Method for detecting optical defects in windshields - Patent Application 20070122999

Abstract

It provides a novel and easy to implement method for detecting certain optical defects that may not be detected by current inspection systems, but may remain visible to the driver's eye. A method (1000) for detecting optical defects in a windshield, taking as input a digital image map of the intensity of the refractive power of the windshield and providing as output a digital image map of optical defects, comprising: (a) an image processing step (1001) of said digital image map of the refractive power to detect and demarcate areas of different intensities of the refractive power; (b) a step (1002) of calculating for each detected area a representative geometric distance and a representative value of the refractive power; and (c) a step (1003) of calculating a representative value of the refractive power such that the product between the representative geometric distance and the representative value of the refractive power is greater than or equal to 2.9×10 -4 and calculating (1003) an image map of the detection area.

Description

[Technical Field]

[0001] The present invention relates to a computer-implemented method for detecting optical defects in a windshield. [Background technology]

[0002] Windshields are well known in the transportation industry, for example, in automobiles, rail transportation, and aircraft. Windshields are typically made of two curved sheets laminated together with a polymer interlayer.

[0003] A windshield is the glazing through which a driver views what lies ahead, such as the road, rails, scenery, etc. For safety reasons, therefore, distortion of objects visible through the windshield must be as low as possible, or at least it must not be so great as to confuse the driver. In this regard, the optical quality of the windshield must meet certain requirements detailed in paragraph 9.2 of Annex 3 to Regulation No. 43 of the United Nations Economic Commission for Europe (UN / ECE).

[0004] In the art, several methods and instruments are described which can be intended to measure the optical distortion of windshields within the framework of Regulation No. 43 of the Agreement.

[0005] EP 0 463 940 A1 describes a process for measuring the optical quality of a windshield based on shadow illumination.

[0006] WO 98 / 17993, GB 2152210 and EP 1061357 describe methods for measuring optical distortions in windscreens through image analysis of transmission or reflection patterns.

[0007] WO 2017 / 008159 describes a method for detecting optical defects in windshields through analysis of a composite image of chromatic aberration. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] European Patent Application Publication No. 0463940 [Patent Document 2] International Publication No. 98 / 17993 [Patent Document 3] British Patent No. 2152210 [Patent Document 4] European Patent Application Publication No. 1061357 [Patent Document 5] International Publication No. 2017 / 008159 Summary of the Invention [Problem to be solved by the invention]

[0009] Although the methods described in the art are efficient, they may fail to detect some optical defects that are detectable by the human eye. Human vision is generally more flexible than most inspection systems that implement the method, and sometimes at certain angles or different angles, optical defects that are only visible on the driver's and / or passenger's side of the windshield may be completely missed by inspection systems on the production line. A direct negative consequence is that windshields that were initially deemed to meet technical specifications may subsequently be rejected by customers. Complaints and production losses may occur.

[0010] Furthermore, it has been found that some of these missed or undetected optical defects may not have the same features or characteristic properties, and thus the inspection system may behave differently with respect to their detection. In some particular setups, some inspection systems may be able to detect these defects, while others may not. Therefore, regardless of the characteristics of the inspection system, it is difficult to find a setup and / or criteria that will cause the inspection system to detect these defects. Furthermore, even if such a setup or standard could be found, it would still need to remain compatible with the overall configuration of the manufacturing line.

[0011] Therefore, a need exists for a novel, easy-to-implement method for detecting these particular optical defects that may not be detected by current inspection systems, but may remain visible to the eye of a human driver. [Means for solving the problem]

[0012] In a first aspect of the present invention there is provided a computer implemented method for detecting optical defects in a windshield as set out in claim 1, the dependent claims being advantageous embodiments.

[0013] In a second aspect of the invention, a data processing system, a computer program and a computer readable medium for implementing the method are provided.

[0014] In a third aspect of the invention there is provided a process for detecting optical defects in a windscreen as claimed in claim 10, the independent claims being advantageous embodiments.

[0015] Both the method and the process may be used in the manufacturing process of windshields.

[0016] A first outstanding benefit of the present invention is that it enables the detection of optical defects in windshields that may remain undetected by most conventional inspection systems, but may still be visible to the human eye, e.g., the driver's eye.

[0017] A second advantage is that the present invention is relatively easy to implement in existing manufacturing processes, thus requiring little, if any, adaptation. More precisely, the present computer-implemented method invention and processes according to the present invention can benefit from facilities for acquiring digital image maps of refractive power that are already available in the manufacturing line and / or quality control processes.

