Rupture Analysis Method for Analyzing Cracks in a Vehicle Glazed Panel

Description

1 / 31 Rupture Analysis Method for Analyzing Cracks in a Vehicle Glazed Panel Divided from BR 11 2018 072977-9, deposited on 05 / 11 / 2017

[0001] INSTRUMENT AND METHOD OF ANALYSIS OF BREAKDOWN FIELD

[0002] This invention relates, in general terms, to an apparatus and a method. Particularly, but not exclusively, the invention relates to an apparatus and a method to be used for analyzing ruptures in vehicle glass panels. More particularly, but not exclusively, the invention relates to a method and an apparatus to be used for analyzing cracks in glass, in particular in a vehicle glass panel. FUNDAMENTALS

[0003] When driving, the presence of debris and other materials on the road can cause such materials to be diverted into the traffic path where, when such material collides with a windshield, it can cause cracks, ruptures or other damage to such windshields that may require repairs or replacement of the vehicle's glass panel.

[0004] For safety and economic reasons, it is imperative that such work be carried out as quickly as possible, since these cracks can spread further into the windshield due to the influence of cold weather, for example, which can cause the crack to grow. This can cause the crack to change from one that needs minor repair to one that requires the replacement of the entire windshield.

[0005] It is necessary to assess whether the damage to a vehicle's glass panel can be remedied by repair. If the assessment indicates that repair is not feasible, then the glass panel will need to be replaced. Petition 870230057435, dated 06 / 30 / 2023, page 15 / 57 2 / 31

[0006] Aspects and modalities were conceived with the above in mind. SUMMARY

[0007] Viewed from a first aspect, a rupture analysis method for analyzing ruptures in a vehicle glass panel is provided, the method comprising: capturing an image of a rupture in a vehicle glass panel; processing the image of the rupture.

[0008] Viewed from a second aspect, a rupture analysis apparatus for analyzing ruptures in a vehicle glass panel is provided, the apparatus comprising: a camera configured to capture an image of a rupture in a vehicle glass panel; an operating processing module for processing the image of the rupture.

[0009] Optionally, the device may include a mobile computing device, which includes the camera. A mobile computing device is an electronic device that is configured to capture images. This may include a mobile phone (e.g., a smartphone), a laptop computer, a tablet, a phablet, or a camera. The mobile computing device includes a camera to capture images.

[0010] The mobile computing device may also include the processing module.

[0011] A method or apparatus according to the first and second aspect allows an image of a rupture on a surface to be used to analyze the rupture. This eliminates the need for physical attendance by someone skilled in the art to perform any analysis on the crack.

[0012] Image processing can be used to determine if the glazed panel needs replacing. Thus, the method may include the step of determining if the glazed panel needs Petition 870230057435, dated 06 / 30 / 2023, page 16 / 57 3 / 31 or not be replaced, and / or whether the glazed panel is suitable for repair, based on image processing.

[0013] The image of the rupture can be captured at an angle relative to the vehicle's glass panel.

[0014] The image can be captured by a mobile computing device, held in contact with the surface of the glazed panel, wherein the mobile computing device includes a camera. The mobile computing device can, for example, be a mobile phone, such as a smartphone equipped with a camera. The invention can be implemented through a software component for processing image data from the camera in order to determine whether the break can be repaired or whether replacement is necessary. The software component can be downloaded to the device as, for example, a standalone application or similar. The software component may comprise an algorithm for performing the determination and also, preferably, display instructions to the user on how to implement the determination method.

[0015] According to another aspect, the invention comprises a computer-implemented method for determining technical information regarding a crack present in a vehicle's glass panel, wherein the method includes the step of downloading a software component to a mobile computing device that has a camera, the software component configured to process the image information captured through the camera in order to determine whether the crack can be repaired or whether replacement is preferable.

[0016] The mobile computing device can be tilted to view the rupture at a predetermined position within the camera's field of view. A datum position can be achieved by ensuring that the computing device is tilted toward the panel with one edge in contact with the panel. Petition 870230057435, dated 06 / 30 / 2023, page 17 / 57 4 / 31

[0017] The predetermined position in which the mobile computing device can be tilted to view the break can be indicated by cues displayed on the camera or on the mobile computing device comprising the camera (e.g., on a screen).

[0018] Additionally or alternatively, the clues may indicate the center of the field of vision (for example, on a screen).

[0019] Optionally, the camera or mobile computing device may initially be placed flat on the surface of the glazed panel. The mobile computing device may then be articulated or rotated away from the glazed panel, wherein at least a portion of the mobile computing device remains in contact with the glazed panel.

[0020] Optionally, the camera is positioned at a predetermined position relative to the break before the mobile computing device is pivoted. For example, the method may include aligning a feature of the mobile computing device at a predetermined position relative to the break.

[0021] In some embodiments, an edge of the image capture module, such as the top edge, may be aligned adjacent to (or immediately below) the lowest point of the break. This means that the initial distance between the camera and the lowest point of the break can be determined using the geometry of the mobile computing device.

[0022] The mobile computing device or image capture module can then be pivoted away from the glazed panel, wherein at least one lower edge of the mobile computing device remains in contact with the glazed panel. The image of the break is then captured.

[0023] The method may involve pivoting or rotating the mobile computing device until the break is visible at a predetermined position in a camera's field of view. Petition 870230057435, dated 06 / 30 / 2023, page 18 / 57 5 / 31

[0024] Optionally, the method includes rotating the mobile computing device until the break is in the center of the camera's field of view.

