Neuromorphic vision-based crack monitoring system for monitoring deformation of a loaded material to detect and track cracking phenomena in the loaded material and computer program product thereof

Abstract

A crack monitoring system (1) for monitoring deformation of a loaded material (2) to detect and track cracking phenomena in the loaded material (2); the crack monitoring system (1) comprises: an imaging sensory system (3) configured to output an output that allows cracking phenomena in the loaded material (2) to be detected and tracked; and electronic processing resources (4) in communication with the imaging sensory system (3) to receive and process the output thereof to detect and track cracking phenomena in the loaded material (2). The imaging sensory system (3) comprises: an event camera (6) arranged to be directed towards an observed portion (2A) of the loaded material (2) to be monitored for cracking phenomena and operable to output an asynchronous stream of events (ES) triggered by changes in brightness in the observed portion (2A) of the loaded material (2); and a digital camera (7) arranged to be directed towards the observed portion of the loaded material (2) to be monitored for cracking phenomena and operable to output a synchronous stream of a plurality of digital images (DI) of the observed portion (2A) of the loaded material (2).

Description

[0001] “NEUROMORPHIC VISION-BASED CRACK MONITORING SYSTEM FOR MONITORING DEFORMATION OF A LOADED MATERIAL TO DETECT AND TRACK CRACKING PHENOMENA IN THE LOADED MATERIAL AND COMPUTER PROGRAM PRODUCT THEREOF”

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This Patent Application claims priority from Italian Patent Application No.

[0004] 102025000006231 filed on March 26, 2025, the entire disclosure of which is incorporated herein by reference.

[0005] TECHNICAL FIELD OF THE INVENTION

[0006] The invention relates in general to the field of crack detection, crack length measurements and damage rate estimation, in particular to a crack monitoring system and a computer program product or software thereof for monitoring deformation of a loaded material (e.g. a specimen, like double-edge notched or structures manufactured in brittle and quasi-brittle materials) to detect and track cracking phenomena while measuring crack propagation parameters (e.g. stress intensity factor and fracture toughness).

[0007] STATE OF THE ART AS is known, in fracture mechanics, crack propagation and stress intensity factor estimation play a crucial role in determining and predicting a failure behaviour and damage rate of coupons and structures (e.g. a specimen, like double-edge notched or structures manufactured in brittle and quasi-brittle materials) under different loading conditions.

[0008] The abovementioned aspect is of utmost importance for fields like industry, environmental conservation and, more in general, society. While extensively studied in metals, crack propagation in brittle (e.g., rock-based and ceramic-based specimens and structures) and quasi-brittle (e.g., concrete-based and composites-based specimens and structures) solids is a sudden and fast phenomenon and determining its evolution in time is challenging.

[0009] Current, known solutions provide approaches for crack monitoring on the abovementioned materials that are typically performed under dynamic testing conditions, where high-speed cameras synchronized with the mechanical testing device are used to capture the propagation process.CN 203 811 458 U discloses a monitoring system for super dynamic strain fields in rock blasting, which can obtain a dynamic fracture mechanics experiment effect under blast load on natural rocks.

[0010] WO 2024 / 199846 Al discloses apparatuses and methods for characterizing an object, as well as an apparatus and a method for training a machine-learning model.

[0011] “ High-Speed Deformation Measurement with Event-Based Cameras’" by Zhu C. el al. discloses an integrated method, from calibration to measurement, using a multi -event camera array for high-speed 3D deformation monitoring of structures in extreme illumination conditions.

[0012] OBJECT AND SUMMARY OF THE INVENTION

[0013] The Applicant notes that current, known solutions are susceptible to improvements. In particular, the Applicant has noted that the current, known approaches are costly and difficult to implement; on this regard, the Applicant has noted that the current, known solutions are challenging to tune and synchronize with external devices, such as universal testing machine, split Hopkinson bar and ultra-high speed cameras and requires large amount of memory to save all the recorded data. Moreover, the known solutions typically produce, as an output, considerable amount of data to record, rendering them impractical for quasi-static loading conditions.