[0018] A third advantage is that the present invention indirectly provides the driver with insight into the relative size at which optical defects may appear when viewing distant objects through the windshield. It is then possible to assess whether optical defects may affect visibility. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a data flow diagram of a computer-implemented method for detecting optical defects according to an embodiment of the first aspect of the present invention; [Figure 2] 1 is an example of a digital image map of the refractive power intensity of a windshield in grayscale. [Figure 3] 1 is a data processing system according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0020] Referring to FIG. 1, in one embodiment of the first aspect of the present invention, a computer-implemented method 1000 for detecting optical defects in a windshield includes: 1. A method of taking a digital image map of the refractive power intensity of a windshield as input I1001 and providing a digital image map of optical imperfections as output O1001, the method comprising: (a) a step 1001 of image processing of said digital image map of refractive power to detect and demarcate areas of different strengths of refractive power; (b) calculating 1002 representative geometric dimensions and representative values ​​of refractive power for each detection area; (c) The product between the representative geometric distance and the representative refractive power is 2.9 × 10 -4 a step 1003 of calculating an image map of the detected area; A method is provided that includes:

[0021] The method 1000 takes as input I1001 a digital image map of the refractive power intensity of a windshield, an example of which is provided in FIG. Refractive power refers to the ability of a lens to focus light. Depending on how much the lens refracts the light, the light can diverge or converge. The SI unit of refractive power is the diopter (dpt), which is the reciprocal of the meter (m -1 )

[0022] In FIG. 2, the refractive power intensity is represented by a grayscale pattern across the windshield, with the white boxed areas described below.

[0023] The measurement of refractive power in windscreens is well described in the art, for example in EP 3012619, EP 0463940, EP 0342127, EP 0685733. Alternatively, or as a complement to the measurements, it is also possible to simulate the refractive power, as in EP 3756114 A1. All of these methods may directly provide a digital image map of the refractive power of the windshield or may be adapted to obtain this digital image map.

[0024] In some embodiments, the digital image map may be acquired at a specific angle or for a range of given angles relative to the windshield normal. In a preferred embodiment, the digital image map may be acquired at angles or within a range of angles corresponding to or representative of the angles at which a driver may view objects through the windshield in use. In fact, since the tilt angle of the windshield relative to the vehicle framework typically varies from vehicle to vehicle, the driver's viewing angle, and subsequently the probability of seeing an optical defect, may also vary depending on this tilt angle. It may therefore be advantageous to take this effect into account when acquiring a digital map of refractive power. The accuracy and reliability of the method can be improved.

[0025] The image processing 1001 of step (a) for detecting and demarcating regions of the digital image map with different intensities of refractive power can be any adapted image processing method for object detection. The image processing method may be a neural network or a non-neural network method.

[0026] In a preferred embodiment, image processing 1001 may be blob detection, in particular blob detection through calculation of the Laplace operator of the Gaussian distribution of the digital image map of refractive power, the difference of the Gaussian distribution of the digital image map of refractive power, or the determinant of the Hessian of the digital image map of refractive power. Blob detection can be easier to implement than neural network methods, yet provides valuable results for most types of windshields and applications.

[0027] In step (b), the representative geometric dimensions of the detection regions may depend on the shapes selected to represent their boundaries in step (a). However, as a general rule of thumb, to be representative, the calculated values ​​for the geometric dimensions should advantageously show little variation regardless of the shape selected, provided that the shape is suitable for delimiting the detection area.

[0028] Many shapes may be suitable and may be more or less complex, depending on the image processing methods used to detect and demarcate the regions and the proximity and accuracy that can be expected for the region boundaries. For example, these shapes may be convex shapes such as convex polygons, eg squares or rectangles, or curvilinear figures, eg circles or ellipses, or concave shapes, eg concave polygons or concave curvilinear figures.

[0029] Since optical defects within the scope of the present invention are often regular and may have a relatively rounded or elongated convex shape, in some advantageous embodiments the detection area in step (a) may be bounded by an ellipse, and the representative geometric distance of the detection area calculated in step (b) is the minor axis of said ellipse.

[0030] In step (b), the calculated representative geometric dimensions of each detection area have units of length, e.g., meters in SI units, and the calculated representative values ​​of refractive power have units of, e.g., m -1 Or has units of the reciprocal of length, such as dpt.