[0025] The method may include the use of geometric parameters of the mobile computing device and camera lens parameters to estimate the rupture parameters. The rupture parameters include one or more spatial dimensions indicative of the rupture size.

[0026] For example, using the method above, the geometric parameters of the mobile computing device and the camera lens parameters can be used to determine the length of one or more legs of the rupture and / or the size (e.g., width / diameter) of the rupture center.

[0027] Geometric parameters of the mobile computing device can be used to determine the rotation angle (or pivot angle) between the mobile computing device and the glazed panel.

[0028] If the estimated size of the crack (e.g., length of one of the crack legs) exceeds a given threshold, then the method can determine that the glazed panel needs to be replaced. If the estimated size of the crack (e.g., length of one of the crack legs) is less than a given threshold, then the method can determine that the glazed panel is suitable for repair.

[0029] Image processing can be based on geometric parameters of the mobile computing device to capture the image of the rupture.

[0030] Image processing can also be based on chip parameters for the camera, and / or mobile computing device.

[0031] Image processing can generate a set of data points that are used to generate a diameter for the rupture. Petition 870230057435, dated 06 / 30 / 2023, page 19 / 57 6 / 31 which can be used to determine the need for replacement for a vehicle's glass panel.

[0032] The method may include outputting a signal or indication showing whether the glazed panel needs to be repaired or replaced.

[0033] A crack in a glazed panel may comprise a center and one or more legs. This crack formation is common when a small stone or other object impacts a glazed panel. One or more legs (or cracks) usually radiate from the center of the crack.

[0034] The center of the rupture may be substantially circular in shape.

[0035] Determining the need to replace or repair a glazed panel may involve generating data indicative of a fracture center and one or more fracture legs.

[0036] The method may include identifying a rupture center and one or more rupture legs.

[0037] The method may include generating a scaling factor indicating the relative length of one or more legs of the rupture compared to the size of the rupture center. The size of the rupture center may be the diameter, width, and / or length of the rupture center.

[0038] The method may include estimating the length of one or more legs of the rupture (i.e., the absolute length in cm or mm, etc.) using the scale factor. For example, the length of one or more legs can be estimated by multiplying the scale factor by a predetermined value.

[0039] The predetermined value can be an estimate of the actual (i.e., absolute) size of the rupture center. This offers the advantage that no calibration object is required, which is at least more convenient for the user. Petition 870230057435, dated 06 / 30 / 2023, page 20 / 57 7 / 31

[0040] Thus, the method of the present invention may comprise determining the scale of the image to estimate the length of one or more legs of the rupture.

[0041] If the estimated length of one or more legs exceeds a given threshold, then the method can determine that the glazed panel needs to be replaced. If the estimated length of one or more legs is less than a given threshold, then the method can determine that the break in the glazed panel is appropriate for repair.

[0042] The method may include emitting a signal indicating that repair of the glazed panel is necessary if the estimated length of one or more legs is less than the determined threshold.

[0043] The method may include emitting a signal indicating that replacement of the glazed panel is necessary if the estimated length of one or more legs exceeds the determined threshold.

[0044] It has been found that the size of the center of the rupture generally varies less than the length of one or more of the rupture legs when comparing different ruptures in glazed panels. As such, the predetermined value may be an average, or mode of the measured sizes of the center of a rupture in a glazed panel.

[0045] Optionally, the estimate of the actual width (or diameter) of the fracture center (i.e., the predetermined value) may be between 1 mm and 3 mm. A particularly preferred predetermined width (or diameter) of the fracture center may be 2 mm. These ranges / values ​​were determined from fracture studies carried out by the applicant.

[0046] The effect of this is that the estimated size of the rupture center, as well as the estimated diameter of the rupture center, can be used to estimate the length of the rupture legs, since the predetermined value can be used as a scale between the relative lengths. Petition 870230057435, dated 06 / 30 / 2023, page 21 / 57 8 / 31 of one or more legs in relation to the size of the rupture center and the estimated actual length of one or more rupture legs.

[0047] For example, if we know that the center of the rupture will always be around 2mm wide (or in diameter) and the generated image data indicates that the legs are twice the length of the diameter of the center of the rupture, then the method can understand multiplying 2mm by a scaling factor of 2. This estimates that the legs are 4mm long. This helps to build an image, in the data, of the dimensions of the rupture.

[0048] The generated (or estimated) length of one or more legs can be used to indicate the estimated size of the break. The size of the break can be compared to a threshold parameter to determine the need for replacement or repair of the glazed panel.

[0049] If the estimated size of the crack exceeds a given threshold, then the method can determine that the glazed panel needs to be replaced. If the estimated size of the crack is less than a determined threshold, then the method can determine that the glazed panel is suitable for repair.

[0050] The comparison can be between the rupture threshold parameter and the greatest distance across the entire rupture.

[0051] Optionally, the predetermined estimate of the rupture center size may depend on one or more parameters. The parameters may be user-entered and / or pre-configured in the processing device or module. For example, parameters may include: one or more properties of the glass panel (such as type, size, etc.), and / or the speed at which the vehicle was traveling when the rupture occurred.