[0014] Thus, the aim of the present invention is to develop a system and a computer program product thereof for monitoring cracks, especially on specimens and structures made with brittle or quasi-brittle solids, thereby solving at least part of the abovementioned inconveniences.

[0015] This aim is achieved by the present invention, that relates to a crack monitoring system and a computer program product thereof for monitoring deformation of a loaded material, e.g. notched specimens and structures, to detect and track cracking phenomena, especially on specimens and structures made with brittle or quasi-brittle solids, as claimed in the appended claims.

[0016] BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 schematically shows a system for monitoring cracks in a specimen or in a structure according to the present invention.

[0018] Figure 2 schematically shows a crack propagation, data acquisition and the postprocessing method implemented by a system for monitoring cracks in a specimen or in astructure according to the present invention.

[0019] Figure 3 schematically shows an application example of a system for monitoring cracks in a specimen or in a structure according to the present invention.

[0020] DESCRIPTION OF PREFERRED EMBODIMENTS OF THE INVENTION

[0021] The present invention will now be described in detail with reference to the accompanying drawings in order to allow a skilled person to implement it and use it. Various modifications to the described embodiments will be readily apparent to those of skill in the art and the general principles described may be applied to other embodiments and applications without however departing from the protective scope of the present invention as defined in the appended claims. Therefore, the present invention should not be regarded as limited to the embodiments described and illustrated herein but should be allowed the broadest protection scope consistent with the features described and claimed herein.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning commonly understood by one of ordinary skill in the art to which the invention belongs. In case of conflict, the present specification, including the definitions provided, will control. Furthermore, the examples are provided for illustrative purposes only and as such should not be considered limiting.

[0023] In particular, the block diagrams included in the attached figures and described below are not to be understood as a representation of the structural features, i.e. constructional limitations, but must be understood as a representation of functional features, i.e. intrinsic properties of the devices defined by the effects obtained, that is to say functional restrictions, which can be implemented in different ways, so as to protect the functionalities thereof (operational capability).

[0024] In order to facilitate the understanding of the embodiments described herein, reference will be made to some specific embodiments and a specific language will be used to describe the same. The terminology used herein is used for the purpose of describing particular embodiments only and is not intended to limit the scope of the present invention.

[0025] Figure 1 schematically shows a crack monitoring system 1 for monitoring deformation of a loaded material, for example and without it being limiting to the present invention a coupon or specimen 2; in particular, the coupon 2 is made of brittle or quasibrittle material or solid, such as rock-based and ceramic-based or concrete-based andcomposites-based coupons. According to an aspect of the present invention, the coupon 2 may be, for example, a double edge notched specimens or the like. According to an aspect of the present invention, the coupon 2 is suitable for fracture parameters estimation, i.e. it is a specimen or a structure which can be analysed to determine relevant fracture parameters in relation to the material with which it is composed of.

[0026] The crack monitoring system 1 comprises:

[0027] - an imaging sensory system 3 configured to output an output that allows cracking phenomena in the coupon 2 to be detected and tracked; and

[0028] - electronic processing resources 4 in communication with the imaging sensory system 3 to receive and process the output thereof to detect and track cracking phenomena in the coupon 2.

[0029] According to an aspect of the present invention, the electronic processing resources 4 are of a local type, e.g. they are implemented as a local unit; according to a further and alternative aspect of the present invention, electronic processing resources 4 are of a distributed type, e.g., on nodes in a network, and implementing paradigms such as cloud computing. Without it limiting to the present invention, the electronic processing resources 4 are hereinafter of the local type, in particular in the form of a workstation with suitable computing capabilities for data recording and post processing, i.e. capable of performing the operations disclosed hereinafter.

[0030] The imaging sensory system 3 comprises:

[0031] - an event camera 6 arranged to be directed towards an observed portion 2A of the coupon 2 to be monitored for cracking phenomena and operable to output an asynchronous stream of events, also referred to as event stream, ES, triggered by changes in brightness in the observed portion 2A of the coupon 2; and

[0032] - a digital camera 7 arranged to be directed towards the observed portion 2A of the coupon 2 to be monitored for cracking phenomena and operable to output a synchronous stream of a plurality of digital images DI of the observed portion 2A of the coupon 2.