[0031] The representative value of the refractive power of each detection region can be calculated through different methods. In some embodiments, the representative value of the refractive power of each detection region is the mean refractive power, the median refractive power, the maximum refractive power, or the maximum / minimum refractive power difference within said detection region.

[0032] The product value calculated in step (c) is a dimensionless number. In some embodiments, this number may be considered to represent an angle that may be expressed in radians (rad). In this regard, the 2.9 × 10 provided in step (c) -4 The minimum value of may then be taken to represent the minimum distortion coefficient at which optical defects can be visible to the human eye, i.e., the driver, when looking through the windshield at an object positioned at a given distance D from the windshield. This 2.9 x 10 expressed in radians -4 The product of the minimum value of ≡ ...

[0033] In step (c), the product between the representative geometric distance and the representative refractive power is 2.9 × 10 -4 For such detection regions, an image map of the detection region is calculated. In this case, 2.9 × 10 -4 can be considered as the lower boundary of the semi-infinite interval. Since any optical defect can have virtually any size larger than what the human eye can perceive, any attempt to define a high boundary value can be seen as purely artificial and arbitrary. In practice, a higher boundary value may be defined depending on the desired requirement for greater optical distortion.

[0034] In some embodiments, the value of the product calculated in step (c) is 5×10 -4 ~2×10 -3 , preferably 7 x 10 -4 ~1.5×10 -3 and more preferably 1×10 -3 It may be. The interval between these values ​​is 2.9×10 -4 is greater than the minimum value of Therefore, these are 2.9 × 10 -4 This may not seem as strict as the minimum value of However, they may be adapted to the most common configurations of windscreens, particularly with respect to the most common vehicle frameworks.

[0035] The method is such that the product between the representative geometric distance and the representative value of refractive power is 2.9×10 -4 The image map of the detection area thus obtained is provided as output O1001. An example of such an image map is provided in Figure 2, where the white boxed area represents the detection area and is superimposed on a digital image map of refractive power.

[0036] In some embodiments, the detection region in step (b) may further be such that its apparent size on the decimal visual acuity scale is between 0.5 and 3; preferably between 0.67 and 1.25, more preferably 1. Visual acuity within the range of human vision is well known in the art and is fully described within the EN ISO 8596:2018 standard. In some applications where high optical quality is required, i.e. aviation, motorsport, premium car applications, the criterion of apparent size can be an advantageous supplement. Such standards can ensure that optical defects to which the human eye may be sensitive can be detected with a high degree of reliability.

[0037] The method 1000 according to the first aspect of the invention may be advantageously used in the manufacturing process of windscreens. Because the manufacturing process may already include equipment for acquiring a digital map of refractive power, few, if any, adaptations to the process may be required to implement the method.

[0038] In a second aspect of the present invention, with reference to Figure 3, there is provided a data processing system 3000 including means for performing the method 1000 according to any one of the embodiments of the first aspect of the present invention, and a computer program I3001 including instructions which, when executed by a computer, cause the computer to perform the method according to any one of the embodiments of the first aspect of the present invention.

[0039] The data processing system 3000 comprises means 3001 for performing the method according to any of the embodiments of the first and second aspects of the present invention. An example of a means for performing a method 3001 may be a device that can be instructed to automatically perform a sequence of arithmetic or logical operations to perform a task or action. Such devices, also referred to as computers, may include one or more central processing units (CPUs) and at least one controller device adapted to perform these operations. It may further include other electronic components such as an input / output interface 3002, a non-volatile or volatile storage device 3003, and a bus, which is a communication system for transferring data between components within a computer or between computers. One of the input / output devices may be a user interface for human-machine interaction, for example a graphical user interface for displaying information that can be understood by humans.

[0040] Because calculations require a large amount of computing power to process substantial amounts of data, the data processing system may advantageously include one or more graphics processing units (GPUs) with a parallel structure that makes them more efficient than CPUs for image processing, particularly in ray tracing.

[0041] With regard to the computer program I3001, any kind of programming language, whether compiled or interpreted, can be used to implement the steps of the method of the present invention. The computer program I3001 may be part of a software solution, for example a collection of executable instructions, code, scripts etc. and / or a database.

[0042] In some embodiments, there is also provided a computer readable storage device or medium 3003 comprising instructions that, when executed by a computer, cause the computer to perform a method according to any of the embodiments of the first aspect of the present invention.

