WO2021136934A1

WO2021136934A1 - Improvements in and relating to detection and mapping of cracks

Info

Publication number: WO2021136934A1
Application number: PCT/GB2020/053379
Authority: WO
Inventors: Saber KHAYATZADEH
Original assignee: Khayatzadeh Saber
Priority date: 2020-01-03
Filing date: 2020-12-30
Publication date: 2021-07-08
Also published as: GB2590731B; GB202000072D0; GB2590731A

Abstract

A method for preparing a surface crack profile, comprising of detecting a first crack in a surface of a component, measuring a first set of crack characteristics of the first crack, deriving a surface model of the first crack and deriving a three-dimensional crack profile based on the surface model and at least a first set of data.

Description

IMPROVEMENTS IN AND RELATING TO DETECTION AND MAPPING OF

CRACKS

Field of the invention

This invention relates to detection of cracks in structures and particularly, but not exclusively, detection and three-dimensional mapping of such cracks.

Background to the invention

Maintenance of structures, such as wind turbines or other free-standing structures, or components thereof can be costly and labour intensive. Typically, wind turbines (as well as other, similar structures) are inspected manually by a person. Since the surfaces of such structures can often be inconvenient or difficult for a person to access and to move about on, manual inspection can be extremely time- consuming. In recent years, drones and certain types of remotely operated robots or vehicles have been employed to carry out inspections. However, the development and deployment of such drones or vehicles is very limited at present, in part due to lack of development in the field as well as high operating costs.

During inspections, the structure typically has to be non-operational to prevent injury to the inspecting person or vehicle. Since inspections can be time consuming using present inspection techniques and equipment (e.g., the aforementioned manual inspection), the structure or component may be offline for a significant period of time. This reduces productivity and efficiency of the structure. For example, in the case of wind turbines, such inspections reduce valuable energy production time.

Furthermore, existing inspection techniques rely only surface detection and mapping. However, the depths or three-dimensional structure of a crack may, in some cases, be even more significant than the surface opening of the crack.

The inventors have appreciated the shortcomings of existing solutions and techniques. Summary of the invention

In accordance with a first aspect of the invention, there is provided a method for preparing a surface crack profile, comprising: detecting a first crack in a surface of a component; measuring a first set of crack characteristics of the first crack; deriving a surface model of the first crack; and deriving a three-dimensional crack profile based on the surface model and at least a first set of data.

The step of deriving a three-dimensional crack profile may comprise: segmenting the surface model of the first crack into a plurality of segments; deriving a predicted first crack profile parameter for each of said plurality of segments based on the at least first set of data; and combining the surface model and the predicted first crack profile parameter into a three-dimensional crack profile.

The predicted first crack profile parameter may comprise a first crack depth. The predicted first crack profile parameter may comprise a first directional crack depth.

The first crack profile parameter may be derived based on a set of material parameters. The set of material parameters may comprise a set of mechanical testing results, and the step of deriving may comprise deriving the predicted first crack profile parameter based on a correlation between the set of mechanical testing results and a derived local stiffness parameter for the component material.

The first crack profile parameter may be derived based on a finite element analysis.

The first crack profile parameter may be derived by a first predictive algorithm based on at least one of: a set of material parameters; a result of a first finite element analysis; or at least a first set of predetermined material characteristics.

The first crack profile parameter may be derived based on two or more of: a set of material parameters; a finite element analysis; or a first predictive algorithm based on at least one of: a set of material parameters, a result of a first finite element analysis, or at least a first set of predetermined material characteristics.

The step of detecting may comprise obtaining at least one image of the surface. The step of detecting a first crack on a surface may comprise: obtaining a plurality of images of the surface; and stitching the plurality of obtained images together to form a combined surface image. The at least one image of the surface may be projected onto a substantially planar surface. The at least one image of the surface may be projected onto a three-dimensional surface.

The step of measuring a set of characteristics may comprise: determining a first crack face boundary; determining a first crack length; and determining a first crack mouth.

Determining a first crack length may comprise: detecting a first crack corner; detecting a second crack corner; and defining the crack length as a distance between the first crack corner and the second crack corner. The step may further comprise: partitioning the first crack into a plurality of partitions if it is determined that at least a portion of the first crack length is positioned outside the first crack boundaries. The step of partitioning may comprise: determining that the first crack length comprises at least a portion that is located substantially outside the first crack boundaries; detecting at least one first crack feature; dividing the first crack into a set of first crack partitions based on the at least one detected first crack feature; and a set of first crack partition lengths is generated based on the derived set of first crack partitions.

In accordance with a second aspect of the invention, there is provided a computer system operable to carry out the method as set out above.

The computer system may comprise: a first device operable to carry out at least a first portion of the method as set out above; and a second device operable to carry out at least the remaining of the method as set out above. In accordance with a third aspect of the invention, there is provided a method for a user device, the method comprising: detecting a first crack in a surface of a component; transmitting data relating to the detected first crack to a remote device, the remote device being operable to at least derive a surface model of the first crack and derive a three-dimensional crack profile based on the surface model and at least a first set of data; and receiving a three-dimensional crack profile from the remote device.

The method may further comprise measuring a first set of crack characteristics of the first crack.

The method may further comprise displaying a representation of the three- dimensional crack profile to the user.

In accordance with a fourth aspect of the invention, there is provided a user device operable to carry out the method as set out above.

In accordance with a fifth aspect of the invention, there is provided a method for a computing device, the method comprising: receiving data relating to a detected first crack from a user device; deriving a surface model of the first crack; deriving a three-dimensional rack profile based on the surface model and at least a first set of data; and transmitting the three-dimensional crack profile to the user device.

In accordance with a sixth aspect of the invention, there is provided a computing device operable to carry out the method as set out above. Brief description of the drawings

An embodiment of the invention will now be described, by way of example, with reference to the drawings, in which:

Figure 1 illustrates an exemplary method in accordance with an aspect of the invention;

Figure 2 schematically illustrates the method of Figure 1 ;

Figure 3 shows an exemplary derivation step of the method illustrated in Figure 1 ;

Figure 4 schematically shows the exemplary step of Figure 3;

Figure 5 shows schematically an exemplary first crack profile parameter as may be derived by the exemplary methods;

Figure 6 illustrates a further exemplary method in accordance with an aspect of the invention;

Figure 7 shows an exemplary method for partitioning a first crack;

Figure 8 shows schematically the exemplary method of Figure 8;

Figures 9 and 10 illustrate exemplary systems in which the exemplary methods may be implemented; and

Figures 11 and 12 illustrate exemplary methods in accordance with an aspect of the invention.