[0052] Image processing may involve filtering the image to remove a portion of the background to identify the break. Petition 870230057435, dated 06 / 30 / 2023, page 22 / 57 9 / 31

[0053] Morphological refinement can be applied to the image to remove any image dirt and improve the quality of the image data used as the basis for determining whether a replacement glazed panel is needed.

[0054] The method may include cleaning the glass panel before capturing an image of the break. This can help in removing any dirt that may affect image processing. For example, there is a risk that dirt may be interpreted as a break by image processing software.

[0055] The method may involve disabling a flash function on the image capture module or device before capturing an image of the break. If flash is used in the photograph, the light may adversely affect the accuracy of the image processing software. For example, the flash may be reflected by the glass panel, which may affect the identification or analysis of the break.

[0056] The method can be implemented using computer-implemented instructions that, when installed in memory, instruct a processor to implement a method as defined above. A downloadable software component (such as an application) is preferred.

[0057] It will be appreciated that any features of the method can be performed using the apparatus of the present invention.

[0058] These and other aspects of the present invention will become apparent and elucidated with reference to the embodiment described in this document. DESCRIPTION

[0059] The first and second embodiments of the present invention will now be described, by way of example only, and with reference to the attached figures, in which:

[0060] Figure 1 illustrates a windshield with a crack; Petition 870230057435, dated 06 / 30 / 2023, page 23 / 57 10 / 31

[0061] Figure 2 illustrates the use of a camera to capture an image of a crack;

[0062] Figure 3 illustrates a processing module that can be used to analyze the crack in the windshield in Figure 1;

[0063] Figure 4 illustrates a flowchart detailing the steps involved in evaluating a crack in the windshield using the system in Figure 3;

[0064] Figure 5 illustrates an image of a transform of Fourier transform of a crack; and Figure 6 illustrates an outlined image of a crack.

[0065] Figure 7 illustrates an arrangement that can be used to model a crack 102;

[0066] Figure 8 illustrates the steps involved in generating rupture parameters;

[0067] Figure 9A schematically illustrates the field of view of a camera with a bird's-eye view from a windshield;

[0068] Figure 9B schematically illustrates the field of view of a camera with an angled view of a windshield;

[0069] Figure 10 illustrates how a camera can be modeled to use the camera parameters to analyze the crack; and

[0070] Figure 11 illustrates an image that is output by the system to determine the size of a crack in a windshield.

[0071] In a first embodiment, Figure 1 illustrates a glass windshield 100 with a crack 102 caused by a stone thrown at the windshield. The driver of the car in which the glass windshield 100 is housed then captures an image of the crack 102 using a cell phone 104, which comprises a camera 106 that is used to capture the image of the crack 102. This arrangement is shown from a side perspective in Figure 2. The focal length Petition 870230057435, dated 06 / 30 / 2023, page 24 / 57 The 11 / 31 inch of the 106 camera is fixed at less than 100 millimeters to allow the camera's focus to be fine-tuned at a short distance.

[0072] The image of the crack 102 is then captured, in response to user input, and the mobile phone 104 is configured to provide a request to the user to indicate whether they would like the image to be transmitted from camera 106 to an image processing module 108 which we will now describe with reference to figure 3. This step allows the user to assess the image quality so that they can choose to capture another image if they do not consider the image clear, such as, for example, in inclement weather conditions in which condensation may be deposited on the lens of camera 106.

[0073] Camera 106 converts the captured image into an image data matrix using any suitable method. Camera 106 can save the image data as an interchangeable image file (EXIF) in which the camera's lens parameters are also stored.

[0074] That is, camera 106 is an example of an image capture module that is operable to capture the image of crack 102 and to transmit the captured image in the form of captured image data to the image processing module 108, where it can be further processed to extract details of the crack.

[0075] The image processing module 108 may be part of the mobile phone 104 or may be geographically distant from the mobile phone 104. The image data is transmitted to the image processing module 108 by any appropriate means, such as a data bus or the internet.

[0076] In a preferred embodiment, the image processing module 108 is implemented as a software component downloaded to the mobile phone. This can be implemented by means Petition 870230057435, dated 06 / 30 / 2023, page 25 / 57 12 / 31 download of the software component as an application. The software component is capable of processing the camera image data to determine if the break is suitable for repair or if replacement of the glass panel may be required as the preferred solution.

[0077] The rupture analysis module 112 may be a software component downloaded to the mobile phone, preferably as a single download in combination with the image processing module 108. A single downloaded software component is preferably arranged to process the image data from the phone's camera and analyze the rupture using one or more algorithms implemented in software.

[0078] In one embodiment, the captured image data is received by the image processing module 108 at the data input interface 110. The image data is then transmitted to the rupture analysis module 112, which is configured to access a routine library 114 where routines can be stored to perform operations on the captured data throughout the analysis of the captured image data.

[0079] The disruption analysis module 112 is also configured to access a device parameter database 116 in which the parameters related to the mobile phone 104 are stored.

[0080] The parameters related to the mobile phone 104 include chip parameters that define the image capture performance of the camera 106, such as, for example, the focal length and the lens sensor size, and the dimensional parameters of the mobile phone 104, such as, for example, the length of the mobile phone 104 and the distance between the top edge of the mobile phone 104 and the center of the camera image 106. Petition 870230057435, dated 06 / 30 / 2023, page 26 / 57 13 / 31

[0081] The rupture analysis module 112 is also operable to interface with a display module 118 which is operable to display image data transmitted from the rupture analysis module 112 on a display and to display parametric data transmitted from the rupture analysis module 112 on a screen.

