METHOD FOR CALIBRATING A DIGITAL IMAGE CORRELATION SYSTEM
A screen-based digital image correlation calibration method and system address the bulkiness and complexity of mechanical target movement by using digital image analysis, offering a compact and ergonomic solution for precise calibration.
Patent Information
- Application Number
- FR2022014340
- Authority / Receiving Office
- FR · FR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-12-22
- Publication Date
- 2025-01-10
- Estimated Expiration
- 2042-12-22
AI Technical Summary
Existing digital image correlation systems for mechanical tests are bulky, complex, and not ergonomic, requiring cumbersome mechanical devices for target movement during calibration, which complicates the calibration process.
A method and system utilizing a screen to display a target at multiple positions, combined with a camera and processing unit to acquire and determine movements, allowing calibration through digital image analysis without mechanical movement devices.
Provides a compact, ergonomic, and accurate calibration system with reduced space requirements and improved ergonomics, enabling precise measurement uncertainty determination.
Smart Images
Figure 00000010_0000 
Figure 00000011_0000 
Figure 00000011_0001
Abstract
Description
Title of the invention: METHOD FOR CALIBRATING A DIGITAL IMAGE CORRELATION SYSTEM Technical field
[0001] The invention relates to the field of monitoring mechanical tests by digital image correlation. It relates in particular to a method for calibrating a digital image correlation system. Prior art
[0002] Digital image correlation (DIC) is a known technique for measuring the deformation of specimens during mechanical tests. This technique uses one or more cameras to measure object displacements.
[0003] Thus, when it is applied to a test piece subjected to deformation, for example under the effect of a hydraulic jaw, it makes it possible to monitor and quantify this deformation via the acquisition of digital images of the test piece in question.
[0004] More precisely, while the test piece is subjected to one or more forces which cause its deformation (i.e. during a test), a camera (for a monovision measurement) or two cameras (for a stereovision measurement) acquire digital images of the test piece (or part of the test piece) at a regular time interval. The displacement, in the images acquired over time, of points of interest on the surface of the test piece, then makes it possible, via a known correlation algorithm, to quantify its deformation.
[0005] As an example, [Fig.l] shows a mechanical testing device 101 on which a digital image correlation system 103 is mounted. The mechanical testing device 101 comprises a hydraulic jaw 105 which makes it possible to crush a test piece (not shown) placed at location 107.
[0006] The digital image correlation system 103 comprises a camera 109 (for example a CCD camera from the English Charge Coupled Device) for acquiring digital images and a processing unit 111 (for example a computer) for recovering these images and processing them (i.e. implementing the algorithm) in order to determine the deformation of the test specimen. Furthermore, in the example shown, a lighting system 113 makes it possible to illuminate the test specimen to optimize the quality of the acquired digital images.
[0007] In the state of the art, the calibration of a CIN system is based on the comparison of the displacement of an object measured, on the one hand, by the CIN system to be calibrated and, on the other hand, by a control physical measurement system. The difference between the two measurements of the same displacement makes it possible to know an error of measure attributed to the CIN system.
[0008] In particular, with regard to the calibration of CIN systems intended for measuring the deformation of test pieces in the context of mechanical tests on materials having an aeronautical application, standard NF EN 4861 specifies a procedure to be followed for the calibration of CIN systems.
[0009] More precisely, according to this standard, a target having a pattern suitable for the CIN, placed in front of the camera, must follow a series of predetermined movements exactly. A measurement by CIN is carried out on this target in order to determine the different movements and the comparison of the result obtained from the CIN system with the actual displacement value (which is known in advance), leads to determining a measurement uncertainty of the CIN system.
[0010] [Fig.2] shows an example of target 201 prescribed by standard NF EN 4861 to carry out such a calibration as well as a series of movements 203 (symbolized by the arrows between the different positions of the target) also prescribed by the standard for the same application.
[0011] In this context, it is imperative to be able to move, very precisely, the target(s) used during calibration. Thus, it is for example possible to use manual or motorized systems allowing a target to be moved on one or more axes in order to carry out a series of chosen movements.