A first exemplary embodiment of the invention will now be described with reference to Figure 1 and Figure 2.

In a first step 101 , a first crack 202 is detected in a surface 204 of a component. The surface may be that of a component or structure under test (such as, without limitation, a pylon for a wind turbine, a transmission tower for RF signals, or any other structure) The first crack may be detected in any suitable fashion using a suitable detection unit, component or apparatus (not shown). The detection unit may use any suitable methodology, means or technology to detect cracks. In some examples, the detection unit receives electromagnetic radiation (e.g., radiation substantially within the visible spectrum, infrared radiation or ultraviolet radiation). In some examples, the detection unit emits and receives electromagnetic radiation. In other examples, the detection unit emits and receives acoustic signals (e.g., ultrasound).

In a particular example, the detection unit is a camera operating substantially within the visual portion of the electromagnetic spectrum. Specifically, the detection unit is a camera located in a mobile device (such as a mobile phone, phablet, tablet, laptop computer or other portable computing device). The ubiquity of mobile devices means that a person inspecting a surface is likely to have a suitable detection unit to hand if cracks are detected. Furthermore, this obviates the need for the inspecting person or vehicle to carry or use specialised or expensive equipment. In other examples, the detection unit is a camera unit mounted on a vehicle (e.g., a remotely operated vehicle). This allows inspections to be conducted without the presence of a human operator at the surface being imaged. In some examples, the detection unit is a camera unit in contact with a mobile device but located remotely therefrom. In a specific example, the detection unit is a camera unit that is wirelessly connected with a user’s mobile device.

The detection unit may obtain any suitable number of data sets or any suitable amount of data. Typically, a detection unit obtains one or more frames 206 of the targeted area or areas. For example, a camera conventionally obtains either individual frames, a predetermined plurality of frames (e.g., ‘double shot’ or ‘multi shot’ cameras) or sequences of frames (e.g., video or time-lapse recordings).

In some instances, it may be necessary or advantageous to obtain a number of frames and to conjoin (or ‘stitch’), these frames into a combined frame or image. This could be because the dimensions of the crack are too large to suitably fit within a single frame in sufficient detail. Conjoining a plurality of frames may also be necessary in circumstances where the geometry of the surface or component in which the crack has formed does not permit imaging the entirety of the crack in a single image. For example, the crack could extend across a curved or angled surface of the component. In an example, the step of detecting a first crack on a surface comprises obtaining a plurality of frames of the surface, and stitching the plurality of obtained images together to form a combined surface image. The plurality of frames may be obtained by a single camera unit, for example as a sequence of frames, each frame showing a different portion of the first crack. In the above example, wherein the camera is located in mobile device, the user may move the camera around so as to capture the entirety of the first crack. The plurality of frames may also be obtained by a plurality of detection units. In some examples, a plurality of detection units each obtain one or more frames, which are subsequently transmitted in a relevant fashion to a central device where the frames are conjoined or stitched together. In some examples, additional information or data is obtained that may be advantageous or useful in the conjoining or stitching process.

In the above examples, the frames have been obtained by one or more cameras operating substantially within the visible portion of the electromagnetic spectrum. However, one or more of the detection devices could, in principle, equally well operate in another portion of the electromagnetic spectrum. For example, a portion of the frames could be obtained by a first camera operating substantially within the visual spectrum, and a second portion of the frames could be obtained by a second camera operating within the ‘infrared’ spectrum. This may necessitate one or more operations or conversion procedures being carried out so as to be able to meaningfully stitch together the component frames that are comprised in the combined frame. This may result in additional processing resource being required in order to form the combined frame. However, by using different types of frames (e.g., a plurality of frames, each obtained by using radiation in a different portion of the electromagnetic spectrum), the overall frame quality may be improved. For example, some features of the first crack may be recordable by using radiation from a certain portion of the electromagnetic spectrum. As a specific example, some features of the first crack may be detectable only by using ‘infrared’ radiation, whereas other features may only be detectable by using ‘ultraviolet’ radiation. Hence, by forming a combined frame based on different portions of the electromagnetic spectrum, an increased accuracy or level of detail of the first crack may be obtained. This, in turn, may increase the speed or accuracy of subsequent method steps.

Any suitable methodology or algorithm for conjoining or stitching together the plurality of images may be used or employed. Typically, one or several of a number of matching or stitching methods will be employed. It will be appreciated that any number of existing methodologies could be employed and implemented within the scope of the present disclosure. Examples include, without limitation, direct matching methods or feature matching methods.

In a second step 102, a first set of crack characteristics of the first crack is measured. The first set may be measured in any suitable manner and using a suitable measuring technology, means or algorithm. It will be appreciated that a number of technologies, means or algorithms could be envisaged within the scope of the present disclosure.

The first set of crack characteristics may comprise any suitable number or type(s) of characteristics. In an example, the first set of crack characteristics comprises one or more of: a first crack length 208; a first crack mouth 210; and a first crack face boundary 212a, 212b.

For substantially linear cracks (i.e. , cracks that predominantly extend in a single linear direction and that do not have any significant bends or other such features), the first crack length is defined as the distance along a line between a first end 214 (also referred to as a “crack corner”) of the first crack and a second end 216 (or crack corner) of the first crack. In some examples, a particular crack can be defined as ‘linear’ if the line 218 defining the first crack length is located entirely between the crack boundaries.

It should be noted that, dependent on the specific shape of the first crack, the first crack length may or may not be representative of the actual length of the first crack. However, for purposes of the present method, the actual length of the first crack is not necessarily required, as will be explained in more detail in the following.

The above approximation of a linear crack does not, as will no doubt be appreciated, encompass all types of cracks. Some cracks may have a more complicated shape, e.g., featuring one or more bends, curves, kinks or other features. For cracks in which the line defining the first crack length falls outside the crack boundary for at least a portion of its length, the first crack length can be determined in a manner which will be discussed in more detail in the following.

The first crack mouth is, in the present example, defined as the width of the first crack at its widest point. The width is determined in a direction perpendicular to the first crack length. The exact position of the first crack mouth is dependent on the properties of the cracked material. In some instances, a crack will be widest near the middle thereof. It will be appreciated that the present definition of the first crack mouth is exemplary only, and that other specific definitions and implementations thereof may be envisaged within the scope of the present disclosure.

The first crack boundary defines the outline of the crack opening on the surface. Typically, the first crack boundary can be subdivided into two opposing boundary edges. The opposing boundary edges meet at the first end of the crack and at the second end of the crack (i.e. , at the two opposing “crack corners”).

The exact shape of the first crack boundary may be dependent on a number of factors, including properties or characteristics of the material(s) of the component or volume in which the first crack is located.

