Non-contact deformation monitoring system

Abstract

A deformation monitoring system for measuring a deformation of an object, the system comprising: a first camera arranged to capture a first image of the object; a second camera arranged to capture a second image of the object, the second camera being in a known direction relative to the first camera; and a controller arranged to cause the deformation monitoring system to perform the steps of: imaging the object with the first camera at a first time; imaging the object with a second camera at a first time; determining a first location of a first measurement point on the object at the first time and a first location of a second measurement point on the object at the first time based on the imaging of the first and second cameras at the first time; imaging the object with the first camera at a second time different from the first time; imaging the object with a second camera at a second time; determining a location of a first measurement point on the object at the second time and a location of a second measurement point on the object at the second time based on imaging of the first and second cameras at the second time; and determining a deformation of the object based on the locations of the first and second measurement points at the first and second times.

Description

Technical Field

[0001] The invention relates to a deformation monitoring system and a deformation monitoring method. Background Art

[0002] All physical objects deform in shape when subjected to forces, temperature changes, residual stress relaxation, etc. For example, the components of a car's suspension (e.g. pushrods, wishbones, drive shafts, etc.) will all bend, twist, compress, and stretch as the car drives along the road. During the design phase of these same components, they may have been placed in a test rig that deforms the component by simulating the forces it will experience during use. By measuring these deformations, engineers are able to optimize the design of the component to reduce cost and weight while ensuring that the component will not fail during use.

[0003] As another example, when a material (e.g., a carbon composite) is manufactured, a sample of the manufactured material may be placed in a testing machine that deforms the sample by applying a force (typically tensile or compressive.) Measuring the resulting deformation allows a property of the material (e.g., modulus) to be measured as part of a quality control or other testing process.

[0004] Whatever the object is, and whether the deformation is produced on a test bench or in actual use, it is often necessary to measure the deformation of the object accurately.

[0005] Traditionally, deformation has been measured using “contact” sensors such as digital test indicators, linear variable differential transformers, and strain gauges. Such devices are typically only able to measure a single location on an object, and can only measure deformation along a single axis. Therefore, obtaining a detailed understanding of the deformation may require mounting multiple such sensors on the object. This often makes the use of such sensors time-consuming and expensive, and the associated wiring can quickly become cumbersome and prone to failure. In addition, each sensor that contacts the object will exert a force on the object, and will therefore affect the deformation being measured to some extent. In some cases, the force applied by the sensor can significantly affect the deformation of the object, making it difficult to obtain accurate deformation data. Summary of the invention

[0006] According to a first aspect of the present invention, there is provided a deformation monitoring system for measuring the deformation of an object, the system comprising: a first camera arranged to capture a first image of the object; a second camera arranged to capture a second image of the object, the second camera being in a known direction relative to the first camera; and a controller arranged to cause the deformation monitoring system to perform the following steps: imaging the object with the first camera at a first time; imaging the object with the second camera at a first time; determining a first location of a first measurement point on the object at the first time and a first location of a second measurement point on the object at the first time based on the imaging of the first camera and the second camera at the first time; imaging the object with the first camera at a second time different from the first time; imaging the object with the second camera at a second time; determining a location of the first measurement point on the object at the second time and a location of the second measurement point on the object at the second time based on the imaging of the first camera and the second camera at the second time; and determining the deformation of the object based on the locations of the first measurement point and the second measurement point at the first time and the second time.

[0007] With this arrangement, the system can more easily and accurately measure deformation of the object. Although the system can work with a single camera, it may be difficult to detect or measure movement of the measurement point along the optical axis of the camera (also known as out-of-plane movement). Determining the location of the measurement point on the object based on the imaging of the first camera and the second camera can include a photogrammetry process, etc.

[0008] Furthermore, the system can monitor more than two measurement points simultaneously, allowing a more in-depth analysis to be performed. Each camera can image at least two, optionally more than two, measurement points on the object. The measurement points can all be captured within a single image taken by each camera.

[0009] The measurement point may be a feature or symbol on the object. The measurement point may be a corner, edge, or hole in the object. Alternatively, the measurement point may be a symbol drawn on the object, such as a circle or a cross. The measurement point may also be referred to as a measurement location. An image of the object captured by the camera may include the measurement point, and the controller may identify the measurement point within the image. In this way, the location of the measurement point may be determined.

[0010] Determining the location of the measurement point may include identifying a symbol or feature on the object.

[0011] Determining the location of the measurement point may include using a Hough transform. Using a Hough transform may allow the measurement point to be located more easily in the image, particularly where features or symbols defining the measurement point may be distorted, for example by imaging at an oblique angle.