[0012] As an example, [Fig. 3] shows a target moving device 301 which comprises a first motorized moving stage 303 (for example having micrometric precision) aligned along a first axis and a second motorized moving stage 305 (for example also having micrometric precision) aligned along a second axis orthogonal to the first axis. The second stage 305 is fixed on the first stage 303 and a set of targets 307 of different sizes is mounted on the whole so that it is possible to move these targets in a plane with the movement precision of the stages.
[0013] Such a device is both bulky, heavy and not very ergonomic (i.e. complex to use and handle). Summary of the invention
[0014] The present invention provides a solution to these drawbacks.
[0015] Thus, one objective of the invention is to reduce the size and complexity of the device used to move a target during the calibration of a digital image correlation system.
[0016] To this end, the invention according to a first aspect relates to a method for calibrating a digital image correlation system, said digital image correlation system comprising at least one camera configured to the acquisition of a plurality of digital images of an object, and a processing unit configured to determine at least one displacement of said object from the plurality of digital images acquired by said camera,
[0017] said method being characterized in that it comprises:
[0018] - the display, on a screen, of a target at a plurality of positions determined on said screen;
[0019] - the acquisition, by the camera, for each position of the target, of digital images of said target; and,
[0020] - the determination, from digital images of the target, of movements associated with each change in position of said target.
[0021] The method according to the invention may comprise one or more of the following characteristics, taken in isolation from one another or in combination with one another:
[0022] - the method further comprises comparing the determined displacements with predetermined displacement values, so as to determine at least one measurement uncertainty value of the digital image correlation system.
[0023] - a displacement corresponds to a translation of the target in the plane of the screen, to a translation of the target normally to the plane of the screen, a rotation of the target on itself or a combination of several of these movements.
[0024] - the digital image correlation system is configured to measure the de formation of a test piece.
[0025] - the digital image correlation system comprises a single camera.
[0026] The invention according to a second aspect also relates to a calibration system of a digital image correlation system, said digital image correlation system comprising at least one camera configured to acquire a plurality of digital images of an object, and a first processing unit configured to determine at least one displacement of said object from the plurality of digital images acquired by said camera,
[0027] said calibration system comprising:
[0028] - a screen configured to display a target at a plurality of determined positions;
[0029] - at least one camera of the digital image correlation system configured to acquire, for each position of the target, digital images of said target; and,
[0030] - a second processing unit configured to determine, from the naked images target's merics, movements associated with each change in position of said target,
[0031] and being configured to implement the method according to the first aspect.
[0032] The system according to the invention may comprise one or more of the characteristics following, taken individually or in combination with each other:
[0033] - the first processing unit and the second processing unit are one and same processing unit, preferably a computer.
[0034] - the system further comprises a mobile support for holding the screen, preferably an articulated arm.
[0035] Finally, the invention according to a third aspect also relates to an assembly comprising a digital image correlation system comprising at least one camera configured to acquire a plurality of digital images of an object, and a first processing unit configured to determine at least one movement of said object from the plurality of digital images acquired by said camera, and a calibration system according to the second aspect.
[0036] Further, in a particular embodiment, the assembly may be configured to be mounted on a mechanical testing device. Brief description of the drawings
[0037] The invention will be better understood with the aid of the following description, given solely by way of example and with reference to the appended drawings in which:
[0038] [Fig.l] is an illustration of an embodiment of a mechanical testing device having mounted thereon a digital image correlation system;
[0039] [Fig.2] is an illustration of a target and a target displacement series as prescribed by a standard for calibrating a digital image correlation system;
[0040] [Fig.3] is an illustration of a target moving device for calibrating a digital image correlation system;
[0041] [Fig.4] is a step diagram of a method for calibrating a digital image correlation system according to an embodiment of the invention;
[0042] [Fig.5] is a schematic representation of a method for calibrating a digital image correlation system according to an embodiment of the invention; and,
[0043] [Fig.6] is a schematic representation of types of movements of a target carried out for a method of calibrating a digital image correlation system according to an embodiment of the invention. Description of the embodiments
[0044] With reference to [Fig.4] and [Fig.5], we will now describe an embodiment of a method for calibrating a digital image correlation system according to the invention.