[0082] Now we will describe, with reference to figure 4, the crack analysis 102 using the rupture analysis module 112.

[0083] The rupture analysis module 112 receives the image data in an S400 step. The rupture analysis module 112 then, in an S402 step, calls a Fourier transform routine from the routine library 114 and uses the Fourier transform routine to apply a discrete two-dimensional Fourier transform to the image data to produce a transformed image as illustrated in Figure 5a.

[0084] In Figure 5a we can see the transformed image. Figure 5a graphically represents the spatial frequency against the magnitude of the respective Fourier component. It can be observed that low spatial frequencies occupy the center of the transformed image, and higher spatial frequencies can be seen as we move away from the center of the transformed image.

[0085] Performing a Fourier transform on the image allows the rupture analysis module 112 to perform image analysis in terms of its spatial frequencies and component phase. As will now be described, this allows the removal of spatial frequencies in which we are not interested and allows us to reconstruct the image we are interested in, while maintaining the spatial frequencies of interest.

[0086] A Butterworth bandpass filter is then applied to the transformed image in an S404 step by the 112 break analysis module. The mask implemented by the Butterworth bandpass filter is Petition 870230057435, dated 06 / 30 / 2023, page 27 / 57 14 / 31 illustrated in Figure 5b. The Butterworth bandpass filter implements a mask on the transformed image shown in Figure 5a and removes low spatial frequencies (shown by the black dot in the center of Figure 5b) and very high spatial frequencies (the dark edge in the image in Figure 5b) that represent dust and dirt spots in the image.

[0087] The Fourier transform of the image data is then inverted in an S406 step by calling a Fourier transform inversion routine from the routine library 114 to perform an inverse two-dimensional Fourier transform on the transformed image data.

[0088] The inverse Fourier transform performance on transformed image data transforms the Fourier domain transformed image data into real domain data to generate real domain image data. The resulting real domain image data are illustrated in Figures 6a and 6b.

[0089] The use of the Fourier transform to produce the image illustrated in figure 6B has the effect of isolating the crack from the background.

[0090] Real-domain image data are compared to an intensity threshold of 4, in an S408 step, to delineate areas of interest more clearly.

[0091] Figure 6a shows the real-domain image data without using a Butterworth bandpass filter. Figure 6b shows the real-domain image data after applying the Butterworth bandpass filter to the transformed data illustrated in Figure 5a and applying a threshold to generate a binary image using a threshold intensity of 4. The Butterworth bandpass filter in this example has a roll-off value of 3.

[0092] The upper and lower cutoff frequencies of the Butterworth bandpass filter can be modeled as being linearly Petition 870230057435, dated 06 / 30 / 2023, page 28 / 57 15 / 31 dependent on the number of pixels on the longer side of the image (indicated as m) and can be expressed respectively as:

[0093] This relationship can be altered using standard tests and numerical experiments.

[0094] The image illustrated in figure 6b is an image that may include more than just crack 102. It may also include image data that passed through steps s400 to S408 but are due to dirt particles on the windshield and other artifacts from the processing performed by the crack analysis module 112.

[0095] The use of threshold intensity 4 to generate the binary image shown in Figure 6b helps to show the areas of interest more clearly. The image shown by the real-domain image data illustrated in Figure 6b highlights the crack – including the central crack area – which is an area of ​​low spatial frequency.

[0096] As can be seen, the Fourier approach does a fairly adequate job of isolating the crack region from the polluted background, assuming that the region is in focus, and the background is not.

[0097] The rupture analysis module 112 can then call a morphology routine from the routine library 114 to remove any pollution from the image illustrated in figure 6b in an S410 step.

[0098] The morphological routine performs several operations on the image illustrated in Figure 6b. This image is a binary image. Black regions have a value of zero, and white regions are evaluated as non-zero. The pixel values ​​are stored in the module's memory. Petition 870230057435, dated 06 / 30 / 2023, page 29 / 57 16 / 31 of interrupt analysis 112 and is the result of processing in steps S400 to S408.

[0099] The first of these operations is a fill operation that uses morphological reconstruction to fill pixel-sized black regions that are surrounded by white regions with white, by replacing the zero value with a non-zero value, according to the set of processes determined in [1]

[00100] The second of these operations is a cleaning operation that discards very small non-zero value regions. Very small non-zero value regions are defined as non-zero value regions that occupy an area smaller than the square of (largest image dimension / 500). The largest image dimension can be determined by the break analysis module simply by comparing the length of the image width with the length of the image.

[00101] The first morphological operation is then repeated to fill any pixel-sized black regions that are surrounded by white regions that were generated by the second morphological operation. This is a third morphological operation.

[00102] A fourth morphological operation is then performed to join all legs in the crack image 102 that have gaps. This is implemented using a morphological closure operation as described in [2]. An erosion is performed followed by a dilation, making use of a disc-shaped structuring element with a radius of (largest image dimension\5312) multiplied by 20. The value of 20 was determined empirically and may change. This value can be determined without any undue burden for different image resolutions.

[00103] The first morphological operation is then repeated to fill any pixel-sized black regions that are Petition 870230057435, dated 06 / 30 / 2023, page 30 / 57 17 / 31 surrounded by white regions that were generated by the fourth morphological operation. This is the fifth morphological operation.