[0033] According to an aspect of the present invention, the event camera 6 is a neuromorphic vision sensor or camera and the digital camera 7 is a digital single-lens reflex, DSLR, camera. In further detail, as also discussed below, the event camera 6 is configured to provide an output relative to asynchronous high temporal resolution per pixel data and the digital camera 7 is configured to provide an output relative to synchronous RGB frames with high spatial resolution.In further detail, the event camera 6 is an asynchronous imaging sensor configured to respond, i.e. trigger an output, to local changes in brightness in the pixels of the event camera 6. In particular, each pixel of the event camera 6 operates independently and asynchronously, reporting changes in brightness as they occur, and staying silent otherwise, meaning that each pixel independently responds to changes in brightness as they occur. In more detail, each pixel of the event camera 6 stores a reference brightness level; the event camera 6, in particular each event pixel of the latter, is configured to compare the reference brightness level to the current brightness level. If the difference in brightness exceeds a threshold, said event pixel updates its reference level and generates an event E, i.e. a discrete packet that contains the pixel address and a timestamp. According to another aspect of the present invention, each event E also contains the polarity (increase or decrease) of the brightness change.

[0034] According to an aspect of the present invention, the event stream ES comprises a plurality of events like the event E acquired in a time interval At; in particular, each event E has respective coordinates (x, y) and timestamp (t), which can be represented in a reference system (x,y,t), and being representative of a change in brightness in the observed portion 2A of the coupon 2. Thus, the event stream ES is a tridimensional depiction formed by the plurality of events E and representing the evolution over time such events E in the time interval At. In particular, an example of the event stream ES is shown in Figure 1, wherein positive and negative events, referred to as PE and NE, are shown in a 3D reference system xyt; as it can be seen, each event PE, NE, is indicative of brightness change (either positive or negative), has respective (x,y,t) coordinates, meaning that each event E has a certain position and is referring to a specific time instant.

[0035] The digital camera 7 is a synchronous device that combines the optics and mechanisms of a single-lens reflex camera with a solid-state image sensor (not shown) and is configured to digitally record the images from the solid-state image sensor.

[0036] As also shown in Figure 1, the plurality of digital images DI are acquired at a frame rate f = 1 / At, where At is the time interval between consecutive images of the plurality of digital images DI. In further detail, each digital image of the plurality of digital images DI is a three-coloured channel, in particular RGB, frame acquired in a corresponding time instant t; thus, At is defined by the time interval between two consecutive images, where the first image is acquired at time t = t’ and the second image is captured at time t = t + At.In further detail, as also discussed in the following, the digital camera 7 is configured to output high resolution RGB images at a fixed frame rate, i.e. f = 1 / At, the abovementioned frames are, for example, acquired at At and 2At, being temporally distant one another of the time interval At. Furthermore, as anticipated above, within each At the event stream ES generated by the event camera 6 covers a period of time which is substantially equal to the duration of the time interval At itself; furthermore, the event stream ES has a high temporal resolution, so that, when the electronic processing resources 4 process such even stream ES, a finer analysis of the crack evolution over time can be carried out. Additionally, the imaging sensory system 3 of the crack monitoring system 1 according to the present invention allows to monitor the observed area 2A by acquiring synchronized relevant information from the event and digital cameras 6, 7 avoiding the sensors operation with a temporal offset with respect to each other.

[0037] As better described in the following paragraphs, the present invention provides a multi -camera approach for detecting and tracking crack propagation in the coupon 2 that exploits the benefits of event cameras 6 and digital cameras 7. In fact, as also described in further detail in the following, the RGB frames captured by the digital camera 7 are used to reconstruct the deformation field during the loading phase via digital image correlation, DIC, techniques, while the event stream ES captured by the event camera 6 is mainly used for tracking the propagation of the crack on the coupon 2 during the cracking process. Therefore, as better disclosed in the following, the electronic processing resources 4 are configured to apply dedicated methods to process the two data sources, i.e. the one from the event camera 6 and the one from the digital camera 7, and fuse the outputted data to detect the crack and reconstruct the surrounding strain fields to allow for a temporal super-resolution of the cracking phenomena occurring on the coupon 2. Thus, the information acquired through the abovementioned processing allows for a more accurate determination of crack-relevant parameters during the failure process involving the coupon 2, as failure phenomena is sudden and its beginning is unlikely predictable.