[0045] The calibration method 401 described in the following applies to a cor- digital image relation such as the system 103 described with reference to [Fig.l].
[0046] In particular, as described above, the digital image correlation system comprises at least one camera 109 (for example one camera for a so-called monovision system or two cameras for a so-called stereovision system) which is configured to acquire a plurality of digital images of an object.
[0047] The camera may be, for example, a CCD camera, i.e. a camera equipped with a CCD sensor which corresponds to a matrix of photodiodes. This system is compatible with any type of camera used for CIN in the visible spectrum.
[0048] The object that the camera 109 images may be, for example, a test piece. In this case, the digital image correlation system is configured to measure the deformation of a test piece, typically, in the context of mechanical testing.
[0049] Step 403 of method 401 consists of displaying, on a screen 501, a target 201 at a plurality of determined positions. The target may be, for example, a geometric figure comprising black areas and white areas such as the target 201 shown in more detail in [Fig.2]. The determined positions designate distinct areas of the screen 501 on which the target 201 is displayed.
[0050] The different positions taken by the target have been determined in advance and knowing the movements made between each position. The display can be controlled by a processing unit, such as for example a computer, such as the processing unit 503 described later with reference to step 407 or by a processing unit separate from the latter.
[0051] Furthermore, a displacement can be quantified (i.e. in distance) thanks to a priori knowledge of the distance separating the pixels which make up the screen 501. We speak of center distance, which can, for example, be expressed in micrometers, and which is data provided by the manufacturer of the screen 501 with a determined tolerance.
[0052] Depending on the characteristics of the screen, a correspondence between a “digital” displacement in pixels and the physical distance of the pixels of the screen is established and ultimately makes it possible to arrive at a known displacement of the target in the plane of the screen.
[0053] Step 405 consists of the acquisition, by the camera 109, for each position of the target, of digital images of the target 201. This is indeed the camera(s) (the non-limiting example shown only involves a single camera) of the CIN system since the calibration carried out is that of said CIN system.
[0054] [Fig. 5] illustrates an example of movement of a target in which an enlargement of a portion of the target 201 is presented for two distinct positions 201a and 201b of the target 201. The enlargement in question reveals pixels 505 of the screen 501, the center distance 509 of which is known, and also reveals the change in color (black or white) of certain pixels of the screen 501 following the movement of the target 201 symbolized by the arrow 507.
[0055] Step 407 consists of determining, from the digital images of the target, displacements associated with each change in position of said target. The processing of the images is carried out by the processing unit 503 which recovers the images acquired by the camera 109 and determines the associated displacements with the same algorithm as that used by the digital image correlation system.
[0056] Furthermore, the processing unit 503 may be a separate processing unit or the same processing unit as the processing unit 111 of the CIN system. It may be, for example, a computer.
[0057] The screen 501, the camera 109 (there may be several cameras) and the processing unit 503 (111) are therefore included in a calibration system 511 of a digital image correlation system which also includes the camera 109 and, potentially, the processing unit 503.
[0058] Furthermore, the assembly formed by the CIN system and the calibration system can thus be mounted on the same mechanical testing device.
[0059] Finally, in the non-limiting embodiment described in [Fig.4], step 409 consists of comparing the displacements determined by the processing unit 503 with predetermined displacement values, so as to determine at least one measurement uncertainty value of the digital image correlation system.
[0060] In other words, it is a matter of comparing the previously known displacements of the target with those obtained by the algorithm of the CIN system and establishing, from this, a measurement error.
[0061] For example, it may be determined that the measurement error is equal to 20 micrometers from a comparison showing a difference of 20 micrometers between the previously known displacement value(s) and the value(s) obtained by the CIN system algorithm.
[0062] The method and the calibration system described above have the particular advantage of providing a system having a small footprint and good ergonomics for calibrating a CIN system.
[0063] Furthermore, the use of a screen, which can be small and flat, allows for a large number of configurations for the location of the system and avoids inaccuracies related to the potential existence of play in a mechanical displacement system.