[00104] A sixth morphological operation is then performed to discard any non-zero small regions. Small regions are defined as regions with an area equal to the square of (largest image dimension / 100).

[00105] A seventh morphological operation is then performed to remove any disconnected objects in the image. Disconnected objects of interest are objects that are more than % of the radius of the largest object closest to the center of the image. This means that the legs of the crack that are still disarticulated are included, but superfluous artifacts are also included. The seventh morphological operation is implemented by finding, for each remaining region in the image, the centroid, i.e., the center of mass of the image, and the length of the region's principal axis. An additional weighting is assigned to each region area based on the proximity of the centroid to the center of the image.

[00106] The weighting w = 1 / d2 where d is the Euclidean distance between the centroid and the center of the image. The largest region, closest to the center of the image, is selected, and the length of its principal axis is used to define a radius (or % of the principal axis length from its centroid) outside which all regions are discarded. That is, the morphology routine and the centroid for boundary calculations are configured to retain all spots at a distance from the crack center equal to the radius of the largest object in the image, plus half the radius to ensure that any discontinuities in the crack are not missed.

[00107] The image data, after morphology is applied to refine the image data, can then be used to determine the dimensions of the crack 102. Petition 870230057435, dated 06 / 30 / 2023, page 31 / 57 18 / 31

[00108] The rupture analysis module 112 additionally applies edge detection, morphology, blurring and threshold to determine the crack center 102.

[00109] It has been observed through experimentation that the center of a crack is generally about 2 mm in diameter. The fracture analysis module 112 is operable to, using refined image data and data resulting from the determination of the crack center 102, estimate the length of the crack legs 102 and determine a proportional value, which characterizes the length of the legs in comparison to the diameter of the crack center 102, which is a scale factor for the legs in relation to the crack center 102. Using the observation that the crack center is generally 2 mm, the scale factor can then be used to determine the length of the legs. This provides an uncalibrated analysis of the size of a crack 102.

[00110] The determined length of the legs can then be used to approximate the size of the crack 102 and allows the rupture analysis module 112 to issue a determination as to whether windshield replacement is necessary or if a repair will suffice, because it is the size of the crack that is important in making this determination and by comparing the size of the crack 102 to a repair / replacement threshold, the rupture analysis module 112 can automate this determination. The rupture analysis module 112 will issue this determination to the display module 118

[00111] This issue, i.e. whether or not a replacement windscreen will be required, is then displayed using display module 118 in an S412 step.

[00112] Using an observed estimate of the center of a crack to estimate the size of the crack legs, which depends on the assumption of a radial peak degree in a crack, means that an image can be taken of a crack and used Petition 870230057435, dated 06 / 30 / 2023, page 32 / 57 19 / 31 to analyze the crack without any calibration on the scene to provide a scale for the crack 102.

[00113] This method allows a crack analysis to be performed under a wide range of conditions and without the presence of a technician.

[00114] In a second embodiment, we describe how to derive crack parameters 102 using mobile phone parameters 104 and camera lens 106. This can help in correcting any influence that the angle may have on the image.

[00115] The second method can be combined with the first method without deviating from this disclosure.

[00116] The arrangement illustrated in Figure 2 can allow the dimensions of the crack to be estimated using the camera chip parameters 106 and the mobile phone geometric parameters 104.

[00117] To calculate the rotation angle (or pivot angle) of mobile phone 104 relative to the windshield, we can use the geometric parameters of mobile phone 104.

[00118] By positioning crack 102 in the center of the field of view of camera lens 106, this allows a right-angled triangle to be drawn. This is described with respect to figure 7.

[00119] After crack 102 has been discovered, mobile phone 104 is placed horizontally against the windshield with its top edge at the base of crack 102. This means that the distance between the bottom edge of mobile phone 104 and the base of the crack is equal to the length of mobile phone 104. Mobile phone 104 is then tilted from its bottom edge until crack 102 is in the center of the field of view of camera 106. Indicators may be provided on the screen of mobile phone 104 to indicate the center of the field of view. Petition 870230057435, dated 06 / 30 / 2023, page 33 / 57 20 / 31

[00120] The distance between the bottom edge of the mobile phone 104 and the camera lens 106 can be retrieved from the device parameter database 116. There is therefore a right-angled triangle formed, defined by the rotation angle between the bottom edge of the mobile phone 104 and the windscreen 100, the z-axis of the camera lens, and the distance formed between the bottom edge and the base of the crack.

[00121] We now describe how the geometric parameters of the mobile phone 104 and the lens parameters can be used to estimate the crack parameters.

[00122] An image of the crack is captured consistently with the process described above, in which mobile phone 104 is rotated until the crack 102 is in the center of the camera's field of view 106

[00123] This allows a right-angled triangle to be formed by the z-axis of the camera lens, the distance formed between the bottom edge and the base of the crack, and the length between the bottom edge and the camera lens.

[00124] With reference to figure 8, we now describe how the geometry of the mobile phone 104 and the lens parameters can be used to estimate the rupture parameters.