[0038] The crack monitoring system 1 further comprises a mounting plate 8 configured to support and carry the imaging sensory system 3 so that the event and digital cameras 6, 7 are positioned and aligned in parallel with each other, i.e. positioned parallel to a XY plane of a Cartesian reference system XYZ at a fixed distance d, to observe the same portion, i.e. the observed area 2A, of the coupon 2; in particular, the mounting plate 8 is configured to rigidly connect the cameras 6, 7, thereby avoiding any separate movement of the event and / or digital cameras 6, 7. The mounting plate 8 is thus configured toguarantee a rigid mechanical attachment among the cameras to keep constant the extrinsic camera calibration parameters of the cameras 6, 7. In other words, the event and digital cameras 6, 7 are positioned at a specific point and oriented parallel to the coupon surface 2 to monitor the crack nucleation and propagation on the coupon 2, thereby not simultaneously considering other regions of the coupon 2 which are, for example, not of interest, i.e. for instance, not involved in the cracking process.

[0039] Thus, the mounting plate 8 is placed at a specific location and orientation to ensure the parallelism between the cameras 6, 7 while facing the region of interest 2A. Moreover, it is noted that the coupon 2 is installed in a testing machine (not shown) in compliance with the required gripping system (not shown) of the same testing machine and the mounting plate 8 is placed according to the disclosure above; in this way, during the testing, the coupon 2 is subjected to a generic external load (e.g. mechanical, thermal, etc.). Such loading condition induces the formation of cracks on the surface of the coupon 2 close to stress concentrators (e.g. crack tips, holes, etc.) the latter being designed to promote the cracking of the coupon 2 in the observed portion 2A.

[0040] According to an aspect of the present invention, the electronic processing resources 4 are configured to:

[0041] - synchronize the data output from the event camera 6 and the digital camera 7; - operate the event camera 6 and the digital camera 7;

[0042] - receive and record the asynchronous stream of events ES from the event camera 6 and the synchronous stream of the plurality of digital images DI from the digital camera 7; and

[0043] - process the asynchronous stream of events ES and the synchronous stream of the plurality of digital images DI to detect and track a cracking phenomenon in the observed portion 2A of the coupon 2.

[0044] Furthermore, the electronic processing resources 4 are configured to:

[0045] - fuse the outputted data, namely the asynchronous stream of events ES from the event camera 6 and the synchronous stream of the plurality of digital images DI from the digital camera 7, to reconstruct the strain fields to allow a high spatio-temporal resolution analysis of the cracking phenomena;

[0046] - estimate the position of the crack tip (and, thus, of the position of the point on the coupon 2 where the crack is originated) and the propagation speed of the crack on the coupon 2 based on the fused outputted data and the reconstructed strain fields; and- determine a stress intensity factor and a fracture toughness related to the coupon 2 based on the estimated position of the crack tip and the propagation speed of the crack on the coupon 2.

[0047] It is noted that, in the context of Figures 1 and 2, i.e. referring to a test set-up for monitoring cracking phenomena, before or at the same time when operating the event and digital cameras 6, 7, a mechanical load (not shown) is progressively applied to the material of the coupon 2 until failure occurs, i.e. a crack is formed on the coupon 2; therefore, the cameras 6, 7, which are synchronized by the electronic processing resources 4, are operated to record the failure phenomena and output the abovementioned streams of events ES and of the plurality of digital images DI. As also disclosed in detail above, during the loading phase of the coupon 2, the acquired plurality of digital images DI are used for the reconstruction of the deformation field except for fast the cracking development instants due to the scarce temporal resolution of the digital camera 7. Furthermore, the acquired event stream ES is mainly recording events during the rapid crack propagation phase allowing for the crack tip tracking and its speed determination.

