Method for determining an optical distortion of a glazing, and associated electronic device

Abstract

The invention relates to a method for determining an optical distortion of a glazing (320) comprising: a determination, at each point of a set of points of a surface of the glazing, of a datum representative of an angular deviation undergone, at the output of the glazing, by a light ray passing through said glazing; and a determination, at each point of the set, of a tangent value, referred to as the "first tangent value", by applying a partial derivative to the datum representative of an angular deviation.

Description

Description Title of the invention: Method for determining the optical distortion of glazing, and associated electronic device Technical Field

[0001] The present invention falls within the general field of optical quality control of glazing. More particularly, it relates to a method for determining the optical distortion of glazing. It also relates to an electronic device configured to implement such a method. It finds a particular, though not limiting, application in the production of glazing intended for use in aircraft. Previous technique

[0002] The optical quality of glazing is a major issue, particularly when intended for use in the aeronautical or aerospace sectors, since glazing – whether cockpit windows or portholes – plays a crucial role in the safety and efficiency of air operations.

[0003] The optical quality of glazing is assessed, in particular, by the optical distortion it induces. Optical distortion manifests as undesirable visual deformations of an object or scene viewed through glazing. The origins of these visual distortions can be varied, and may result, for example, from local variations in the thickness of the glazing, significant curvature of the glazing, misalignment of the panes composing the glazing, or heterogeneity in the optical index of the glazing's constituent elements, for example, due to the presence of heating elements – such as heating wire arrays – etc.

[0004] In an environment where visibility and information perception are paramount, the optical quality of glazing must meet strict standards and rigorous technical requirements. To assess the quality of glazing intended for use in the aeronautical or aerospace sectors, the American Society for Testing and Materials (ASTM) has published a method for quantifying optical distortion induced by a transparent object, using the grid line slope method. This quantification method is described in the document "Standard Test Method for Measuring Optical Distortion in Transparent Parts using Grid Line Slope," F2156, version 17, published in 2022 by ASTM.According to this method, the optical distortion induced by a transparent part is characterized by the tangent of the apparent rotation angle of vertical and horizontal lines observed through the glazing.

[0005] Figure 1 illustrates a 100 optical distortion measurement system conforming to the previously mentioned ASTM F2156 standard, from a side view defined by a plane (

[0006] As illustrated in Figure 1, the measuring system 100 comprises a camera 110 and a grid 130. A transparent piece 120 is positioned between the camera 110 and the grid 130, at a predetermined distance from both the camera and the grid. More precisely, the ASTM F2156 standard requires one of the following configurations (or geometries):

[0007] - Configuration A: the distance (referenced (Figure 1) The distance between camera 110 and grid 130 is 1000 cm, the distance between camera 110 and transparent piece 120 (L2) is 550 cm, and the distance between transparent piece 120 and grid 130 (L3) is 450 cm. Implementing this configuration requires installing the measurement system in a room approximately 12 meters in size.

[0008] - Configuration B: The distance (L1) between camera 110 and grid 130 is 450 cm, the distance between camera 110 and transparent part 120 (L2) is 150 cm, and the distance between transparent part 120 and grid 130 (L3) is 300 cm. Implementing this configuration requires installing the measurement system in a room approximately 7 meters in diameter.

[0009] - configuration C: user-defined. The distance between camera 110 and transparent part 120 (L2) is for example between 500 and 3000 cm, and that between transparent part 120 and grid 130 (L3) is for example between 500 and 10,000 cm.

[0010] Generally, the greater the distances L1, L2, and L3, the more visible the distortions, and the more reliable their measurement. While it is common for L2 to represent a value representative of viewing conditions through the glazing (measurements known as "eye-eye" measurements), it is more difficult to test glazing with a value L3 representative of normal viewing conditions through the glazing. Therefore, the quantification of optical performance as measured during quality control is not entirely representative of the performance perceived by a glazing user. Furthermore, since not all portions of the glazing are at the same distance from the target, measurement biases are not uniform across all sections.

[0011] The camera 110 is positioned to record the pattern of the grid 130 as seen through the transparent piece 120 from its observation position. To achieve this, the camera 110 is focused on the grid 130.

[0012] The captured image is then printed and analyzed to assess the level of visual distortion. To do this, the slope of the 130 grid lines is calculated manually or using a plotter mounted on a drawing table.

[0013] Figure 2 illustrates the principle of calculating the optical distortion induced by a transparent part, in accordance with the ASTM F2156 standard.

[0014] As illustrated by Figure 2, a grid 220 - corresponding for example to grid 130 of Figure 1 - is partly seen through a transparent piece corresponding for example to piece 120 of Figure 1 and having a border 210, according to a view defined by a plane (%, y).

[0015] This 220 grid appears distorted when viewed through the transparent part. Indeed, the grid lines, when viewed through the transparent piece, appear curved, inclined, and offset horizontally and vertically.