[00125] In step S800, the crack analysis module 112 retrieves the distance formed between the bottom edge of the mobile phone 104 and the base of the crack (i.e., the length of the mobile phone 104) and the length between the bottom edge of the mobile phone 104 and the camera lens from the device parameter database 116. The rotation angle of the mobile phone 104 can then be calculated in step S802 using the cosine relationship between the distance formed between the bottom edge and the base of the crack and the length between the bottom edge and the camera lens. Petition 870230057435, dated 06 / 30 / 2023, page 34 / 57 21 / 31

[00126] Next, we need to use the camera parameters to derive plane-to-plane homographic mapping between the camera pixels and the real-world spatial dimensions of the image. A plane-to-plane homographic mapping routine is then called from the 114 routine library in an S804 step to derive the real-world spatial dimensions of the image.

[00127] The derivation of the homographic map to provide the real-world spatial dimensions of the image is based on the pinhole camera model, in which a camera views the camera's field of view as a cone with a rectangular base expanding outward relative to the camera lens 106. This is illustrated schematically in figures 9a and 9b.

[00128] Figure 9a is for illustration purposes only and illustrates the case where mobile phone 104 is located directly above the windshield 100. That is, camera 106 provides a panoramic view of the windshield 100. In this example, the display area A1 is a rectangle and each pixel occupies the same amount of space as the real world (in millimeters).

[00129] In this case, and as illustrated in figure 9b, the cell phone 104 is at an angle 100 to the windshield. The angle was calculated in step S802. The viewing area A2 then becomes a trapezoid, meaning that pixels closer to the camera represent fewer millimeters of pixels that are further away.

[00130] We describe the theoretical basis of how the plane-to-plane homographic map is derived, but it will be understood that this will be implemented numerically using routines that will be available to the rupture analysis module 112 using the routines provided by the routine library 114.

[00131] Consider a rectangular image sensor that is part of camera 106 and an angle sensor of a plane flattened by a rotation angle 0, the area observed by the sensor maps to an isosceles trapezoid. The width of the bases of this trapezoid is directly dependent Petition 870230057435, dated 06 / 30 / 2023, page 35 / 57 22 / 31 on 9. Using the plane-to-plane homographic mapping routine, we can use this principle to numerically estimate the crack parameters 102 using knowledge of the pixels in the camera 106.

[00132] We define a 3D rotation matrix, about the x-axis, as a function of 0, as: T o cos Θ sen β —sen COS Θ

[00133] It will be understood that 0 is the angle of the mobile phone 104 relative to the windscreen. We can define an origin in Cartesian dimensions x, y, z at (0, 0, 0), that is, the world origin. This is the point in the middle of the base-edge of the mobile phone 104 that is aligned with the x-axis. The y-axis of this coordinate system is then directed vertically from the base to the top of the phone. If we assume, for simplicity and without loss of generality, that the camera is on the y-axis, at some distance dc from the base of the phone. The center of the camera is therefore defined as: C = ( ¾ Cy . 1¾);. = (0,4,, 0 iT.

[00134] The focal length and vertical and horizontal sensor sizes of the camera lens 106 can then be retrieved from the device parameter database 116 in an S806 step. These parameters can be called chip parameters. This allows us to calculate the camera's field of view. The field of view is defined by two quantities which are called the horizontal and vertical angle of view. Petition 870230057435, dated 06 / 30 / 2023, pp. 36 / 57 23 / 31 (respectively indicated as om and av) and are defined by the following equations:

[00135] Where sx and sy are the horizontal and vertical sensor sizes and ef is the focal length.

[00136] Having calculated the horizontal and vertical viewing angle, the break analysis module 112 uses the plane-to-plane homographic mapping routine to calculate the edges of the view pyramid to give us a field of view over the windscreen 100 in one S808 step. This gives the trapezoid illustrated schematically in Figure 9b, i.e., the trapezoid that we need to correct to compensate for the different amounts of space occupied by pixels further from the lens compared to pixels closer to the lens. That is, we need to scale the trapezoid to ensure that the process calculations assign equal amounts of real-world space to each pixel.

[00137] This is modeled in the plane-to-plane homographic mapping routine used by the crack analysis module 112 by a line, i.e., a ray, extending from the lens along the line of sight between the lens and the crack 102. This line will intersect the plane represented by the windshield - i.e., the plane-to-plane homographic mapping routine is modeled as a plane.

[00138] In step S810, the plane-to-plane homographic mapping routine calls a numerical solver routine from the routine library 114 to solve the simultaneous equations that define the windshield plane and the line extending from the lens along the line of sight between the lens and the crack 102. The homographic mapping routine Petition 870230057435, dated 06 / 30 / 2023, pp. 37 / 57 24 / 31 plane-by-plane is programmed under the assumption that the plane defining the windshield 100 is flattened and the camera 106 is rotated relative to it. This provides the intersection between the line extending from the lens along the line of sight and the windshield plane 100.

[00139] Theoretically, this can be expressed as the calculation of the rays emanating from the point in the center of the camera through the corners of the sensor / image-plane and to the windscreen that forms the trapezoid mentioned above.

[00140] First, we obtain the intersection of the rays with a plane, parallel to the image plane, at unit distance, given horizontal and vertical viewing angles of OIH and av, respectively, as defined above.