The measurement step can be carried out by any suitable device, entity or component thereof. In some examples, the measurement step is carried out by the user’s mobile device. In other examples, the measurement step is performed by a device, or component of a device, located remotely from the user’s mobile device. In some specific examples, the measurement step comprises a transmission sub-step in which the measurement results are transmitted from the measuring component to the user’s mobile device.

In a third step 103, a surface model 220 of the first crack is derived. The surface model of the first crack may be derived based on any suitable amount of data or obtained characteristics. In essence, the surface model of the first crack is a digital facsimile of the first crack 202 in the surface 204 of the component. The derivation step may involve a number of operations or processes to detect the edges of the crack based on the one or more frames obtained of the crack.

It will be appreciated that a number of methodologies exist for detecting and deriving edges of an object (such as the edges of the first crack) within a frame. A number of such methodologies that could be used as part of the first step may easily be envisaged within the scope of the present disclosure.

In some examples, one or more processing operations may be carried out on the surface model of the first crack, whether prior to and/or subsequently to the derivation step. This may involve operations to reduce excess noise or digitisation effects. This may be necessary to take into account digital artefacts or errors that result from an edge detection algorithm or process.

The derivation step may also involve additional image analysis or processing steps. In some examples, the derived crack boundary may contain one or more artefacts or other unwanted effects. In some instances, the edge detection algorithm may result in the modelled boundary edge containing deviations from the actual crack boundary. The derivation step may accordingly contain one or more error detection algorithms intended to detect apparent deviations between the actual crack boundary and the modelled crack boundary. In some instances, this may involve a user prompt or interaction, whereas in other instances it may be automatically detected and corrected.

The derivation step may be carried by any suitable entity, component or unit. In an example, the derivation step is performed by a component of a user’s mobile device. In other examples, the derivation step is carried out by a device located remotely from the user’s device (e.g., a remotely located computer, server, ‘cloud’ device or other processing device). In such examples, the derivation step comprises a transmission sub-step, in which the measurement data is transmitted to the remotely located device. Furthermore, the derivation step comprises a retransmission step, in which the results of the derivation step are retransmitted to the user’s device. It will be appreciated that a number of implementations of such data transmission sub-steps may be envisaged within the scope of the present disclosure.

The surface model of the first crack may comprise data relating to the geometry of the surface of the component. In some examples, the surface is assumed to be a substantially planar surface. Typically, this may be a default choice if no other information is available or immediately derivable from the obtained frames or images. In other examples, the surface is assumed to be a non-planar surface. In particular examples, the surface is (without limitation) assumed to be one of a cylindrical, spherical or polygonal shape. In some examples, the derivation step comprises a surface geometry detection sub-step in which the shape or geometry of the surface is detected or derived based on the obtained frames or images. In such examples, the surface of the component is analysed in a relevant or advantageous manner and is fitted to a geometry. By fitting the surface model to a specific geometry, the accuracy of the subsequent steps may be improved. In other examples, active detection or sensing of the surface is performed as part of the geometry detection sub-step.

In a fourth step 104, a three-dimensional crack profile 222 is derived based on the surface model and at least a first set of data. The three-dimensional crack profile may be derived in any suitable fashion and using any suitable derivation methodology or algorithm. The at least first set of data may comprise any suitable or relevant data or information on which to base the derivation. An exemplary derivation step will be described in more detail below.

The three-dimensional crack profile may comprise any suitable parameters, characteristics and variables that describe or otherwise reflect the properties of the first crack. In an example, the three-dimensional crack profile comprises: a first crack boundary; a first crack mouth; a first crack length; and a first crack depth. Each of these may be provided in any suitable or advantageous format or manner. In some examples, the three-dimensional crack profile contains additional or ‘meta’ data. Examples of such additional data includes (without limitation) data relating to the material in which the first crack is located; environmental data; or data relating to standard factors, values or constants. The step of deriving the three-dimensional crack profile may be carried out by any suitable device, unit or element. In some examples, the step of deriving the three- dimensional crack profile is carried out by the user’s device. In other examples, the step of deriving the three-dimensional crack profile is carried out by a device located remotely from the user’s device (e.g., a laptop computer, desktop computer, server, tablet, or ‘cloud’ device). The device carrying out the step of deriving the three-dimensional crack profile can, in some examples, be the same device that carries out the step of deriving the surface model for the first crack. In examples wherein the step of deriving the three-dimensional crack profile is performed by a device located remotely from the user’s device or the device performing the surface model derivation step, the step of deriving the three- dimensional crack profile additionally comprises one or more transmission sub steps, in which the relevant surface model data is transmitted to the relevant device performing the derivation of the three-dimensional crack profile. It will be appreciated, of course, that a number of specific implementations of such data transmission steps may be envisaged within the scope of the present disclosure.

An exemplary step for deriving a three-dimensional crack profile will now be described with reference to Figure 3 and Figure 4. For ease of comparison with Figure 2, elements of Figure 4 similar to corresponding elements of Figure 2 are labelled with reference signs similar to those used in Figure 2, but with prefix “4” instead of “2”.

In a first deriving step 301 , the surface model 420 of the first crack is segmented into a plurality of segments 424. The segmentation may be carried out in any suitable fashion and may be segmented into any suitable number of segments.

The segments may be arranged in any suitable or advantageous pattern. In some examples, the plurality of segments are arranged along a pre-defined line. In a specific example, the plurality of segments are arranged along the line defining the first crack length. In other examples, the plurality of segments are arranged in a pre-defined pattern. In some examples, the pre-defined pattern is a two- dimensional pattern. In the present example, the first deriving step comprises dividing the surface model of the first crack into a plurality of crack segments, each segment being oriented with a main axis perpendicular to a direction of the crack length 418 and parallel with the first surface.

For each segment, the segment length may be defined in any suitable fashion. In an example, the segment length 426 is defined as the distance between one crack boundary and the other crack boundary in a direction perpendicular to the first crack length. It will, however, be appreciated that this is for exemplary purposes only, and that alternative segment lengths could equally well be used. While the segment length may not precisely correspond to the first crack width, it simplifies the processing and modelling of the first crack.

Each of the plurality of segments may have any suitable segment width (the width being defined as a direction parallel with the first crack length). In an example, each of the plurality of segments has the same segment width. In another example, the segment width varies for at least one of the plurality of segments.

As described above, the surface model of the first crack may be segmented into any suitable number of segments. It will be appreciated that the lowest number of segments possible is one. Reducing the number of segments reduces the calculative burden of the method, but may also reduce the accuracy of the three- dimensional crack profile. Conversely, increasing the number of segments may increase the calculative burden (or other resource requirements), but may also increase the accuracy of the three-dimensional crack profile.

