the reflection 337 of the spatial frequency pattern 336. Thus, the reflection data 355 obtained from the reflection image 350 can include any information about the corresponding reflection 337 of the spatial frequency pattern 336 in the glass ribbon 103.

  The method further includes determining a correspondence between the target data 341 and the reflection data 355. As described above, the correspondence relationship may include, for example, a comparison, a correlation, or any other one or more relationships between all or a portion of the target data 341 and all or a portion of the reflection data 355. . For example, the target data 341 can be analyzed. In another example, the reflection data 355 can be analyzed. In yet another embodiment, the target data 341 and the reflection data 355 can be analyzed. The correspondence may be determined in any of a variety of ways, including computation by computer or manual processing, mathematical calculations, or other methods. In one embodiment, the correspondence includes determining a distortion of the reflection 337 of the spatial frequency pattern 336 from the reflection data 355 of the reflection image 350 compared to the spatial frequency pattern 336 from the target data 341 of the target structure 331. be able to.

  In one embodiment, the correspondence may include a fast Fourier transform. In yet another embodiment, the spatial frequency pattern of the target structure 331 may be based on which target data 341 can be obtained and the reflection 337 of the spatial frequency pattern captured in the reflected image 350 from which the reflection data 355 can be obtained. A fast Fourier transform can be performed on 336. For example, a dominant frequency and reflection data in the reflection image corresponding to the target data can be determined. For example, starting from a known data point of the reflection data, another plurality of data points can be calculated. In another embodiment, a dominant frequency can be calculated using a fast Fourier transform, and can also include a correction of the most recent to the highest harmonic. In yet another embodiment, where the structure of interest has a spatial frequency pattern that exhibits a substantially linear dependence on frequency from column coordinates, deviations from linear dependence in the reflected image can cause distortion or movement of the glass ribbon 103. Can be shown.

  The method further includes determining the shape of the glass ribbon 103 using the correspondence and the calibration data. In one embodiment, all or part of the correspondence may be used. In another embodiment, all or a portion of the calibration data may be used. In yet another embodiment, all or some of the correspondences can be used and all or some of the calibration data can be used. For example, this step may include performing a shape recovery algorithm. The shape recovery algorithm can determine the shape of the glass ribbon 103 using arbitrary data. For example, the shape of the glass ribbon 103 estimates the contour or profile of the glass ribbon that will generate a captured reflected image 350 of the target structure 331 based on the restoration, recovery, back calculation, or correspondence and calibration data. Can be determined.

  As shown in FIG. 8, in one embodiment, the cross-sectional profile 70 of the glass ribbon 103 can be approximated by the shape. The cross-sectional profile 70 may be, for example, a cross-section of the glass ribbon 103 at a plane 75 that intersects the glass ribbon 103. In another embodiment, if the glass ribbon is thin and has a thickness substantially less than its length and width, the cross-sectional profile 70 may be approximated as a line or curve 71 lying in a plane 75 that intersects the glass ribbon 103. Can be. For glass ribbons 103, glass plates 104, or other transparent materials where reflection of an object or object structure can occur on both surfaces, the shape can be determined by considering the Fresnel reflection coefficient. In yet another embodiment, the method can further include determining a plurality of shapes 72 of the glass ribbon 103. For example, the cross-sectional profile 70 or the curve 71 of the glass ribbon 103 can be approximated by each shape 73 of the plurality of shapes 72.

  In yet another embodiment, shown in FIG. 9, the method may further include approximating a surface profile 74 of the glass ribbon 103 based on the plurality of shapes 72. The surface profile 74 can be determined, for example, by arranging the plurality of shapes 72 spatially ordered based on the relationship between them. In one embodiment, the plurality of shapes 72 can be assembled digitally to produce a rendered image that can approximate the entire surface profile 77 of the glass ribbon 103. For example, the shape may approximate or simulate the actual shape of some or all of the glass ribbon 103 and / or the actual shape of some or all of the glass plate 104 cut from the glass ribbon.

  Any of the foregoing steps can be performed at any of the same or different time frequencies. For example, as shown in FIG. 10, a step 501 of acquiring calibration data, a step 502 of acquiring target data, a step 503 of capturing a reflection image, a step 503 of acquiring reflection data, a step 505 of determining a correspondence, and a step 505 of determining correspondence. The method step 500 comprising the step 506 of determining the shape of the specular surface using the correspondence and the calibration data can be performed at any same or different time frequency. In one embodiment, any of the steps can be performed at least once per second. In another embodiment, any of the steps can be repeated at a rate such that the repetition period approaches zero. For example, if the shape change of the glass ribbon between repetitions is not significant, any of the steps can be performed at a substantially continuous rate in time. In yet another embodiment, any of the steps can be performed at a rate defined by any number of variables. In one embodiment, both steps can be performed at a rate consistent with one per glass sheet. In another embodiment, any step may contribute to or change the size of the glass sheet, the quality of the glass sheet being manufactured or already manufactured, or glass forming equipment and other processes. At an adjusted speed based on the Further, each step can be performed once. In one embodiment, one or more steps can be performed once, while other steps can be performed multiple times.

  Obtaining the calibration data, obtaining the target data, defining the target line from the target data, capturing the reflection image, obtaining the reflection data, defining the reflection line, and determining the correspondence. Various computer, numerical, mathematical, linear, non-linear, scientific, digital, and / or numerical steps may be used for any of the steps of the method, including determining the shape of the glass ribbon using the correspondences and calibration data. Electronic or other techniques can be used. Any configuration, estimation, operation, or calculation may be performed together or alone for any of the method steps described herein.

  It will be apparent to those skilled in the art that various modifications and variations can be made without departing from the spirit and scope of the claimed invention.