[00141] There are four rays, one for each corner of the rectangular sensor. The minimum and maximum values ​​of x can be defined as: . . .. \ ~~} ) ... mrà: — Lí 111· i

[00142] Similarly, we can define the minimum and maximum values ​​of y as: Pmin ~ )

[00143] We can then define the corners of the rectangular sensor as: Petition 870230057435, dated 06 / 30 / 2023, pp. 38 / 57 25 / 31 3m / '—·' 1 / X&r — V^wi# ? / / mm: >1)XM —”1 / ·

[00144] Normalizing these coordinates by their magnitude gives us the direction of the 4 rays. We define the ray direction for each of these coordinates as: Ilxdl

[00145] If we assume that the phone is rotated, on the x-axis, by 0, we can calculate that the position of the camera-center is now: c = (4,4. ¾):. - nr** · (o, rnr

[00146] This allows us to define the direction of the rays as: xj^ = Ar(W · i € (tLbr br, bl}.

[00147] This gives us the rays in Cartesian coordinates with a known point of intersection with the plane parallel to the image plane, and we know that this intersection occurs only once. This provides a trapezoid that indicates the field of view in the real world.

[00148] We define the corners of the trapezoid as: € (éL tr. br. W)

[00149] We calculated the vertices of the trapezoid using the line plane intersection formula described in [3] Petition 870230057435, dated 06 / 30 / 2023, pp. 39 / 57 26 / 31

[00150] We know that the normal to the windshield plane is the vector n = (0, 0, -l) and that it is located at the origin of the world, which means that the intersection formula simplifies to: —Gr t — --------. η ox^' \'i = ê 4- tx? °, i E (th tr. br. bl). V;. i Ç (tl. ír. br, bli

[00151] Where the points define the vertices of the trapezoid that we need to define the homographic mapping H from the image plane to the plane in the real world using the four-point matching technique between the trapezoid vertices and the image coordinates: Utf = (0.0)5(h. w)T= (0. w) ',

[00152] Where w is the image width and h is the image height. The algorithm that discusses how this homographic map is obtained is discussed in [4].

[00153] The camera height above the windshield can be calculated by the break analysis module 112 using the Pythagorean theorem, since the distance formed between the bottom edge of the mobile phone 104 and the base of the crack (i.e., the length of the mobile phone 104) and the length between the bottom edge of the mobile phone 104 and the camera lens were retrieved from the device parameter database 116 in step S800 and are still in the memory of the break analysis module 112. Petition 870230057435, dated 06 / 30 / 2023, pp. 40 / 57 27 / 31

[00154] The output of step S810 is the real-world view trapezoid (Xi, X2, X3, X4). A comparison between the parameters (Xi, X2, X3, X4) and the corners of the image captured on the windshield (performed by the break analysis module 112 in step S812) provides the scale needed to map the location of the camera pixels 106 to locations in millimeters in the windshield field of view 100. This gives us the plane-to-plane homographic map. The scale is in the form of a 3x3 matrix representing the scale, rotation, twist, and translation between the camera and windshield field of view 100.

[00155] The plane-to-plane homographic map allows correction of the perspective effect on the captured image and conversion of pixel dimensions to millimeters, which allows the rupture analysis module 112 to derive dimensional parameters for the crack 102.

[00156] The plane-to-plane homographic map is a matrix that maps the two-dimensional image plane of camera 106 to a plane representing the windshield.

[00157] The issuance of the planar-planar homographic map provides an orthorectified mask, in millimeters, indicating the position and shape of the crack.

[00158] Responding to this plane-by-plane homographic map output, which will be, as will be understood, the output of the plane-by-plane homographic mapping routine, the rupture analysis module 112 calls a convex envelope calculation routine from the routine library 114. The locations in millimeters in the windshield field of view are provided to the convex envelope calculation routine from the routine library 114.

[00159] A convex surface is, in short, a space that encompasses each of the millimeter-sized locations in the field of view. The result of the convex envelope calculation routine is data that can be expressed, in simple terms, as a spot that will be the same size. Petition 870230057435, dated 06 / 30 / 2023, pp. 41 / 57 28 / 31 size that crack 102 detected. This allows the analysis to be performed on crack 102 detected using the stain.

[00160] The rupture analysis module 112 then calls a smallest circle routine from the routine library 114 that implements a numerical solution to the smallest circle problem for the convex envelope that is the result of the convex envelope calculation routine. This module outputs the smallest circle that encompasses each of the points in the convex envelope and thus provides a minimum radius for the crack 102.

[00161] The data representing the convex envelope, the data representing the solution to the smallest circle problem for the convex envelope, and the calculated radius for the crack are stored by the stored rupture analysis module 112, which is local with respect to processing module 108 or remote with respect to processing module 108.

[00162] In other words, the rupture analysis module 112 used the geometric parameters of the mobile phone 104 and the parameters of the camera 106 to generate a radius for the crack 102.

[00163] The parameters and circle output of the smaller circle routine can then be displayed using display module 118 in an S814 step.

[00164] An example image that can be provided by display module 118 is illustrated in Figure 11. In this case, the diameter of the smaller circle is indicated as 16 mm, which means a radius of 8 mm. The largest estimated crack diameter in this case is 16 mm. The effect here is that a minimum size for the crack is estimated and can be used to determine the need to replace a windshield.

[00165] The estimated radius can be compared to a replacement / repair threshold by the rupture analysis module 112 to determine whether crack 102 requires replacement or whether repair will suffice. Petition 870230057435, dated 06 / 30 / 2023, pp. 42 / 57 29 / 31

[00166] The presence of a case on mobile phone 104 may introduce an error in the measured parameters, as it will add length to mobile phone 104, but the error is generally around 3%. A 3% margin of error is incorporated into the calculations of the break analysis module 112 and provided on the screen by the display module 118.