The specific number of segments chosen for any given crack may be dependent on a number of factors, including (but not limited to): required accuracy; time or resource constraints; length of the first crack; other properties of the first crack; material parameters or characteristics of the material in which the first crack is located. It should be noted that, in the above, it has been assumed that the first crack is substantially or close to linear, i.e. , that the first crack length as determined above lies substantially within the crack boundaries. However, this will not be the case in all instances, in particular for longer cracks. In such instances, however, the present exemplary method can be modified to partition the first crack into a number of crack partitions, as will be described in more detail in the following.

In a second deriving step 302, a predicted first crack profile parameter is derived for each of said plurality of segments based on the at least first set of data. Any suitable first crack profile parameter may be derived, including (but not limited to): first crack depth; first directional crack depth; or first crack side profile.

In the present example, the first crack profile parameter comprises a first crack depth 428. This allows a basic three-dimensional crack profile to be determined and visualised. The first crack depth may be arranged in any suitable configuration relative to any of the other parameters or characteristics.

In one example, the first crack depth is assumed to be a normal to the surface in which the first crack is formed. In another example, the first crack depth is assumed to be a normal to a derived surface defined by the surface model of the first crack. The line defining the first crack depth may be positioned relative to its corresponding segment in any suitable or desirable location. In a specific example, the line defining the first crack depth intersects the plane of its corresponding segment substantially in the centre of the segment.

It will, of course, be appreciated that this is for exemplary purposes only, and that the first crack depth may be selected to extend from other points. For example, in some examples, the first crack depth is aligned with one crack boundary. In other examples, the first crack depth is aligned with the other crack boundary. In yet other examples, the alignment of the first crack depth with each respective crack segment is determined individually for each crack segment.

In the present example, as described above, the first crack depth is assumed to be a normal to the surface of the surface model for the first crack. It will be appreciated that other geometric arrangements could equally well be envisaged within the scope of the present disclosure. For example, the first crack depth could be arranged at a non-normal angle relative to the surface of the surface model.

The first crack profile parameter may be derived based on any suitable or relevant parameters, characteristics or other information. In other terms, the at least first set of data may comprise any suitable or relevant parameters, characteristics or general information. In some examples, wherein the first crack profile parameter comprises the first crack depth, the first crack profile parameter is derived based on one or more of: material thickness of a first object containing the surface in which the first crack is formed; mechanical properties of the first object; operating or environmental conditions of the first object; operating or environmental conditions of the surface; environmental or other outside effects on the first object and/or the surface. Any suitable or relevant material parameters or mechanical properties of the first object may be used, including (but not limited to): Young’s modulus; yield point; work hardening; fracture toughness; crack growth resistance curve (also known as ‘R-Curve’); or fatigue behaviour. It will be appreciated that these parameters are exemplary only and not intended to be limiting. A number of exemplary deriving steps will be presented in more detail below.

In a third deriving step 303, the surface model and the predicted first crack profile parameter are combined into a three-dimensional crack profile. The three- dimensional crack profile may comprise any suitable information, data, parameters or characteristics of the first crack. In an example, the three-dimensional crack profile comprises a first crack length, a first crack mouth, a first crack boundary and a first crack depth.

In the preceding example, the first crack profile parameter comprises a first crack depth. This example comprises a number of assumptions and simplifications that serve to simplify the determination of the three-dimensional crack profile.

However, it will no doubt be appreciated that this is an approximation only, and that the approximation does not necessarily apply in all cases. Further, in some instances, the approximation may yield only limited accuracy. A further example of a first crack profile parameter will now be described with reference to Figure 5. In this example, the first crack profile parameter comprises a first directional crack depth 528. The first directional crack depth is substantially similar to the first crack depth described above, but additionally comprising directional information 530. This allows a crack to be more precisely modelled and predicted, but also introduces additional complexity to the three-dimensional crack profile.

The directional information may be formatted or provided in any suitable fashion.

In some examples, the directional information may be provided as a direction relative to a reference direction 532. The reference direction, in some examples, is defined as the normal to the surface. In other examples, the reference direction is defined with respect to a predefined referential coordinate system. The referential coordinate system may be chosen advantageously so as to facilitate or simplify the processing and calculations.

In the present example, first crack profile parameter, which includes the directional information, is recalculated for each segment of the surface model of the first crack.

Depending on the complexity desired, the directional information can be provided in one, two or three dimensions. In the case of a single dimension, for example, the directional is substantially identical to the preceding example, where the first crack profile parameter is a first crack depth along the normal to the surface of the surface model of the first crack. In such an example, the only required information is an indication of the depth.

It will be appreciated that a number of methodologies or processes may be implemented to derive the predicted first crack profile parameter. A number of exemplary processes for deriving the predicted first crack profile parameter will now be discussed. It will be appreciated that whilst below examples are each described in isolation, this is purely for conciseness and exemplary purposes. It is, in principle, entirely possible for the below examples to be used in combination. For example, the first example may be used in combination with one or both of the second and third examples. Alternatively, the second example may be used in combination with either of the first example or the third example.

In a first example, the predicted first crack profile parameter is derived based on a set of material parameters. It is well known to use material parameters to determine, calculate or predict material behaviour, e.g., by way of sets of known material equations. As described above, mechanical parameters that may be used include (without limitation): Young’s modulus; yield point; work hardening; fracture toughness; crack growth resistance curve (also known as ‘R-Curve’); or fatigue behaviour. Based on the results of such determinations or calculations, predictions or estimates on material behaviour can be made within the limitations set by the equations in question.

In the present example, a set of material characteristics comprising experimental data obtained during load-controlled mechanical testing is correlated to a combined boundary condition by considering the stiffness of the component material. Specifically, the equations allow the calculation or derivation of a local stiffness parameter for the component material, which is subsequently compared to or correlated with the mechanical testing results (e.g., by way of a normalised crack growth resistance curve, otherwise known as a normalised R-Curve) in order to derive a predicted first crack profile parameter.

In a second example, the predicted first crack profile parameter is derived based on a finite element analysis. Finite element analysis covers a group of methodologies for solving specific computational problems. In general, such methodologies divides a specific system into a plurality of a smaller geometric portions or parts (typically referred to as ‘finite elements’). By dividing the system into discrete elements, it becomes possible to predict the behaviour of a complex problem (typically described by partial differential equations). The general principles of finite element analyses will not be discussed in more detail in the present disclosure. Examples of finite element analysis methods that may (without limitation) be used within the scope of the present disclosure include (but are not limited to): extended finite element method (XFEM); generalised finite element method (GFEM); or mixed finite element method. The finite element analysis may use any relevant or suitable parameters, characteristics or variables in respect of the component or the materials comprised in the component. The finite element analysis may use, without limitation: material parameters or characteristics (e.g., Young’s modulus, yield point, work hardening, fracture toughness, crack growth resistance curve, or fatigue behaviour); component geometry; component dimensions; boundary conditions of the component or structure; operating conditions of the component or structure; environmental conditions surrounding the component or structure; or loading history of the component or structure.