[0043] The computer readable storage device 3003 may preferably be a non-volatile storage device or memory, such as a hard disk drive or solid state drive. The computer readable storage device may be a removable or non-removable storage medium that is part of the computer.

[0044] Alternatively, the computer readable storage device may be a volatile memory within a removable medium.

[0045] The computer-readable storage device 3003 may be part of a computer used as a server that can download executable instructions that, when executed by the computer, cause the computer to perform a method according to any of the embodiments described herein.

[0046] Alternatively, program I3001 may be implemented in a distributed computing environment, such as cloud computing. The instructions are executable on a server to which client computers can connect and provide encoded data as input to the method. Once the data has been processed, the output can be downloaded and decoded on the client computer or sent directly, for example as instructions. This type of implementation can be advantageous as it can be realized within a distributed computing environment, such as a cloud computing solution.

[0047] In one embodiment of the third aspect of the present invention, in a process for detecting optical defects in a windshield, the method comprises: (a) Acquisition of a digital image map of the refractive power intensity of the windshield; (b) processing the digital image map with a computing system to detect and demarcate areas of different refractive power intensities; (c) The product between the representative geometric distance and the representative refractive power is 2.9 × 10 -4 calculating, using a computing system, the image map of the detection area; A process is provided which includes:

[0048] The technical aspects and features of the different embodiments detailed with respect to the first and second aspects of the invention may also be applied to the third aspect of the invention. It is within the ability of a person skilled in the art to modify, convert or adapt them in the process according to the third aspect of the invention. [Explanation of symbols]

[0049] 1000 Computer Implementation Methods 1001 image processing steps 1002 Steps for calculating geometric dimensions and representative values ​​of refractive power 1003 Image map calculation step for the detected area 3000 Data Processing System 3001 Means for implementing 3002 Input / Output Interface 3003 Storage device

Claims

1. 1. A computer-implemented method for detecting optical defects in a windshield, the method taking as input a digital image map of the refractive power intensity of the windshield and providing as output a digital image map of the optical defects, the method comprising: (a) image processing of said digital image map of refractive power to detect and demarcate areas of different strengths of refractive power; (b) calculating representative geometric dimensions and representative values ​​of refractive power for each detection area; (c) The product between the representative geometric dimension and the representative refractive power is 2.9×10 －４ calculating an image map of the detection area; A method comprising:

2. The product in step (c) is 5×10 －４ ~2 x 10 －３ , preferably 7 × 10 －４ ~1.5 x 10 －３ , more preferably 1×10 －３ The method of claim 1, wherein

3. 3. The method according to claim 1 or 2, wherein the image processing of step (a) is blob detection, in particular blob detection through calculation of the Laplace operator of the Gaussian distribution of the digital image map of refractive power, the difference of the Gaussian distribution of the digital image map of refractive power, or the determinant of the Hessian of the digital image map of refractive power.

4. 3. The method according to claim 1 or 2, wherein in step (b), representative geometric dimensions and representative values ​​of refractive power are calculated for a detection area whose apparent size on the decimal visual acuity scale is between 0.5 and 3, preferably between 0.67 and 1.25, more preferably 1.

5. 3. The method of claim 1, wherein the detection area in step (a) is bounded by an ellipse, and the representative geometric dimension of the detection area calculated in step (b) is the minor axis of the ellipse.

6. 3. The method according to claim 1, wherein the representative value of the refractive power of each detection area is the mean refractive power, the median refractive power, the maximum refractive power, or the maximum / minimum refractive power difference within the detection area.

7. A data processing system comprising means for carrying out the method according to claim 1 or 2.

8. A computer program comprising instructions that cause a computer to carry out the method of claim 1 or 2 when the program is executed by a computer.

9. A computer-readable medium containing instructions that, when executed by a computer, cause the computer to perform the method of claim 1 or 2.

10. 1. A method for detecting optical defects in a windshield, the method comprising: (a) obtaining a digital image map of the refractive power intensity of the windshield; (b) processing the digital image map with a computing system to detect and demarcate areas of different refractive power intensities; (c) The product between the representative geometric dimension and the representative refractive power is 2.9×10 －４ calculating, using a computing system, the image map of the detection area; A method comprising:

11. The product in step (c) is 5×10 －４ ~2 x 10 －３ , preferably 7 × 10 －４ ~1.5 x 10 －３ , more preferably 1×10 －３ The method of claim 10, wherein

12. 3. Use of the method according to claim 1 or 2 in a windscreen manufacturing process.

Images