[00167] It is also possible that the distance between the mobile phone base 104 and the camera 106 will not be available from the device parameter database 116. In this case, we can estimate the parameter to improve the robustness of the described method.

[00168] We can use an inclinometer built into the mobile phone 104 to obtain the angle of the mobile phone when the image of the crack 102 is being captured. This can be used to calculate the height h using the equation:

[00169] Where 1 is the length of the 104 mobile phone and angle 0 is the angle obtained from the inclinometer.

[00170] Similarly, the phone's angle can be estimated using the viewing angle, image resolution, and sensor size.

[00171] As described in relation to the first technique, in a preferred embodiment, the image processing module 108 is implemented as a software component downloaded to the mobile phone. This can be implemented by downloading the software component as an application. The software component is capable of processing the camera's image data to determine whether the break is suitable for repair or whether replacement of the glass panel may be required as the preferred solution.

[00172] The 112 disruption analysis module can be a software component downloaded to the mobile phone preferably as Petition 870230057435, dated 06 / 30 / 2023, pp. 43 / 57 30 / 31 a single download in combination with the image processing module 108. A single downloaded software component is preferably arranged to process the phone's camera image data and analyze the breakup using one or more algorithms implemented in software.

[00173] The crack analysis module 112 is operable to provide an on-screen alert about whether a complete windshield replacement is necessary based on the radius of the smallest circle. If the crack 102 is above a specified threshold, then the crack analysis module 112 will indicate whether or not a windshield replacement is needed. The alert can be displayed on a mobile phone screen 104.

[00174] It should be noted that the embodiments mentioned above illustrate, rather than limit, the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the invention as defined by the appended claims. In the claims, any references placed in parentheses should not be interpreted as limiting the claims. The words comprising, includes, and the like do not exclude the presence of elements or steps other than those listed in any claim or in the descriptive report as a whole. In the present descriptive report, includes means includes or consists of, and comprising means including or consisting of. The singular reference to an element does not exclude the plural reference to such elements and vice versa. The invention can be implemented by means of hardware comprising several distinct elements and by means of a suitably programmed computer.In a device claim with multiple meanings, several of those meanings may be modalized by one and the same hardware item. The mere fact that certain measurements are cited in dependent claims. Petition 870230057435, dated 06 / 30 / 2023, pp. 44 / 57 31 / 31 mutually different does not indicate that a combination of these measures cannot be used advantageously. References [1] http: / / www.mathworks.com / tagteam / 64199_91822v00_eddins_final.pdf [2] https: / / en.wikipedia.org / wiki / closing (morphology) [3] Joseph O'Rourke “Computational Geometry in C”, Cambridge University Press, 2012 [4] Richard Hartley and Andrew Zisserman “Multiple View Geometry in Computer Vision”, Cambridge University Press, 2011 Petition 870230057435, dated 06 / 30 / 2023, pages 45 / 57

Claims

1 / 3 CLAIMS 1. A rupture analysis method for analyzing ruptures in a vehicle glazed panel (100), the method comprising: capturing an image of a rupture (102) in a vehicle glazed panel; processing the image of the rupture; wherein the image is captured by a mobile computing device (104) having a camera (106); characterized in that the mobile computing device (104) is held in contact with the surface of the glazed panel and tilted towards the surface of the glazed panel (100) to view the rupture (102) at a predetermined position in a camera field of view while the computing device is held in contact with the surface of the glazed panel.

2. Method according to claim 1, characterized in that it further comprises: determining whether the glazed panel needs to be replaced or repaired based on image processing.

3. Method, according to claim 1 or 2, characterized in that the image of the rupture is captured at an inclined angle relative to the vehicle's glazed panel.

4. A method, according to any one of claims 1 to 3, characterized in that the predetermined position is indicated by cues displayed on the mobile computing device.

5. A method, according to any one of claims 1 to 4, characterized in that it comprises: positioning the mobile computing device flat on the surface of the glazed panel; and pivoting the mobile computing device away from the glazed panel, wherein at least part of the mobile computing device remains in contact with the glazed panel. Petition 870230057435, dated 06 / 30 / 2023, pp. 46 / 57 2 / 3 6. Method according to claim 5, characterized in that the camera is positioned in a predetermined position relative to the break before the mobile computing device is pivoted.

7. A method according to claim 6, characterized in that it comprises aligning a feature of the mobile computing device in a predetermined position relative to the break, before the mobile computing device is pivoted.

8. A method according to claim 7, characterized in that it comprises aligning an edge of the mobile computing device with the lowest point of the break before pivoting the mobile computing device.

9. A method, according to any one of claims 5 to 8, characterized in that it comprises pivoting the mobile computing device until the break is visible at a predetermined position in a camera's field of view, preferably where the predetermined position is the center of the camera's field of view.

10. A method, according to any one of claims 5 to 9, characterized in that it includes the use of geometric parameters of the mobile computing device and camera lens parameters to estimate the size of the break, preferably in which the geometric parameters of the mobile computing device are used to determine the pivot angle between the mobile computing device and the glazed panel.

11. Method, according to claim 10, characterized in that the image processing is also based on chip parameters for the camera, and / or mobile computing device.

12. Method, according to claim 10 or 11, characterized in that the image processing generates a set of data points that are used to generate a diameter for the rupture or a zone within the rupture.

Images