[0048] In particular, in order to fuse the data from the asynchronous stream of events ES and the synchronous stream of the plurality of digital images DI, the electronic processing resources 4 are configured to:

[0049] - process the received asynchronous stream of events ES to track the propagation over time of the crack on the coupon 2 and output data thereof;

[0050] - process the received synchronous stream of the plurality of digital images DI to determine a deformation field during a loading phase of the loaded material and output data thereof; and

[0051] - fuse the processed data from the asynchronous stream of events ES and the synchronous stream of the plurality of digital images DI to carry out an analysis on the cracking phenomena and its evolution on the coupon 2.

[0052] Referring to Figures 1 and 2 jointly, the Applicant notes that the present invention is intended, without it being limiting for the present invention, for analysing fast and sudden crack propagation phenomena, which is typical of destructive mechanical tests, to provide information about the deformation and the stress-intensity factor of the coupon 2; such information is particularly important for structural design.

[0053] Considering Figures 1 and 2 jointly, the starting point of the approach is the installation of the coupon 2, the latter being suitable for the estimation of fracturemechanical parameters within a testing machine (not shown), the latter being observed by the crack monitoring system 1 according to the present invention.

[0054] According to an aspect of the present invention, the coupon 2 is painted with a high contrast speckle pattern to allow the deformation field reconstruction by the electronic processing resources 4 via digital image correlation, DIC, techniques (see, e.g., “ Digital Image Correlation: Advanced Methods and Applications". 2017, David Chambers Editor), the latter being applied to the synchronous stream of the plurality of digital images DI, as also disclosed in the following. It is noted that the illumination condition of the scene where the coupon 2 and the system 1 is kept constant during the whole monitoring of the coupon 2.

[0055] Furthermore, thanks to the mounting plate 8, the set-up shown in particular in Figure 1 satisfies the requirement of both cameras 6, 7 pointing towards the coupon 2 such that the structural detail of interest around the crack initiation region and its possible deformation, caused by the cracking action, lies inside the Field of View, FoV, of all cameras 6, 7, i.e. the cameras 6, 7 are arranged so as to observe the same portion of the coupon 2. Furthermore, the set of cameras 6, 7 is mechanically attached via the mounting plate 8 to facilitate a multi-sensor intrinsic and extrinsic calibration, i.e. estimating the camera internal parameters and the multi-camera transformations of the imaging sensory system 3, and ensure repeatability of the abovementioned monitoring.

[0056] Furthermore, in view of operating the event and digital camera 6, 7, the electronic processing resources 4 comprise a synchronization module 9 designed to operate the event and digital cameras 6, 7 so as to trigger the acquisition by the event and digital cameras 6, 7 at the same time instant and synchronize the output data from both cameras 6, 7. In other words, the synchronization module 9 is configured to synchronize the acquisition by the cameras 6, 7, thereby avoiding situations wherein, for example, the capturing by the event and digital cameras 6, 7 is offset one another, thereby introducing a delay in the acquisition which would have to be compensated by the same electronic processing resources 4.

[0057] In order to process the received synchronous stream of the plurality of digital images DI, the electronic processing resources 4 are configured to apply digital image correlation, DIC, techniques to the plurality of digital images DI to determine the deformation field of the coupon 2 and to track the crack evolution on the coupon 2.

[0058] In particular, DIC techniques employ tracking and registration techniques for measuring, bidimensionally or tri dimensionally, variations between images; in particular,such techniques allow to measure full-field displacements and strains, with a high spatial resolution level, especially in providing fine details about deformation. In the case of the present invention, the abovementioned DIC techniques are used by the electronic processing resources 4 to process the outputs provided by the digital camera 7 to map the variations in the full-field on the coupon 2 including the crack position and opening until the unstable crack condition is met, meaning that any change in the dimension of the crack (thus, a propagation of it on the coupon 2) and the surrounding deformation field is mapped or tracked to provide information for further processing steps in its stable propagation phase, as also reported in the following disclosure. In particular, to guarantee a spatial resolution of the quantities of interest, the full Field of View, FoV, is divided into small subsets SFOV by the electronic processing resources 4. Each subset SFOV contains several speckles, ensuring it has a unique pattern that can be identified in subsequent digital images of the plurality of digital images DI. Using cross-correlation algorithms, the displacement of each subset SFOV is determined by tracking the change in position of the speckle pattern between a reference pattern and the deformed one in the following digital images of the plurality of digital images DI. For each subset SFOV, the displacement vector can be computed whose derivation allows the reconstruction of the full strain-field.