[0016] To measure the distortion, the line with the steepest tangent curve is visually identified. In this example, point 230 corresponds to the point where the slope of the tangent (to a curve of a vertical line) is maximum. The tangent 240 to the curve and passing through point 230 then defines an angle and its associated tangent tan( ).

[0017] In this example, tan(Φ) = 0.2

[0018] This method allows the optical performance of glazing to be characterized by a single quantity (or by a quantity per optical zone of the glazing). However, it remains true that this method does not allow the detection of localized distortion between the grids of the target. Reducing the size of the target squares to obtain satisfactory spatial resolution would make the measurement impossible (overlapping lines and difficulty in visually visualizing line rotations). Furthermore, this method is not very robust since it is subject to human error, and its accuracy depends on the tools used (rulers, set squares, protractors).

[0019] Documents WO2021063847A1 and WO2023275486A1 disclose, respectively, a method for measuring and a method for analyzing the optical quality of a delimited area of ​​glazing, this delimited area being intended to be placed in front of a camera. Description of the invention

[0020] The present invention aims to remedy all or part of the drawbacks of the prior art, in particular those set out above, by proposing a solution which makes it possible to determine (quantify) an optical distortion induced by glazing, and to do so in a robust manner.

[0021] To this end, and according to a first aspect, the invention relates to a method for determining the optical distortion of glazing implemented by an electronic processing device, the method comprising:

[0022] - a determination, at each point of a set of points on a glazing surface, of a value representing the angular deflection undergone, upon exiting the glazing, by a light ray passing through said glazing; and,

[0023] - a determination, at each point of the set, of a tangent value, called the "first tangent value", by applying a partial derivative to the data representing an angular deviation.

[0024] The term "representative data for an angular deviation" refers to the angular deviation itself, or a spatial deviation on a plane defined by the screen that is a function of the angular deviation experienced by the light ray passing through the glazing as it exits the screen. When the deviations are uniform across the glazing, the spatial deviation is proportional to the angular deviation.

[0025] Generally speaking, the steps of a process should not be interpreted as being linked to a notion of temporal succession.

[0026] In particular modes of implementation, the method of determining an optical distortion according to this first aspect may further include one or more of the following characteristics, taken individually or according to all technically possible combinations.

[0027] In certain implementations, the entire glazing surface is discretized into a set of points, and the tangent value is determined at each point. In other implementations, the points are distributed regularly and uniformly across the surface. Alternatively, samples are randomly selected from the surface. Alternatively, the density of the sampling points is adjusted according to areas of interest.

[0028] In other specific implementation methods, the process is applied to one or more areas of interest, for example, the central area of ​​the glazing. This variant is advantageous because it eliminates potential edge effects in the glazing.

[0029] In particular embodiments, the electronic device is an electronic image processing device of a measurement system further comprising an image acquisition device connected to the electronic image processing device and a screen, the glazing being positioned between the image acquisition device and the screen, and the data representative of an angular deviation is determined from images captured by the image acquisition device, each image illustrating an effect of the glazing on a pattern of a series of phase-shifted fringe patterns or a series of phase-shifted grids displayed on the screen.

[0030] The effect induced by this glazing corresponds to the optical distortion that the process according to the invention aims to evaluate.

[0031] In certain implementations, the screen is an electronic device for displaying images. Alternatively, this screen is a medium onto which patterns are fixed or projected.

[0032] In certain implementation modes, the data representing an angular deviation corresponds to a spatial deviation on a plane defined by the screen, and the first d dx ''Ox — a with x, y the coordinates of a radius to ' x = ~ incident on said plane, dx and dy the horizontal and vertical spatial deviations, and d the partial derivative symbol.

[0033] In specific implementation modes, the data representing an angular deviation is determined by applying a fringe demodulation method including:

[0034] - obtaining the images captured by the image acquisition device;

[0035] - the determination of a discontinuous phase map from the images obtained; and,

[0036] - a phase unwinding applied to the discontinuous phase map, so as to obtain a continuous phase map corresponding to the representative data of an angular deviation determined at each of the points.

[0037] By "discontinuous phase map", we mean a spatial distribution of discontinuous phase.

[0038] In particular implementation modes, the data representing an angular deviation is determined at least from glazing thicknesses at each point of the assembly, for example from local variations in thicknesses.

[0039] Thus, in certain specific implementation modes, the data representing an angular deviation is determined from local variations in thickness and from the field of incidence (i.e., the angle formed between the incident light ray and the normal to the surface).

[0040] In this particular case, the data representing an angular deviation is, for example, determined by applying a partial derivative to each thickness value, and this data representing an angular deviation then corresponds to slopes or angular deviations.

[0041] In particular implementation modes, the data representing an angular deviation is further determined as a function of a refractive index of the media traversed by the light ray.