In a third example, the predicted first crack profile parameter is derived by a first predictive algorithm. The predictive algorithm may use any suitable parameters or data sets in order to derive the predicted first crack profile parameter. In an example, the first predictive algorithm derives the first crack profile parameter based on at least one of: a set of material parameters (e.g., as substantially described above); a result of a first finite element analysis; or at least a first set of predetermined material characteristics (e.g., the above-mentioned mechanical testing results).

In some examples, the predictive algorithm is a trainable algorithm or learning algorithm. Trainable predictive algorithms typically require one or more sets of initial data in order to provide meaningful results. The amount of data required may depend on the algorithm and the properties thereof, as well as on the requirements for the intended results to be derived from the trainable predictive algorithm. Such sets of initial data are commonly referred to as training data.

Any suitable set (or sets) of training data may be used in order to ‘train’ the predictive algorithm. The predictive algorithm, in some examples, uses as an initial input a set of material data obtained by way of a set material parameters used to determine material behaviour. In some examples, the predictive algorithm uses as an initial input a set of finite element data obtained by way of a finite element analysis. In some examples, the predictive algorithm uses as an initial input both a set of material and a set of finite element data. In other examples, the predictive algorithm uses one or more of: a set of material data; a set of finite element data; or other input data relating to the component or the surrounding environment thereof.

As described above, a trainable predictive algorithm may, in some instances, require a significant amount of training data to provide meaningful results. As such, in some examples, the trainable predictive algorithm may not initially yield satisfactory results. In such examples, the trainable predictive algorithm may be used in parallel with another derivation methodology (e.g., the aforementioned material parameter calculations and/or the finite element analysis calculations). Output data from the predictive algorithm may be compared with a reference to determine whether the output data from the trainable predictive algorithm is within a first threshold of a reference result. If the output data is not within the first threshold, the output data from the predictive algorithm is not used. Instead, the results from the material parameter calculations and/or the results from the finite element analysis calculations are used.

It will be appreciated that other types of predictive algorithms may be envisaged within the scope of the present disclosure

As discussed above, whilst described in isolation above, each of the above examples may be used in combination. One such example will now be discussed with respect to Figure 6.

In a first step 601 , a first crack is detected in a surface. The detection may be carried out in any suitable fashion. In the present example, the detection is carried out substantially as described with reference to Figures 1 and 2 above.

In a second step 602, a first set of crack characteristics of the first cracks is measured. Similarly to the detection step, the measurement step is in the present example carried out substantially as described with reference to Figure 1 and 2 above. It will, however, be appreciated that numerous methods for measuring the characteristics of a detected crack may be envisaged within the scope of the present disclosure. In a third step 603, at least a first set of data is provided. The first set of data may be provided in any suitable fashion and may contain any suitable information pertaining to any relevant aspect, characteristic or parameter of the first crack (or relevant aspects thereof).

In the present example, purely for exemplary reasons, there is provided a first set of data and a second set of data. The first set of data comprises information relating to the specification or characteristics of the component under test. This comprises (without limitation): design or specification data for the component; geometry data of the component or parts thereof; thickness of the component; length of the component; width of the component; tensile load; compression; bending data; internal pressure data; environment temperature, environment moisture content; environmental wind speed; or other environmental or external data. In other terms, the first set of data, in the present example, comprises observational and measurement data relating to the component or the environment in which the component is positioned. The first set of data may be obtained in any suitable or relevant fashion.

The second set of data, in the present example, comprises data relating to properties, characteristics or parameters of the material (or materials) of which the component is comprised.

The second set of data may be obtained in any suitable way. For example, at least a portion of the second set of data may be obtained by way of the above- described correlation between testing data and the real situation. Typically, at least a portion of the second set of data is obtained by performing mechanical testing on the component (or a representation thereof). For example, the mechanical testing may include determining mechanical properties or characteristics of the component material or materials. It may also include determining crack mouth opening displacement under load control conditions with different values for stiffness (geometry sizes) in order to generate a normalized R- Curve (crack growth resistance curve) for the component. The mechanical testing may also include one or more of: fracture mechanic parameters (e.g., K, J or C* parameters); standard fatigue testing; standard tensile testing; or standard fracture toughness testing.

It will be appreciated that some or all of the data comprised in the at least first set of data and the second set of data may be obtained prior to the commencement of the present exemplary method.

In a fourth step 604, a surface model of the first crack is derived. The surface model may, as described above, be derived in any suitable fashion. In the present example, purely for exemplary purposes, the surface model is derived substantially as described with reference to Figures 1 and 2 above.

In a fifth step 605, a predicted first crack profile parameter is derived based on a set of material parameters. As described above, material parameters can be used to determine, calculate or predict material behaviour, e.g., by way of sets of known material equations. Based on the results of such determinations or calculations, predictions or estimates on material behaviour can be made within the limitations set by the equations in question.

In a sixth step 606, the predicted first crack profile parameter is derived based on a finite element analysis. As described above, finite element analysis allows a complex problem, e.g., the behaviour of a component, in an efficient manner. The finite element analysis may use any relevant or suitable parameters, characteristics or variables in respect of the component or the materials comprised in the component. The finite element analysis may use, without limitation: material parameters or characteristics (e.g., Young’s modulus, yield point, work hardening, fracture toughness, crack growth resistance curve, or fatigue behaviour); component geometry; component dimensions; boundary conditions of the component or structure; operating conditions of the component or structure; environmental conditions surrounding the component or structure; or loading history of the component or structure.

In a seventh step 607, the predicted first crack profile parameter is derived by a first predictive algorithm. Any suitable predictive algorithm may be employed. In some examples, the predictive algorithm is a trainable algorithm or learning algorithm.

The first predictive algorithm may use any suitable data or information as input to carry out the derivation. In some examples, one or more of the surface model, the first set of data or second data is used as input for the first predictive algorithm. In some examples, some or all of the results from the fifth or sixth steps may be used as input for the first predictive algorithm.