[0059] In order to process the received asynchronous stream of events ES, the electronic processing resources 4 comprise an event-based feature tracking module 10, e.g. HASTE (see, e.g., “ Haste: multi-hypothesis asynchronous speeded-up tracking of events." by Alzugaray I. and Chli, M.), configured to process the asynchronous stream of events ES to track the visual features on the deformation field reconstructed by processing the synchronous stream of the plurality of digital images DI. Furthermore, the electronic processing resources 4 include a specialized module configured to compute the crack field displacement during the quick crack propagation process by applying event-based imagining velocimetry techniques, e.g. EBIV (see, e.g., “ Event-based imaging velocimetry: an assessment of event-based cameras for the measurement of fluid flows." by Willert C.E. and Klinner J.), based on event-based optical flow, e.g. estimated via contrast maximization (see, e.g.,

[0060]

[0061] unifying contrast maximization framework for event cameras, with applications to motion, depth, and optical flow estimation.” by Gallego G., Rebecq H. and Scaramuzza D.), from the stream of events ES.

[0062] In order to fuse the asynchronous stream of events ES and the synchronous stream of the plurality of digital images DI to detect and track the cracking phenomenon in theobserved portion 2A of the coupon 2 during the complete cracking process, the electronic processing resources 4 are configured to combine the outputs generated from processing the asynchronous stream of events ES and the synchronous stream of the plurality of digital images DI to reconstruct, in particular render, the evolution of the crack deformation being formed on the coupon 2. In particular, the output data from the imaging sensory system 3 are fused by reprojecting the features from the deformation field reconstructed by processing the synchronous stream of the plurality of digital images DI into the plane of the event camera 7 assuming a constant and known depth along the surface of the coupon 2, and matching them, in space and time, with the event features tracks and optical flow estimations obtained by processing the asynchronous stream of events ES.

[0063] According to another aspect of the present invention, the electronic processing resources 4 are further configured to determine the stress intensity factor, the fracture toughness and the crack propagation velocity based on the determined data from the asynchronous event stream ES and the synchronous stream of the plurality of digital images DI, i.e. the output indicative of the tracking of the propagation of a crack on the coupon 2 and its deformation field .

[0064] Referring in particular to Figure 2, a crack monitoring method applied by the crack monitoring system 1 in compliance with to the present invention is shown; in particular, the phases of the method implemented by the crack monitoring system 1 according to the present invention are the following:

[0065] a) during the test and data recording phase the experiment is carried out while collecting the event stream ES and the plurality of digital images DI during the entire duration of the test;

[0066] b) in the second phase the crack evolution is analysed from the recorded data as follows:

[0067] - the plurality of digital images DI corresponding to the quasi-static loading phase are processed, in particular by means of DIC techniques, to reconstruct the deformation field of the coupon 2 and to track visual features of the same coupon 2 during the application of the load (S10);

[0068] - the event stream ES concentrated during the fast crack propagation phase is processed by the event-based feature tracking module 10 to follow the crack propagation. In particular, event tracking features, derived from the event stream ES, are initialized by transferring the last tracked visual features from the plurality of digital images DI to theevent camera reference frame. Furthermore, the event-based optical flow EOF is computed by the electronic processing resources 4 to determine the displacement field (S12) following an event-based image velocimetry technique; and

[0069] c) in the data fusion, the estimations from event stream ES and the plurality of digital images DI are combined to reconstruct the full deformation field and cracking evolution during the experiment and estimate the material stress intensity factor (S13) and fracture toughness.

[0070] Therefore, the present invention provides a method that allows to combine two data sources, i.e. the one from the digital camera 7 and the one from the event camera 6 to reconstruct the inter-frame displacement of the digital camera 7 and the strain fields to allow for a temporal super-resolution of the cracking phenomena. The dedicated method fuses information extracted from the imaging sensory system 3 to track the crack displacement and reconstruct the strain fields to allow a high spatio-temporal resolution analysis of the cracking phenomena on the coupon 2.