[0042] The first tangent value is determined when the measuring system is configured according to a first geometry. In certain implementations, this first geometry does not correspond to a "target geometry," and the process further includes:

[0043] - obtaining a proportionality factor that allows calculating a second tangent value according to a second geometry of the measurement system, starting from the first tangent value; and,

[0044] - a determination, at each point of the set of points, of the second value of tangent according to the second geometry of the measurement system, from the first value of tangent and the proportionality factor.

[0045] Within the framework of the invention, the geometry of the measurement system is defined in particular by a distance between the image acquisition device and the screen, a distance between the image acquisition device and the glazing and / or a distance between the glazing and the screen.

[0046] This feature is advantageous because it allows tangent values ​​to be obtained at so-called "target" distances, for example, those imposed by a standard such as ASTM F2156, without requiring a measurement system whose geometry conforms to the distances defined by that standard. In other words, this feature allows data requiring a spacious environment and a large screen to be obtained using a small screen in a similarly small environment. This feature is particularly advantageous when evaluating the distortion seen by a pilot observing a scene at a distance that can be considered infinite.

[0047] In certain implementation modes, the proportionality factor A is expressed as A = / (~^ x x )> avec f une mathematical function, l0 the distance between the image acquisition device and glazing according to the first geometry, l the distance between the image acquisition device and the glazing according to the second geometry, L o the distance between the glazing and the screen according to the first geometry and L the distance between the glazing and the screen according to the second geometry.

[0048] In certain implementation modes, the distances l and l0 are (approximately) identical, the distance L o is considered to be infinite (i.e. for example 100 meters when the glazing is 3 meters from the image acquisition device, and the proportionality factor is then expressed such that A = f 1 +.

[0049] In particular modes of implementation, a first proportionality factor A is applied to a first zone of the glazing, a second proportionality factor A is applied to a second zone of the glazing, the first and second zones having different geometries.

[0050] In particular implementation modes, the process further includes data filtering aimed at mimicking human visual perception.

[0051] In specific implementation modes, the filtering includes at least one of the following elements:

[0052] - an application of a denoising function to data representing an angular deviation;

[0053] - an application of a smoothing function to the first tangent values;

[0054] - a variable weight smoothing filtering based on the extremum of the data representing an angular deviation;

[0055] - a smoothing filter derived from a deep learning analysis of local glazing defects; and,

[0056] - a variable weight smoothing filter, depending on areas of the glazing.

[0057] In particular implementation modes, the filtering includes applying a smoothing function to the first tangent values, and applying a denoising function to the second tangent values.

[0058] In particular modes of implementation, the process further includes a classification of the glazing, for example according to the first and / or second tangent value at each of the points.

[0059] In specific implementation methods, the glazing is intended to equip a vehicle.

[0060] The vehicle is for example in a land vehicle (e.g., a car, a truck, a train) or an aerial vehicle (e.g., an aircraft, a helicopter, a plane, a drone).

[0061] According to a second aspect, the invention relates to an electronic processing device configured to implement a determination method according to the first aspect of the invention.

[0062] According to a third aspect, the invention relates to an electronic processing device comprising:

[0063] - a module for determining, at each point of a set of points on a glazing surface, a data point representing an angular deflection undergone, upon exiting the glazing, by a light ray passing through said glazing; and,

[0064] - a module for determining, at each point of the set, a tangent value, called "first tangent value", by applying a partial derivative to the data representing an angular deviation.

[0065] According to a fourth aspect, the invention relates to a method for determining the optical distortion of glazing implemented by an electronic image processing device of a measurement system further comprising an image acquisition device and a screen, the glazing being positioned between the image acquisition device and the screen, the method comprising:

[0066] - a determination, at each point of a set of points of a surface of the glazing, of a first value of tangent representative of an angular deviation undergone, at the exit of the glazing, by a light ray passing through said glazing, the first value being determined when the measurement system is configured according to a first geometry;

[0067] - obtaining a proportionality factor that allows calculating a second tangent value according to a second geometry of the measurement system, starting from the first tangent value determined when the measurement system is configured according to the first geometry; and,

[0068] - a determination, at each point of the set of points, of the second value of tangent according to the second geometry of the measurement system, from the first value of tangent and the proportionality factor.

[0069] The method according to the invention is advantageous since it allows tangent values ​​to be obtained relative to so-called "target" distances, for example imposed by a standard such as the ASTM F2156 standard, without it being necessary to use a measurement system whose geometry respects the distances defined by this standard.

[0070] In particular modes of implementation, the method for determining an optical distortion according to the fourth aspect may further include one or more of the following characteristics, taken individually or in all technically possible combinations.

[0071] In particular embodiments, the method further includes determining, at each point, a data representative of the angular deviation; and the first tangent value is determined from the data representative of the angular deviation.

[0072] In particular implementation modes, the data representing an angular deviation is determined from images captured by the image acquisition device, each image illustrating an effect of the glazing on a pattern of a series of phase-shifted fringe patterns or a series of phase-shifted grids displayed on the screen.