[0064] Thus, in a particular embodiment of the calibration system 511, the latter further comprises a mobile support for holding the screen such as for example an articulated arm which makes it possible to position the screen in front of the camera at a time chosen for calibration and then to remove it to carry out measurements.
[0065] Furthermore, the use of a screen also makes it possible to change the shape of the target infinitely since this only depends on the display command received by the screen which can be modified according to the wishes of an operator.
[0066] Finally, as can be seen in [Fig.6], the use of a screen also makes it possible to make the target perform more complex movements than a translation of the target in the plane of the screen as illustrated in [Fig.5].
[0067] Indeed, [Fig.6] shows four configurations 601a, 601b, 601c and 601d which correspond respectively to the display of the target 201 in an initial position 603, to the display of the target 201 in a position which corresponds to a translation of the target in the plane of the screen (as in the case of [Fig.5]), to a translation of the target 201 normally to the plane of the screen and to a rotation of the target 201 on itself.
[0068] The last two configurations simulate more complex movements and can be achieved respectively, by a change in scale of the displayed target for the first and by interpolation methods (supported by the processing unit controlling the display) for the second.
[0069] The use of a screen therefore also makes it possible to simply carry out a calibration potentially based on complex movements of a target.
[0070] Furthermore, a move can also be a combination of several of the three moves listed above.
[0071] Thus, advantageously, the fact that a movement is carried out digitally makes possible a large number of types of movements without involving an increase in bulk linked to the use of one or more devices for each movement. In addition, this type of movement also avoids the existence of a functional clearance typical of the use of a mechanical assembly having different degrees of freedom.
Claims
Claims
1. Method (401) for calibrating a digital image correlation system (103), said digital image correlation system (103) comprising at least one camera (109) configured to acquire a plurality of digital images of an object, and a processing unit (111) configured to determine at least one displacement of said object from the plurality of digital images acquired by said camera (109), said method being characterized in that it comprises: - displaying (403), on a screen (501), a target (201) at a plurality of determined positions; - acquiring (405), by the camera (109), for each position of the target (201), digital images of said target (201); and, - determining (407), from the digital images of the target (201), displacements associated with each change in position of said target (201).
2. The method (401) of claim 1, further comprising comparing (409) the determined displacements with predetermined displacement values, so as to determine at least one measurement uncertainty value of the digital image correlation system (103).
3. Method (401) according to claim 1 or 2, in which a displacement corresponds to a translation of the target (201) in the plane of the screen (501), to a translation of the target (201) normally to the plane of the screen (501), to a rotation of the target (201) on itself or to a combination of several of these displacements.
4. The method (401) of any preceding claim, wherein the digital image correlation system (103) is configured to measure deformation of a specimen.
5. A method (401) according to any preceding claim, wherein the digital image correlation system (103) comprises a single camera (109).
6. A system (511) for calibrating a digital image correlation system (103), said digital image correlation system (103) comprising at least one camera (109) configured to acquire a plurality of digital images of an object, and a first processing unit (111) configured to determine at least a movement of said object from the plurality of digital images acquired by said camera (109), said calibration system comprising: - a screen (501) configured to display a target (201) at a plurality of determined positions; - at least one camera (109) of the digital image correlation system (103) configured to acquire, for each position of the target (201), digital images of said target (201); and, - a second processing unit (503) configured to determine, from the digital images of the target (201), movements associated with each change of position of said target (201), and being configured to implement the method (401) according to any one of the preceding claims.
7. The system (511) of claim 6, wherein the first processing unit (111) and the second processing unit (503) are a single processing unit, preferably a computer.
8. System (511) according to claim 6 or claim 7, further comprising a movable support for holding the screen, preferably an articulated arm.
9. An assembly comprising a digital image correlation system (103) comprising at least one camera (109) configured to acquire a plurality of digital images of an object, and a processing unit (111) configured to determine at least one displacement of said object from the plurality of digital images acquired by said camera (109), and a calibration system (511) according to any one of claims 6 to 8.
10. An assembly according to claim 9, configured to be mounted on a mechanical testing device.