[0071] Figure 3 schematically shows an application example of the present system 1 on a stable crack propagation test, wherein the coupon 2 is a double cantilever beam 2 suitable for the estimation of the mode I interlaminar fracture toughness of reinforced plastics. It is noted that the abovementioned test is an international standard for estimating the energy required to have a crack propagating between material interfaces and it is part of the certification process of aerospace-grade composite systems. In this case, the digital camera 7 is configured to record, via generation of the plurality of digital images DI, the loading step of the coupon 2 while a synthetic event stream ES generated by the event camera 6 is reconstructed by the electronic processing resources 4 from the outputs of the imaging sensory system 3. In particular, the events recorded by the outputs of the artificial vision system 5 are represented using a frame-based visualization (in particular, 2D), where the event occurrence is rendered over a white canvas; in this way, starting from the analysis of the processed data, originated from the outputs of the imaging sensory system 3, it is possible to monitor the crack phenomena on the coupon 2 for the whole duration of the test.

[0072] The present solution described herein offers several advantages.

[0073] In particular, the present system provides a multi-camera approach to enhance the analysis of material cracking processes using a digital camera, namely a DLSR camera, and an event camera, namely a neuro-m orphic vision sensor or camera. Furthermore, the present invention provides an approach to increase the spatio-temporal resolution ofvision-based cracking monitoring by fusing the information coming from frame-based sensors, i.e. the digital camera, and event cameras.

[0074] Furthermore, the present invention provides a vision-based method to determine the stress intensity factor of brittle and quasi -brittle solids during quasi-static test using visual information with enhanced spatio-temporal resolution obtained by combining data extracted from frames (high spatial resolution) and events (high temporal resolution).

[0075] In addition, the present invention provides an alternative method with respect to the state of the art. In fact, the Applicant notes that alternative approaches, based on highspeed cameras, have been proposed for dynamic tests; however, the application of such methods in quasi-static experiments results unpractical for the considerable amount of data to be recorded and processed. Consequently, the present invention has the advantage of being more advantageous to apply when the cracking phenomena has to be analysed in detail.

Claims

CLAIMS1. A crack monitoring system (1) for monitoring deformation of a loaded material (2) to detect and track cracking phenomena in the loaded material (2);the crack monitoring system (1) comprises:- an imaging sensory system (3) configured to output an output that allows cracking phenomena in the loaded material (2) to be detected and tracked; and- electronic processing resources (4) in communication with the imaging sensory system (3) to receive and process the output thereof to detect and track cracking phenomena in the loaded material (2),the imaging sensory system (3) comprises:- an event camera (6) arranged to be directed towards an observed portion (2A) of the loaded material (2) to be monitored for cracking phenomena and operable to output an asynchronous stream of events (ES) triggered by changes in brightness in the observed portion (2A) of the loaded material (2); and- a digital camera (7) arranged to be directed towards the observed portion (2A) of the loaded material (2) to be monitored for cracking phenomena and operable to output a synchronous stream of a plurality of digital images (DI) of the observed portion (2 A) of the loaded material (2),wherein the electronic processing resources (4) are configured to:- synchronize the data output from the event camera (6) and the digital camera (7); - operate the event camera (6) and the digital camera (7);- receive and record the asynchronous stream of events (ES) from the event camera (6) and the synchronous stream of the plurality of digital images (DI) from the digital camera (7); and- process the asynchronous stream of events (ES) and the synchronous stream of the plurality of digital images (DI) to detect and track a cracking phenomenon in the observed portion (2A) of the loaded material (2).

2. The crack monitoring system (1) according to claim 1, wherein the electronic processing resources (4) are further configured to:- fuse the data from the asynchronous stream of events (ES) from the event camera (6) and the synchronous stream of the plurality of digital images (DI) from the digital camera (7) to reconstruct the strain fields to allow a high spatio-temporal resolution analysis of the cracking phenomena;- estimate the position of the point on the coupon (2) and a propagation speed of the crack on the coupon (2) based on the fused outputted data and the reconstructed strain fields; and- determine a stress intensity factor and a fracture toughness related to the coupon (2) based on the estimated position of the crack tip and the propagation speed of the crack on the coupon (2).