[0073] In particular implementation modes, the first tangent value is estimated by applying a partial derivative to the data representing an angular deviation.

[0074] In this particular case, we assume a so-called "small angle" approximation, meaning that the angles of the deviations are relatively small, and that tan(0) ~ 6.

[0075] In particular modes of implementation, obtaining the proportionality factor is a determination of this proportionality factor as a function of the first and second geometries.

[0076] To do this, considering the plan in lateral view of the measurement system, the field of view of the acquisition device is defined by two light rays passing through the glazing at two extrema, and the proportionality factor is calculated assuming that the distance (or "step") between these extrema remains the same, regardless of the first or second geometry considered.

[0077] In particular implementation methods, the process further includes a classification of the glazing, according to the second tangent value at each of the points.

[0078] According to a fifth aspect, the invention relates to an electronic image processing device configured to implement a determination method according to the fourth aspect of the invention.

[0079] According to a sixth aspect, the invention relates to an electronic image processing device for a measurement system further comprising an image acquisition device and a screen, the glazing being positioned between the image acquisition device and the screen, the electronic image processing device comprising:

[0080] - a module for determining, at each point of a set of points of a surface of the glazing, a first value of tangent representative of an angular deviation undergone, at the exit of the glazing, by a light ray passing through said glazing, the first value being determined when the measurement system is configured according to a first geometry;

[0081] - a module for obtaining a proportionality factor that allows calculating a second tangent value according to a second geometry of the measurement system, starting from the first tangent value determined when the measurement system is configured according to the first geometry; and,

[0082] - a module for determining, at each point of the set of points, the second value of tangent according to the second geometry of the measurement system, from the first value of tangent and the proportionality factor.

[0083] According to a seventh aspect, the invention relates to a computer program comprising instructions for implementing a method for determining optical distortion according to the first or fourth aspect, when said program is executed by a processor.

[0084] This program can use any programming language, and be in the form of source code, object code, or code somewhere between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0085] According to an eighth aspect, the invention relates to a computer-readable recording medium on which the computer program according to the invention is recorded.

[0086] The information or recording medium can be any entity or device capable of storing the program. For example, the medium may include a storage means, such as a ROM, for example a CD-ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a hard drive.

[0087] On the other hand, the information or recording medium can be a transmissible medium such as an electrical or optical signal, which can be transmitted via an electrical or optical cable, by radio, or by other means. The program according to the invention can, in particular, be uploaded to an internet-type network.

[0088] Alternatively, the information or recording medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the process in question. Brief description of the drawings

[0089] Other features and advantages of the present invention will become apparent from the description below, with reference to the accompanying drawings, which illustrate an example of an embodiment without being limiting in any way. In the figures:

[0090] [Fig.1] Figure 1, described previously, illustrates an optical distortion measurement system in accordance with the ASTM F2156 standard.

[0091] [Fig.2] Figure 2, described previously, illustrates the principle of calculating the optical distortion induced by a transparent part, in accordance with the ASTM F2156 standard.

[0092] [Fig.3] Figure 3 is a representation of a measurement system, according to a particular embodiment of the invention.

[0093] [Fig.4] Figure 4 represents modules embedded in an electronic image processing device, according to a particular implementation method of the invention;

[0094] [Fig.5] Figure 5 schematically represents an example of the hardware architecture of an electronic image processing device;

[0095] [Fig.6] Figure 6 illustrates, in general, the principle of the invention and the notations adopted in the description;

[0096] [Fig.7] Figure 7 illustrates a change in the geometry of the measurement system, according to a particular method of implementing the invention;

[0097] [Fig.8] Figure 8 represents, in the form of a flowchart, a particular method of implementing a process for determining the optical distortion of a glazing, for example carried out by the electronic device of figures 4 or 5. Description of the implementation methods

[0098] The terms "first(s)", "second(s)", etc. are used in this document by arbitrary convention to enable identification and distinction of different elements considered in the embodiments described below, and do not imply any particular sequencing unless explicitly stated.

[0099] Figure 3 is a representation of a 300 optical distortion measurement system, according to a particular embodiment of the invention.

[0100] More specifically, the 300 measurement system is configured to determine the optical distortion induced by a 320 glazing intended to be positioned in the field of view (i.e., on the optical path) of an observer or image acquisition device.

[0101] This 320 glazing, for example, is curved in one or more directions, transparent with a uniform color in reflection and / or transmission, and / or made of one or more sheets of glass, which may be uniform or vary in composition. 320 glazing is intended for use in vehicles, such as land vehicles (e.g., cars, trucks, trains) or aircraft (e.g., airplanes, helicopters, planes, drones).

[0102] In this embodiment, and for the sake of simplicity, the system 300 is considered to measure the optical distortion of a single pane of glass 320. It should be noted, however, that there is no limitation on the number of panes of glass tested. Indeed, in one particular implementation, the measuring system 300 is installed within a glass manufacturing plant, and the quality of each pane of glass produced by the plant is then tested.