In some examples, results from one of the fifth or sixth steps are used as input for the first predictive algorithm and the results of the other of the fifth or sixth steps are used as a verification benchmark so that the output of the predictive algorithm can be verified. In such examples, the seventh step comprises a verification sub step. The verification sub-step may be carried out in any suitable fashion. Purely by way of example, the verification sub-step may comprise checking whether the result of the seventh step is within a pre-defined threshold of one of the fifth or sixth step. The pre-defined threshold may be a percentage of the result of the fifth or sixth step, or it may be a pre-defined value.

It will be appreciated that the fifth, sixth and seventh steps, whilst described sequentially, are described sequentially for exemplary purposes only, and that these steps may, in principle, be carried out in any suitable order. In some examples, two or more of the steps are carried out substantially simultaneously.

In some examples, one or more of the fifth, sixth or seventh steps may not be relevant or advantageous to carry out. For example, under certain circumstances, as part of the aforementioned verification sub-step it may be determined that the fifth step yields a result within an acceptable accuracy, thereby obviating the need for the sixth step to be carried out. As another example, it may be determined as part of the verification sub-step that the seventh step yields a result with an acceptable accuracy or margin of error, which obviates the need for carrying out the fifth or sixth steps. The verification sub-step may be carried out in any suitable fashion. It will be appreciated that a number of implementations of a verification sub-step may be envisaged within the scope of the present disclosure. Purely by way of example, a verification sub-step may comprise determining whether the result of the seventh step is within a pre-defined threshold value. The pre-defined threshold value may, for example, be a percentage of the result of a preceding calculation (e.g., earlier instances of carrying out either or both of the fifth or sixth steps), or it may be a fixed pre-defined value. In other examples, the threshold is calculated or determined dynamically based on a relevant set of criteria.

In an eighth step 608, one of the at least one derived predicted first crack profile parameter is selected for use. The selection may be based on any suitable or relevant criteria.

In some examples, the results of the one or more of the fifth, sixth or seventh steps may be verified as part of the selection step. This may, for example, be achieved by use of a suitable verification sub-step. Such a verification sub-step, similarly to that described above, may be carried out in any suitable fashion. The verification sub-step may, in some examples, perform a comparison with a suitable verification value (or set of values). Purely by way of example, a verification sub-step may comprise determining whether the result of the seventh step is within a pre-defined threshold value of one or both of the fifth or sixth step. The pre-defined threshold value may be a percentage of the result of the fifth or sixth step, or it may be a pre-defined value. In other examples, the threshold is calculated or determined dynamically based on a relevant set of criteria.

In some examples, one methodology is used as the default determining method and one or both of the remaining methodologies is used only as a verificatory calculation. Accordingly, such verificatory determinations may be performed only a portion of the time. If, purely by way of example, the seventh step is used by default to determine the predicted first crack profile, the fifth step and/or sixth step may be used for a percentage of the overall number of calculations to verify the accuracy of the result of the determination. Specifically, in an example, the fifth step is performed to verify the accuracy of 25% of the results of the seventh step. In other examples, the accuracy of a prediction result may be verified less often.

In yet other examples, the accuracy of a prediction result may be verified more often. In a ninth step 609, the surface model and the predicted first crack profile parameter is combined into a three-dimensional crack profile. This may be performed in any suitable fashion or manner. In an example, the combination step is carried out substantially as described with reference to Figure 3 above.

As previously described, the above examples have assumed that the first crack is substantially linear, i.e. , that the line defining the first crack length is located substantially between the crack boundaries. However, it will be appreciated this assumption is not always valid. For example, some cracks may be curved, have ‘corners’ or ‘kinks’, or may consist of a number of those features in combination.

In such circumstances, a line between the crack corners may not accurately approximate the actual length of the crack. This, in turn, could negatively influence the accuracy of the calculations.

It may therefore be advantageous, in some instances, to divide the first crack into a number of partitions. An exemplary method for partitioning a first crack will now be described with reference to Figure 7 and Figure 8.

In a first step 701 , a first crack corner 814 is detected. In a second step 702, a second crack corner 816 is detected.

Both of the first and second steps may be carried out in a suitable fashion. In an example, these steps are carried out in a manner substantially identical to the steps described above.

In a third step 703, the first crack is partitioned into a plurality of partitions if it is determined that at least a portion of the first crack length 808 is positioned outside the first crack boundaries. The determination sub-step may be carried out in a suitable fashion.

The partitioning step may be carried out in a suitable fashion. It will no doubt be appreciated that a number of specific implementations may be envisaged within the scope of the present disclosure. Accordingly, the partitioning will be discussed in general terms below.

In the present example, the partitioning step comprises a plurality of sub-steps. In a first sub-step 703a, it is determined that the first crack length 808 comprises at least a portion that is located substantially outside the first crack boundary 812a, 812b.

In a second sub-step 703b, at least one first crack feature 834 is detected. The first crack feature may be detected in any suitable fashion. In some examples, the at least one first crack feature is detected in a manner similar to that described above in respect of detecting crack boundaries and/or crack corners. Any suitable, relevant or advantageous crack feature may be detected, including (without limitation) bends, kinks, corners, curves or discontinuities. It will generally be appreciated that a number of specific methodologies may be envisaged by which such a detection can be made within the scope of the present disclosure.

In a third sub-step 703c, the first crack is divided into a set of first crack partitions 836a, 836b based on the at least one detected first crack feature and a set of criteria. The first crack may be divided based on any suitable set of criteria. In an example, the first crack is partitioned along at least one partitioning line 838, the at least one partitioning line being positioned substantially at or near the first crack feature 834.

The partitioning line may be oriented and positioned in any suitable way so as to divide the surface model of the first crack into two separate partitions. In the present example, the partitioning line is located substantially at the first crack feature. The partitioning line, in the present example, is oriented to be substantially perpendicular to the line representing the crack boundary.

Any suitable criteria may be used to determine the position of the partitioning line. Suitable criteria include, but are not limited to: curvature of first crack boundary; presence of kinks in the first crack boundary; or distance of first crack length to first crack boundary. In a fourth sub-step 703d, a set of first crack partition lengths 840a, 840b is generated based on the derived set of first crack partitions. Each of the first crack partition lengths may be derived in a suitable fashion. In an example, each of the first crack partition lengths is derived in a manner substantially identical to the first crack length discussed in the above examples.

Each first crack partition length may extend between any suitable end points. For crack partitions located at either end of the crack, one end point is typically the crack corner. For the remaining end points, it is possible to derive and select any number of suitable points. In the present example, the other end point is located on the partitioning line 838, substantially in the middle thereof. It will be appreciated that this is merely for exemplary purposes, and that a number of specific end point choices may be made within the scope of the present disclosure.