  Hereinafter, preferred embodiments of the present invention will be described separately.

Embodiment 1
In a method for determining a shape of a substantially cylindrical specular reflecting surface,
(I) obtaining calibration data;
(II) obtaining target data on a target structure having a spatial frequency pattern;
(III) capturing a reflection image of the target structure on the specular reflection surface;
(IV) obtaining reflection data from the reflection image;
(V) determining a correspondence between the target data and the reflection data; and (VI) determining the shape of the specular reflection surface using the correspondence and the calibration data.

Embodiment 2
The method of embodiment 1, wherein step (V) comprises a fast Fourier transform.

Embodiment 3
2. The method of embodiment 1, wherein the spatial frequency pattern comprises a function of spatial position.

Embodiment 4
2. The method of embodiment 1, wherein the specularly reflective surface extends substantially along a plane, and the target structure is substantially perpendicular to the plane.

Embodiment 5
5. The method of embodiment 4, wherein the spatial frequency pattern has a periodic pattern in a direction substantially parallel to a generatrix of the specular reflection surface.

Embodiment 6
6. The method of embodiment 5, wherein the spatial frequency pattern includes a frequency that monotonically changes in a direction substantially perpendicular to a generatrix of the specular reflection surface.

Embodiment 7
The method of embodiment 1, wherein the specularly reflective surface comprises a major surface of a blank.

Embodiment 8
The method of embodiment 1, wherein the shape approximates a cross-sectional profile of the specular reflection surface.

Embodiment 9
2. The method of embodiment 1, wherein determining a plurality of shapes of the specular surface further comprises the step of each shape approximating a cross-sectional profile of the specular surface.

Embodiment 10
10. The method of embodiment 9, further comprising approximating a surface profile of the specular reflection surface based on the plurality of shapes.

Embodiment 11
In a method of determining the shape of a glass ribbon drawn from a large amount of molten glass,
(I) obtaining calibration data;
(II) obtaining target data on the target structure having a spatial frequency pattern;
(III) capturing a reflection image of the target structure on the glass ribbon;
(IV) obtaining reflection data from the reflection image;
(V) determining a correspondence between the target data and the reflection data; and (VI) determining the shape of the glass ribbon using the correspondence and the calibration data.

Embodiment 12
12. The method of embodiment 11, wherein step (V) comprises a fast Fourier transform.

Embodiment 13
12. The method of embodiment 11, wherein the glass ribbon extends substantially along a plane, and the structure of interest is substantially perpendicular to the plane.

Embodiment 14
14. The method of embodiment 13, wherein the spatial frequency pattern has a periodic pattern in a direction substantially parallel to a bus of the glass ribbon.

Embodiment 15
15. The method of embodiment 14, wherein the spatial frequency pattern comprises a frequency that monotonically changes in a direction substantially perpendicular to the glass ribbon generatrix.

Embodiment 16
Embodiment 12. The method of embodiment 11, wherein the glass ribbon moves continuously in a stretching direction.

Embodiment 17
12. The method of embodiment 11, wherein the shape is used to control an upstream parameter of a glass forming method.

Embodiment 18
12. The method of embodiment 11, wherein the shape is used to control downstream process parameters.

Embodiment 19
Embodiment 12. The method of embodiment 11, wherein the shape is used to control an upstream parameter and a downstream process parameter of the glass forming method.

Embodiment 20
12. The method of embodiment 11, wherein the shape is used to determine an attribute of the glass ribbon, and the quality of the glass ribbon is classified based on the attribute.

10a, 10b, 10c Cylindrical surface 20 Specular reflection surface 31, 331 Target structure 36, 336 Spatial frequency pattern 41, 341 Target data 50, 350 Reflected image 55, 355 Reflected data 51, 351 Image capture device 52, 352 Computer 101 Fusion Draw device 103 Glass ribbon 104 Glass plate 105 Melting container 127 Refining container 131 Stirring container 133 Supply container 143 Molding container 201 Molding wedge 207, 209 Molding surface portion 213 Root bottom 215 Stretched surface 300 System 600, 601 Spatial frequency pattern

Claims (10)

A method for determining a shape of a substantially cylindrical specular reflective surface, the method comprising:
(I) obtaining calibration data;
(II) a target data relating to the target structure, the step of the target data including spatial frequency pattern having a frequency which varies monotonically in the direction perpendicular to the generatrices of the specular surface, acquires the target data,
(III) capturing a reflection image of the target structure on the specular reflection surface having a continuously moving glass ribbon;
(IV) obtaining reflection data from the reflection image;
(V) determining a correspondence between the target data and the reflection data; and (VI) determining the shape of the specular reflection surface using the correspondence and the calibration data. A method comprising:   The method of claim 1, wherein the specularly reflective surface extends substantially along a plane, and the target structure is substantially perpendicular to the plane.   3. The method of claim 2, wherein said spatial frequency pattern comprises a periodic pattern in a direction substantially parallel to a generatrix of said specular reflection surface.   The method of claim 1, wherein step (V) comprises a fast Fourier transform.   5. The method according to claim 1, wherein the specular reflection surface comprises a main surface of a blank.   The method according to claim 1, wherein the shape approximates a cross-sectional profile of the specular reflection surface.   6. The method according to claim 1, further comprising the step of determining a plurality of shapes of the specular reflection surface, wherein each shape further includes a step of approximating a cross-sectional profile of the specular reflection surface. Or the method of claim 1.   The method of claim 1, wherein the shape is used to determine an attribute of the glass ribbon, and the quality of the glass ribbon is classified based on the attribute.   9. The method according to claim 6, wherein the shape is used to control an upstream parameter of a glass forming method.   The method according to any one of claims 6 to 9, wherein the shape is used to control downstream parameters of a glass forming method.

Images