[0103] As illustrated in Figure 3, the optical distortion measurement system 300 comprises an image acquisition device 310 and a screen 330.

[0104] The image acquisition device 310 is positioned so as to be able to capture the effect induced by a glazing on a pattern displayed on the screen 330. To do this, the glazing 320 whose optical distortion is measured is positioned between the image acquisition device 310 and the screen 330. More specifically, this positioning is carried out so that the screen 330 is positioned on the "external face" side of the glazing 320 (i.e. the side intended to be part of the external environment of the vehicle), the image acquisition device 310 being located on the "internal face" side of the glazing (i.e. the side intended to be part of the internal environment of the vehicle).

[0105] The 310 image acquisition device corresponds, for example, to a camera equipped with a CCD (Charge Coupled Device) type photographic sensor commonly used in digital imaging.

[0106] The 330 screen is considered to be positioned in a plane orthogonal to the optical axis. In a particular implementation, the 330 screen is an electronic image display device. As discussed in more detail below, the patterns it displays correspond, for example, to sequences of grids or phase-shifted fringes. Alternatively, a projection device (not shown) separate from the 330 screen is used to display patterns on this screen.

[0107] The measurement system 300 further includes an electronic image processing device 340, connected to the image acquisition device 310 and configured to process the images acquired by this image acquisition device 310, so as to identify the optical distortions induced by the glazing, by implementing the method for determining an optical distortion of glazing according to the invention.

[0108] In a particular implementation method, the optical quality of a portion 321 of the glazing is evaluated. This portion 321 corresponds, for example, to a central area of ​​the glazing 320, which then offers the advantage of eliminating possible edge effects of the glazing.

[0109] Figure 4 represents modules embedded in an electronic image processing device 340, according to a particular implementation method of the invention.

[0110] As illustrated in Figure 4, the 340 electronic image processing device includes, in particular:

[0111] - a M0D_DA module for determining, at each point of a set of points on a glazing surface, a data point representing the angular deflection undergone, at the exit of the glazing, by a light ray passing through said glazing; and,

[0112] - a MOD_TAN1 module for determining, at each point of the set, a tangent value, called "first tangent value", by applying a partial derivative to the data representing an angular deviation.

[0113] Their functionalities are described in more detail below with reference to different implementation methods.

[0114] Figure 5 schematically represents an example of the hardware architecture of an electronic image processing device 340.

[0115] As illustrated in Figure 5, the image processing electronic device 340 has the hardware architecture of a computer. Thus, the image processing electronic device 340 includes, in particular, a processor 1, random access memory 2, read-only memory 3, and non-volatile memory 4. It also has communication means 5.

[0116] The read-only memory 3 of the image processing electronic device 340 constitutes a storage medium according to the invention, readable by the processor 1, on which a computer program PROG according to the invention is stored, comprising instructions for executing steps in the method for determining the optical distortion of a pane of glass according to the invention. The PROG program defines functional modules of the image processing electronic device 340, which rely on or control the hardware elements 1 to 5 of the image processing electronic device 340 mentioned above. These functional modules are illustrated in Figure 4 by way of no limitation and are described in more detail below with reference to different implementation methods.

[0117] In certain implementations, the communication means 5 enable the image processing electronic device 340 to obtain images captured by the image acquisition device 310. To this end, the communication means 5 include a wired or wireless communication interface capable of implementing any suitable communication protocol.

[0118] Figure 6 illustrates, in general terms, the principle of the invention, and the notations adopted in the rest of the description.

[0119] As mentioned previously, the invention aims to determine the optical distortion induced by glazing. This optical distortion is reflected in particular by a deviation of a ray exiting the glazing it passes through, and is measured, at different points of the glazing, by a metric called "tangent".

[0120] As illustrated in Figure 6, the ray represents the path followed by the light between the image capture device 310 and the screen 330, in the absence of the glazing 320 along this path. Thus, this ray does not undergo any deviation and reaches the 330 screen at a refpos point y The ray r2, for its part, represents the path followed by the light between the image capture device 310 and the screen 330 in the presence of the glazing 320 along this path. For this reason, this ray r2 undergoes a deviation and reaches the screen at a point pos y different from the refpos point y Figure 6 illustrates the measurement system from a lateral view defined by a (y, z) plane. However, a similar reasoning applies along the x-axis of the (x, y) coordinate system.

[0121] Also, the horizontal dx and vertical dy components of the spatial deviation (i.e., the deviation in terms of distance on a plane defined by the 330 screen), at each point % and y, are expressed as follows: (dx = pos Y — refpos Y [ L 0122] J , (dy = pos y x - D J spos y x

[0123] The horizontal components a x and vertical a y The angle of optical deviation induced by the 320 glazing is then expressed as follows: dx I a x tan '( L0+ï

[0124] < dy I tan '( a y L0+ï 0

[0125] with l0 the distance between the image acquisition device 310 and the glazing 320, and L0 the distance between the glazing 320 and the screen 330.