In the above exemplary methods, the individual steps have been described as being carried out on any one of a number of possible devices. For example, it is possible for substantially all of the method steps of the above-discussed methods to be carried out on a user’s device. In other examples, some of the steps may be carried out on the user’s devices, whereas other may be carried out on devices located remotely from the user’s device. In such examples, some of the method steps comprise relevant transmission and retransmission sub-steps as described in more detail above.

Below, a number of specific implementation examples are described purely for exemplary purposes. It will be appreciated that variations on these examples may be envisaged within the scope of the present disclosure.

A first exemplary implementation will now be discussed with reference to Figure 9. For purely exemplary purposes, the present example describes a system in which a method that is substantially similar to the method described above with reference to Figure 1 and Figure 2 may be implemented. It will, however, be appreciated that any of the remaining methods described in the above could, in principle, be implemented in substantially the same manner.

In the present example, a first device 950 is used to detect a first crack in a surface of a component 952. In the present example, the first device is a user device, such as a mobile phone or other personal mobile device. The second device may comprise any suitable components, elements or systems. The detection step is carried out in a suitable manner, e.g., (without limitation) by way of a camera unit comprised in the first device, wherein the camera unit is used to obtain at least an image or frame of the first crack.

Subsequently or simultaneously, the first device 950 is used to measure a first set of crack characteristics of the first cracks. The measuring is performed in a suitable manner, e.g., (without limitation) by analysis of an image or frame obtained during the detection step. It should be noted, however, that whilst the first device is used for the measurement in the present example, it is possible for a remote device (e.g., a second device, such as is described below). This may be particularly relevant in instances where the first device may lack in processing power or other resources.

Subsequently, a surface model of the first crack is derived. In the present example, this step is performed by a second device 954 located remotely from the first device. The second device may comprise any suitable components, elements or systems. Examples of a second device include, but are by no means limited to: mobile phone; tablet computer; laptop computer; desktop computer; server; or cloud-based device. Prior to the derivation, relevant data is transmitted from the first device to the second device via a suitable connection 956.

Subsequently to the derivation of the surface model, a three-dimensional crack profile based on the surface model and at least a first set of data is derived at the second device 954. The three-dimensional crack profile may be derived in a suitable manner, for example in one of the exemplary ways described in the preceding examples above. Subsequently to the derivation of the three-dimensional crack profile, resulting data is transmitted from the second device 954 to the first device 950 by way of a suitable connection 958 (which, in some examples, is identical to the connection 956). The results of the derivations can then be displayed to the user in a convenient manner.

A second exemplary implementation will now be discussed with reference to Figure 10. For purely exemplary purposes, the present example describes a system in which a method that is substantially similar to the method described above with reference to Figure 1 and Figure 2 may be implemented. It will, however, be appreciated that any of the remaining methods described in the above could, in principle, be implemented in substantially the same manner.

In the present example, a first device 1050 is used to detect a first crack in a surface of a component 1052. In the present example, the first device is a user device, such as a mobile phone or other personal mobile device. The detection step is carried out in a suitable manner, e.g., (without limitation) by way of a camera unit comprised in the first device, wherein the camera unit is used to obtain at least an image or frame of the first crack.

In the preceding example, the first device 1050 was used to measure a first set of crack characteristics of the first cracks. Flowever, this may not be feasible or convenient in all circumstances. As mentioned above, the first device may not have the required processing power or resources to carry out this measurement step. In some instances, it may be undesirable to carry out any substantial processing on the first device (e.g., to save battery power or if the first device is a shared or public device). In such situations, the measurement step can be carried out by a remotely located device or system, such as a second device 1054 or, as in the present example, a third device 1060. Prior to the measuring, relevant data is transmitted from the first device to the third device by way of a suitable connection 1062. As described above, the measuring is performed in a suitable manner, e.g., (without limitation) by analysis of an image or frame obtained during the detection step.

Subsequently, a surface model of the first crack is derived. In the present example, this step is performed by a second device 1054 located remotely from the first device 1050 and the third device 1060. It will, however, be appreciated that the surface model derivation step could, in principle, also be performed by either of the third device or the first device. Prior to the derivation, relevant data is transmitted from the third device to the second device. This may be performed by way of a direct connection 1064 between the third device and the second device, or by way of a suitable connection between the third device and first device 1066 and a suitable connection 1056 between the first device and the second device.

Subsequently to the derivation of the surface model, a three-dimensional crack profile based on the surface model and at least a first set of data is derived at the second device 1054. The three-dimensional crack profile may be derived in a suitable manner, for example in one of the exemplary ways described in the preceding examples above.

Subsequently to the derivation of the three-dimensional crack profile, resulting data is transmitted from the second device 1054 to the first device 1050 by way of a suitable connection 1058 (which, in some examples, is identical to the connection 1056). The results of the derivations can then be displayed to the user in a convenient manner.

As mentioned above, the above-discussed methods may be implemented in a number of specific ways. A number of exemplary methods that could be implemented in at least parts of the systems discussed with reference to Figure 9 and Figure 10 above will now be discussed. It should be noted that, for purposes of conciseness and clarity, only features and elements that differ substantially from corresponding features and elements described above will be discussed in detail in the following. In one such exemplary method, which will now be discussed with reference to Figure 11 , there is provided a method for a user device. Any suitable user device may be employed, including, but not limited to, a mobile phone, a tablet, a laptop computer or a digital camera.

In a first step 1101, a first crack in a surface of a component is detected. The first crack may be detected in any suitable fashion, such as (without limitation) a camera in a manner substantially described with reference to Figures 1 and 2 above.

In a second step 1102, data relating to the detected first crack is transmitted to a remote device, the remote device being operable to at least derive a surface model of the first crack and derive a three-dimensional crack profile based on the surface model and at least a first set of data. The data may be transmitted in any suitable fashion using a suitable connection. It will be appreciated that a number of specific implementations of such connections may be envisaged within the scope of the present disclosure.

In a third step 1103, a three-dimensional crack profile is received from the remote device. The three-dimensional crack profile may be received in any suitable fashion using a suitable connection. It will be appreciated that a number of specific implementations of such connections may be envisaged within the scope of the present disclosure.

In an optional fourth step 1104, the three-dimensional crack profile is displayed to the user of the user device in a suitable fashion. This could, for example, be accomplished by way of a display component comprised in the user device.

In an optional fifth step 1105, a first set of crack characteristics of the first crack is measured. Typically, the measurement step is performed prior to transmission of data to the remote device. In such instances, the first set of crack characteristics is transmitted along with the other data relating to the detected first crack. The measurement step may be carried out in a suitable fashion, such as is, for example, described above. It should be noted that the above-described exemplary method is exemplary only and not intended to be limiting. It will be appreciated that the exemplary method may comprise alternative or additional steps. In some examples, the exemplary method comprises one or more such optional or alternative steps as are described in the aforementioned methods.