3. The crack monitoring system (1) according to claim 2, wherein, in order to fuse the data from the asynchronous stream of events (ES) and the synchronous stream of the plurality of digital images (DI), the crack monitoring system (1) is configured to:- process the received asynchronous stream of events (ES) to track the propagation over time of the crack on the loaded material (2) and output data thereof;- process the received synchronous stream of the plurality of digital images (DI) to determine a deformation field of the loaded material (2) and output data thereof;- fuse the processed data from the asynchronous stream of events (ES) and the synchronous stream of the plurality of digital images (DI) to carry out an analysis on the cracking phenomena and its evolution on the loaded material (2).

4. The crack monitoring system (1) according to any one of the preceding claims, wherein the event camera (6) is a neuromorphic vision sensor or camera and the digital camera (7) is a digital single-lens reflex, DSLR, camera.

5. The crack monitoring system (1) according to any one of the preceding claims, wherein the asynchronous stream of events (ES) is a tridimensional depiction representing the evolution over time of events (E) forming the asynchronous stream of events (ES),and wherein each event (E) comprises coordinates and a timestamp representative of a change in brightness in the observed portion (2A) of the loaded material (2).

6. The crack monitoring system (1) according to any one of the preceding claims, wherein each digital image of the synchronous stream of the plurality of digital images (DI) is a three-coloured channel frame acquired in a corresponding time instant (t).

7. The crack monitoring system (1) according to any one of claims 1-6, wherein the electronic processing resources (4) comprise a synchronization module (9) designed to operate the event and digital cameras (6, 7) so as to trigger the acquisition by the event and digital cameras (6, 7) at the same time instant and synchronize the output data from both cameras (6, 7).

8. The crack monitoring system (1) according to any one of claims 1-7, wherein, in order to process the received synchronous stream of the plurality of digital images (DI), the electronic processing resources (4) are configured to apply digital image correlation, DIC, techniques to the digital images of the synchronous stream of the plurality of digital images (DI) to determine a deformation field of the loaded material (2) and to track the crack evolution on the loaded material (2).

9. The crack monitoring system (1) according to claim 8, wherein, in order to process the received asynchronous stream of events (ES), the electronic processing resources (4) comprise an event-based feature tracking module (10)to process the asynchronous stream of events (ES) to track the visual features on the deformation field reconstructed by processing the synchronous stream of the plurality of digital images (DI) and a specialized module configured to compute the crack field displacement during the quick crack propagation process by applying event-based imagining velocimetry techniques based on event-based optical flow from the stream of events (ES).

10. The crack monitoring system (1) according to claim 9, wherein, in order to fuse the asynchronous stream of events (ES) and the synchronous stream of the plurality of digital images (DI) to detect and track a cracking phenomenon in the observed portion (2A) of the loaded material (2), the electronic processing resources (4) are configured to combine the outputs generated from processing the asynchronous stream of events (ES) and the synchronous stream of the plurality of digital images (DI) to reconstruct the evolution of the crack deformation formed on the loaded material (2).

11. The crack monitoring system (1) according to claim 10, wherein the electronic processing resources (4) are further configured to determine the stress intensity factor, the fracture toughness and the crack propagation velocity based on the determined data fromthe asynchronous stream of events (ES) and the synchronous stream of the plurality of digital images (DI).

12. The crack monitoring system (1) according to any one of the preceding claims and further comprising a mounting plate (8) configured to support and carry the imaging sensory system (5) so that the event and digital cameras (6, 7) are positioned and oriented to observe the same observed portion (2A) of the loaded material (2).

13. Computer program product loadable in and executable by electronic processing resources (4) of a crack monitoring system (1) according to any one of claims 1-12, the computer program product comprising instructions which, when the program is executed by the electronic processing resources (4), cause the crack monitoring system (1) to operate according to any one of claims 1-12.17

Images