[0126] In the case of a so-called "small angle" approximation, that is, when the angle measures are relatively small, then tan -1 (0) ~ 6, and the components a x and a y Optical deviations are then expressed as follows: dx has x = - L °dy° ü v = - y Lo + lo

[0128] The tangent t', also called the "master" tangent, is defined as the derivative of the angular deviation in a certain direction (horizontal along the x-axis or vertical along the y-axis), with respect to the perpendicular direction in the same plane (%), y. More formally, the horizontal components t x and vertical tÿ of this master tangent are then expressed as follows: doty 1 ddy = X dx LQ + IQ dx

[0129] J X da x 1 dd x — - = - X — - dy L0+l0dy

[0130] The tangent t0 "to the observation distance" (e.g., according to a "first geometry") is then expressed as follows:

[0131] t0= (L o + l0) xt r

[0132] and the horizontal components t O x and vertical t Oy The tangents to the observation distance of this tangent can be expressed, for example, as follows: ddy I ^0,x dx

[0133] < dd x I to, y dy

[0134] Thus, the tangent t0 to the observation distance is expressed independently of the geometry of the measurement system, and in particular independently of the distance L o + l0 between the image acquisition device 310 and the screen 330.

[0135] Thus, this solution is advantageous in that it allows for the robust determination of optical distortion induced by glazing.

[0136] Figure 7 illustrates a change in the geometry of the measurement system, according to a particular embodiment of the invention.

[0137] More precisely, this figure illustrates, according to a side view plan of an optical distortion measurement system, a first geometry adopted by this measurement system, and a second geometry that this measurement system should adopt, for example in order to meet the constraints of the ASTM F2156 standard.

[0138] The first geometry of the measurement system is defined by the following characteristics:

[0139] - the image capture device is located at a position 710-0, and at a distance l0 from the position 720 of the glazing;

[0140] - the screen is located at position 730-0, and at a distance L o from position 720 of the glazing.

[0141] The second geometry of the measurement system is defined by the following characteristics:

[0142] - the image capture device is located at position 710-1, and at a distance l from position 720 of the glazing; and

[0143] - the screen is located at a position 730-1, and at a distance L from the position 720 of the glazing.

[0144] Furthermore, in a particular implementation mode, the distance between the image capture device and the screen according to the second geometry is greater than the same distance according to the first geometry, i.e., L o + l0 < L + l.

[0145] Furthermore, as illustrated in Figure 7, the field of view FOV0 of the image acquisition device according to the first geometry is defined by two light rays passing through the glazing at two extrema P ± and P2, and the field of view FOV1 of the image acquisition device according to the second geometry is defined by two light rays passing through the glazing at these same two extrema P ±and P2.

[0146] In other words, to determine the proportionality factor A to be applied to the tangent t0 to the observation distance, we assume that the distance (also called "step") between these extrema remains the same, regardless of the geometry considered.

[0147] The proportionality factor A can then be expressed, for example, as follows:

[0148] A = — x — x-^-. L

[0149] The second value of tangent or target tangent t is expressed as follows:

[0150] t = A x t0

[0151] It should be noted that the proportionality factor A can still admit other formulations. For example, in the case where the distance L o tends towards infinity and the distances lo and 1 are (approximately) identical, this factor A can then be expressed according to the following approximation: [ L 0152] A — x — = 1 + - J LL o L

[0153] Figure 8 represents, in the form of a flowchart, a particular method of implementing a process for determining the optical distortion of a glazing, for example carried out by the electronic image processing device 340.

[0154] As illustrated in Figure 8, the determination process includes a first step S100 in which a series of phase-shifted patterns displayed on the screen 330 is captured by the image acquisition device 330 and then obtained (e.g., received or accessed) by the image processing device 340. Each of these images illustrates an effect of the glazing 320 on a pattern displayed on the screen 330.

[0155] In one particular implementation, these patterns correspond to out-of-phase fringe patterns. Alternatively, these patterns correspond to out-of-phase grid patterns.

[0156] The method further includes a step S200 for determining, at each point of a set of points on a glazing surface, a value representing the angular deflection experienced by a light ray passing through the glazing as it exits the glazing. This step S200 is implemented, for example, by the M0D_DA module of the 340 image processing electronic device. In a particular implementation, the step S200 for determining a value representing the angular deflection includes the steps S210 and S220 described below.

[0157] During step S210, a spatial phase distribution <p "discontinue" (également nommée "carte de phase discontinue") est déterminée à partir des captures obtenues aucours de l'étape S100. Différentes méthodes de démodulation de phase existent qui utilisent une modulation de la phase optique par laquelle un terme de phase supplémentaire est ajouté à la phase dite "utile". Ces méthodes de démodulation de phase se distinguent notamment selon qu'elles sont locales ou globales. Une méthode locale s'attache à calculer la phase optique en un seul point, et ce, en utilisant des valeurs successives du signal en ce point ou en utilisant un ensemble de valeurs d'un groupe de points dans le voisinage du point considéré, tandis qu'une méthode globale calcule la phase optique en tous les points pour lesquels les motifs (de franges) ont été enregistrés. Ces méthodes de démodulation de phase se distinguent également selon qu'elles sont purement temporelles ou spatiales.In the first case, the data needed to evaluate the phase are obtained at different time intervals, while in the second case, all the data are obtained simultaneously.