In another exemplary method, which will be discussed with reference to Figure 12, there is provided a computing device. The computing device may comprise any suitable components, elements or systems. Examples of computing devices include, but are by no means limited to: mobile phones; tablet computers; laptop computers; desktop computers; servers; or cloud-based devices.

In a first step 1201 , data relating to a detected first crack is received from a user device. The data may be received in any suitable manner using a suitable connection. It will be appreciated that a number of specific implementations of such connections may be envisaged within the scope of the present disclosure.

In a second step 1202, a surface model of the first crack is derived. The surface model may be derived in a suitable fashion, such as (purely by way of example) in a manner such as described with reference to Figure 1 and Figure 2 above.

In a third step 1203, a three-dimensional crack profile is derived based on the surface model and at least a first set of data. The three-dimensional crack profile may be derived in a suitable fashion, such as (purely by way of example) in a manner such as described with reference to Figure 1 and Figure 2 above.

In a fourth step 1204, the three-dimensional crack profile is transmitted to the user device. The three-dimensional crack profile may be transmitted in any suitable manner and using a suitable connection. It will, once again, be appreciated that a number of specific implementations of such connections may be envisaged within the scope of the present disclosure. In an optional fifth step 1205, a first set of crack characteristics of the first crack is measured. Typically, the measurement step is performed prior to transmission of data to the remote device. In such instances, the first set of crack characteristics is transmitted along with the other data relating to the detected first crack. The measurement step may be carried out in a suitable fashion, such as is, for example, described above.

It should be noted that the above-described exemplary method is exemplary only and not intended to be limiting. It will be appreciated that the exemplary method may comprise alternative or additional steps. In some examples, the exemplary method comprises one or more such optional or alternative steps as are described in the aforementioned methods.

While specific embodiments of the invention have been described above, it will be appreciated that an embodiment of the invention may be practiced otherwise than as described. For example, an embodiment of the invention may take the form of a computer program containing one or more sequences of machine-readable instructions describing a method as disclosed above, or a data storage medium (e.g. semiconductor memory, magnetic or optical disk) having such a computer program stored therein.

The descriptions above are intended to be illustrative, not limiting. Thus, it will be apparent to one skilled in the art that modifications may be made to the invention as described without departing from the scope of the claims set out below.

Claims

1. A method for preparing a surface crack profile, comprising: detecting a first crack in a surface of a component; measuring a first set of crack characteristics of the first crack; deriving a surface model of the first crack; and deriving a three-dimensional crack profile based on the surface model and at least a first set of data.

2. A method according to claim 1 , wherein the step of deriving a three- dimensional crack profile comprises: segmenting the surface model of the first crack into a plurality of segments; deriving a predicted first crack profile parameter for each of said plurality of segments based on the at least first set of data; and combining the surface model and the predicted first crack profile parameter into a three-dimensional crack profile.

3. A method according to claim 2, wherein the predicted first crack profile parameter comprises a first crack depth.

4. A method according to claim 2, wherein the predicted first crack profile parameter comprises a first directional crack depth.

5. A method according to any of claims 2 to 4, wherein the first crack profile parameter is derived based on a set of material parameters.

6. A method according to claim 5, wherein the set of material parameters comprises a set of mechanical testing results, and wherein the step of deriving comprises deriving the predicted first crack profile parameter based on a correlation between the set of mechanical testing results and a derived local stiffness parameter for the component material.

7. A method according to any of claims 2 to 4, wherein the first crack profile parameter is derived based on a finite element analysis.

8. A method according to any of claims 2 to 4, wherein the first crack profile parameter is derived by a first predictive algorithm based on at least one of: a set of material parameters; a result of a first finite element analysis; or at least a first set of predetermined material characteristics.

9. A method according to any of claims 5 to 8, wherein the first crack profile parameter is derived based on two or more of: a set of material parameters; a finite element analysis; or a predictive algorithm based on at least one of: a set of material parameters, a result of a first finite element analysis, or at least a first set of predetermined material characteristics.

10 A method according to any preceding claim, wherein the step of detecting comprises obtaining at least one image of the surface.

11. A method according to claim 10, wherein the step of detecting a first crack on a surface comprises: obtaining a plurality of images of the surface; and stitching the plurality of obtained images together to form a combined surface image.

12. A method according to claim 10 or claim 11 , wherein the at least one image of the surface is projected onto a substantially planar surface.

13. A method according to claim 10 or claim 11 , wherein the at least one image of the surface is projected onto a three-dimensional surface.

14. A method according to any preceding claim, wherein the step of measuring a set of characteristics comprises: determining a first crack face boundary; determining a first crack length; and determining a first crack mouth.

15. A method according to claim 14, wherein determining a first crack length comprises: detecting a first crack corner; detecting a second crack corner; and defining the crack length as a distance between the first crack corner and the second crack corner.

16. A method according to claim 15, further comprising: partitioning the first crack into a plurality of partitions if it is determined that at least a portion of the first crack length is positioned outside the first crack boundaries.

17. A method according to any preceding claim, wherein the step of partitioning comprises: determining that the first crack length comprises at least a portion that is located substantially outside the first crack boundaries; detecting at least one first crack feature; dividing the first crack into a set of first crack partitions based on the at least one detected first crack feature; and a set of first crack partition lengths is generated based on the derived set of first crack partitions.

18. A computer system operable to carry out the method of any of claims 1 to 17.

19. A computer system according to claim 18 comprising: a first device operable to carry out at least a first portion of the method steps of any of claims 1 to 17; and a second device operable to carry out at least the remaining of the method steps of any of claims 1 to 17.

20. A method for a user device, the method comprising: detecting a first crack in a surface of a component; transmitting data relating to the detected first crack to a remote device, the remote device being operable to at least derive a surface model of the first crack and derive a three-dimensional crack profile based on the surface model and at least a first set of data; and receiving a three-dimensional crack profile from the remote device.

21. A method according to claim 20, further comprising: measuring a first set of crack characteristics of the first crack.

22. A method according to claim 20 or claim 21 , further comprising: displaying a representation of the three-dimensional crack profile to the user.

23. A user device operable to carry out the method of any of claims 20 to 22.

24. A method for a computing device, the method comprising: receiving data relating to a detected first crack from a user device; deriving a surface model of the first crack; deriving a three-dimensional rack profile based on the surface model and at least a first set of data; and transmitting the three-dimensional crack profile to the user device.

25. A method according to claim 24, further comprising; measuring a first set of crack characteristics of the first crack.

26. A computing device operable to carry out the method of either claim 24 or claim 25.