[0158] This phase distribution Ap is said to be "discontinuous" in the sense that it is calculated over a certain interval, for example [- n; ÎT]. A discontinuity of the phase then manifests itself by the fact that when an extreme value is reached (eg, +n or -n), the phase "jumps" to the other end of the interval (eg, - n or H-ÎT), although the physically real optical phase is continuous and increasing or decreasing in a relatively smooth way.

[0159] To reflect this physical continuity, phase unwrapping is applied to the discontinuous phase map during an S220 step, resulting in a continuous phase map. More precisely, phase unwrapping is implemented, for example, by adding or subtracting a digit ( / c) that is an integer multiple of 2^n, i.e., Ap = 2^kx^n.

[0160] In a particular mode of implementation, the data representing an angular deviation correspond to the horizontal dx and vertical dy components of the spatial deviation (i.e., the deviation in terms of distance on a plane defined by the screen 330).

[0161] The process further includes a step S300 for determining, from the representative angular deviation data determined during step S200, a first tangent value t0 at each point of the set. This step is implemented, for example, by the MOD_TAN1 module of the 340 electronic image processing device.

[0162] To do this, a partial derivative is applied to the data representing an angular deviation. More precisely, a calculation similar to that described with reference to Figure 6 is implemented, for example. In this case, the horizontal components t O x and vertical t O y This first tangent value can be expressed, for example, as follows: { t — _dd L y v o,x tQ 'y dy

[0164] Then, during an S400 step, the proportionality factor A to be applied to the first tangent value t0 so as to obtain a second tangent value t is determined.

[0165] This step is implemented, for example, by the MOD_FCT module of the 340 image processing electronic device. The S400 step is implemented, for example, in a similar manner to the geometry change described with reference to Figure 7. In this particular case, the proportionality factor A is then expressed, for example, as follows:

[0166] A = yxyx, with l0 the distance between the image acquisition device 310 and the glazing 320 according to the first geometry, l the distance between the image acquisition device 310 and the glazing 320 according to the second geometry, L othe distance between the glazing 320 and the screen 330 according to the first geometry and L the distance between the glazing 320 and the screen 330 according to the second geometry.

[0167] The process also includes a step S500 in which the second tangent value is determined at each of the points of the set, by applying the proportionality factor A determined in step S400 to the first tangent values ​​determined in step S300.

[0168] More formally, this S500 step of determining the second value of tangent t is expressed, for example, as follows:

[0169] t = A x t0

[0170] Finally, in a particular implementation method, the determination process includes an S600 step in which the glazing for which the optical distortion was determined in the S500 step is classified.

[0171] To do this, at each point of the set, the second tangent value is compared, for example, to a threshold value representing the expected quality for that glazing. If, for a predetermined number of points (for example, 10% of the points), this threshold value is reached or even exceeded, then the glazing is considered to be of insufficient quality.

[0172] In a particular embodiment, the method further includes data filtering, so as to mimic human visual perception and / or to eliminate unwanted terms, for example, those introduced into the phase and which carry no useful information about the optical deviation induced by the glazing. These terms were introduced, for example, during fringe formation and / or during step S210 of determining the spatial distribution of the "discontinuous" Ap phase.

[0173] Data filtering includes, for example, one or more elements from among:

[0174] - an application of a denoising function to the data representing an angular deviation. This denoising function corresponds, for example, to a median or Gaussian filter;

[0175] - an application of a smoothing function to the first tangent value. This smoothing function corresponds, for example, to a Gaussian filter with a width of 5 m;

[0176] - applying a smoothing function to the first tangent value, then applying a denoising function to the second tangent value;

[0177] - a variable weight smoothing filtering based on the extremum of the data representing an angular deviation;

[0178] - a smoothing filter resulting from a deep learning analysis of local defects in the glazing;

[0179] - a variable weight smoothing filter, depending on areas of the glazing.

[0180] The invention has so far been described in the case where the data representing an angular deviation is determined from images captured by the image acquisition device 310, each image illustrating an effect of the glazing 320 on a pattern displayed on the screen 330. However, the invention remains applicable in the case where the data representing an angular deviation is determined from a thickness at different points of the glazing.

[0181] The thickness at different points of the glazing is then determined, for example, using a profilometer connected to an image processing device.

[0182] As is well known, a profilometer is a measuring instrument used to determine the surface relief of glazing. It is preferably an optical matrix profilometer, which determines the relief from images of the glazing surface. To do this, the profilometer includes (or is connected to) an image acquisition device, such as a camera equipped with a CCD (Charge-Coupled Device) image sensor, commonly used in digital imaging. The use of an optical matrix profilometer offers several advantages, including a faster measurement speed than a contact profilometer. Furthermore, the absence of physical contact with the surface eliminates the risk of damage (i.e., scratching). Finally, the measurement provided by such a profilometer is free from noise caused by any lateral movement during the measurement process.

[0183] It is also important to note that the data representing an angular deviation, according to this implementation method, is determined by applying a first partial derivative to the thicknesses. The data representing an angular deviation then corresponds to a slope or to the angular deviation itself. As mentioned previously, the "first tangent value" is then determined by applying a second partial derivative to the data representing an angular deviation.

[0184] Finally, according to a specific implementation method, the data representing an angular deviation is determined from a local variation in the optical path. For example, the data representing an angular deviation is determined from the thicknesses at different points of the glazing and the refractive index of the media through which a light ray passes.

Claims

Demands

1. A method for determining the optical distortion of glazing (320) implemented by an electronic processing device (340), the method comprising: - a determination (S200), at each point of a set of points on a surface of the glazing, of a data point representing an angular deviation undergone, at the exit of the glazing, by a light ray passing through said glazing; and, - a determination (S300), at each point of the set, of a tangent value, called "first tangent value", by applying a partial derivative to the data representing an angular deviation.

2. Method of determination according to claim 1, the electronic device (340) being an electronic image processing device (340) of a measurement system (300) further comprising an image acquisition device (310) connected to the electronic image processing device (340) and a screen (330), the glazing (320) being positioned between the image acquisition device (310) and the screen (330), and the data representative of an angular deviation is determined from images captured by the image acquisition device (310), each image illustrating an effect of the glazing (320) on a pattern of a series of phase-shifted fringe patterns or a series of phase-shifted grids displayed on the screen (330).

3. A method for determining the angular deviation according to claim 2, wherein the data representing an angular deviation corresponds to a spatial deviation on a plane defined by the screen (330), and the first tangent value t0 is expressed such that d dx '-O,x — g a with x, y the coordinates of a ray incident on said plane, dx and dy the to ' x = ~ horizontal and vertical spatial deviations, and d the partial derivative symbol.

4. A method for determining the angular deviation according to claim 2 or 3, wherein the data representative of an angular deviation is determined by applying a fringe demodulation method including: an acquisition (S100) of the images captured by the image acquisition device (310); - a determination (S210) of a discontinuous phase map from the images obtained; and, - a phase unwinding (S220) applied to the discontinuous phase map, so as to obtain a continuous phase map corresponding to the representative data of an angular deviation determined at each of the points.

5. Method of determination according to claim 1, wherein the data representative of an angular deviation is determined at least from thicknesses at each point of the assembly.

6. A method for determining according to any one of claims 2 to 4, wherein the first tangent value is determined when the measuring system (800) is configured according to a first geometry, the method further comprising: - obtaining (S400) a proportionality factor (A) allowing the calculation of a second tangent value according to a second geometry of the measurement system (800) from the first tangent value; and, - a determination (S500), at each point of the set of points, of the second tangent value according to the second geometry of the measurement system, from the first tangent value and the proportionality factor (A).

7. A method for determining according to claim 6, wherein the proportionality factor A is expressed such that A = f(-x - x) L+l ), with f a function Mathematics, l0 the distance between the image acquisition device (310) and the glazing (320) according to the first geometry, l the distance between the image acquisition device (310) and the glazing (320) according to the second geometry, L o the distance between the glazing (320) and the screen (330) according to the first geometry and L the distance between the glazing (320) and the screen (330) according to the second geometry.

8. A method of determination according to claim 7, wherein a first proportionality factor A is applied to a first zone of the glazing, a second proportionality factor A is applied to a second zone of the glazing, the first and second zones having different geometries.

9. A method of determination according to any one of claims 1 to 8, further comprising data filtering intended to mimic human visual perception.

10. A determination method according to claim 9, wherein the filtering comprises at least one element from: - an application of a denoising function to data representing an angular deviation; - an application of a smoothing function to the first tangent values; - a variable weight smoothing filtering based on the extremum of the data representing an angular deviation; - a smoothing filter resulting from a deep learning analysis of local defects in the glazing; - a variable weight smoothing filter, depending on areas of the glazing.

11. Method for determining according to any one of claims 1 to 10, further comprising a classification (S600) of the glazing (320)

12. A method of determination according to any one of claims 1 to 11, wherein the glazing is intended to be fitted to a vehicle.

13. Electronic processing device (340) configured to implement a determination method according to any one of claims 1 to 12.

14. Computer program comprising instructions for carrying out a determination method according to any one of claims 1 to 12, when said program is executed by a processor.

15. Computer-readable recording medium on which a computer program according to claim 14 is recorded.

Images