[0012] The controller may be arranged to determine the position of one or more holes, edges and / or corners of the object, and the determination of the location of the measurement point may be based at least in part on the determined one or more locations of the one or more holes, edges and / or corners. This may help locate the measurement point and may reduce the likelihood of misidentifying the measurement point.

[0013] Determining the location of the measurement point may include determining the location in an object-based coordinate system or in another non-natural coordinate system. Although the system may use a coordinate system based on the measurement system (e.g., with the origin at the camera), using a different coordinate system may allow for simpler deformation calculations and, therefore, may allow the system to be used with lower latency. The non-natural coordinate system may also be more useful to the user, requiring less downstream processing to obtain meaningful results.

[0014] The system may also include an artificial light source arranged to illuminate the object. The light source may improve the image quality. The signal-to-noise ratio of the image data may also be increased. This may allow the controller to more accurately locate the measurement point. In addition, the light source may also improve the accuracy of the measurement, thereby reducing the exposure time required for the image, allowing the location of the measurement point to be determined at a more precise time.

[0015] The artificial light source may be arranged to produce monochromatic light, and optionally the first and / or second camera may comprise a bandpass filter arranged to allow light of the wavelength of the artificial light source to pass. This may allow the image to be taken with reduced interference, such as radiation generated by the object due to its temperature or other ambient light. This may in particular improve the measurement accuracy in the event of any changes in the spectrum of the ambient light or the light radiated by the test sample (e.g. due to heating of the sample). Otherwise, such changes in light conditions may lead to inaccurate measurements, due to chromatic aberrations of the lens resulting in apparent shifts of the measurement position in the image and thus inaccurate measurement data.

[0016] The artificial light source may be arranged to produce polarized light. Optionally, the first camera and / or the second camera may include a polarization filter arranged orthogonally to the polarization direction of the artificial light source. This may reduce glare from the light source detected by the camera, as only non-specular reflected light from the light source may be detected. The reduction in glare may reduce the chance of the controller misidentifying the measurement point, improving the reliability of the measurement.

[0017] The light source may be a strobe light. Optionally, the strobe light may be synchronized with the first camera and the second camera so that the object is illuminated during imaging of the object with the first camera and the second camera. In this way, the strobe light may be "over-rated", meaning that high instantaneous power may be used for the light, providing a high level of illumination and allowing images to be produced with short exposure times. This may allow for more precise timing of the image of the object, allowing for greater deformation measurement accuracy. The overall latency of the device may also be reduced, which may be particularly beneficial in closed-loop control of measurement and imaging. Motion blur may also be reduced, improving measurement accuracy.

[0018] The first camera and the second camera may be arranged in a fixed orientation. In other words, the optical axes of the first camera and the second camera may be fixed relative to each other so that the position and orientation of the second camera relative to the first camera are known and constant; for example, the cameras may be connected to each other or mounted on a common platform. Alternatively, the second camera may be movable relative to the first camera, and the orientation of the second camera relative to the first camera may be measured and transmitted to the controller, for example, by an encoder.

[0019] The first camera and the second camera may each have a focal length, and the focal length may be the same at the first time and the second time. This may allow for a simpler calculation of the position of the measurement location from the image.

[0020] The controller may be arranged to attempt to determine the locations of the first and second measurement points at a first time (i.e., in an image acquired at the first time), and based on a failure to determine, may be arranged to instruct the first camera and / or the second camera to acquire a further image of the object based on the failure. In other words, if the controller is unable to identify one or more measurement points in an image from the first or second camera at a first time, the controller may instruct the camera to acquire a further image, which may then be analyzed to determine the locations of the measurement points. Thus, the controller may be arranged to determine whether to instruct the first and / or second camera to acquire at least one further image of the object at a first time based on the degree of success of the determination of the locations of the measurement points.

[0021] The controller may be arranged to attempt to determine the location of the measurement point in the image from the first camera and in the image from the second camera, and to determine the degree of success separately for each image.

[0022] Based on a failure to determine at a first time the location of the measurement point in images from only one of the first and second cameras, the controller may be arranged to instruct only the corresponding camera of the first and second cameras to acquire at least one further image.

[0023] The controller may attempt to determine the locations of the measurement points in the further images, and based on the determination of the locations of all the measurement points in the further images, the instructions for acquiring the further images may be stopped. The controller may start measuring the object at this stage, for example measuring the distance between the measurement points. This may make the system more automated and reduce the burden on the user.

[0024] The controller may be arranged to cause the deformation monitoring system to acquire images of the measurement points at a further time after the second time until the position of at least one of the measurement locations cannot be determined. This may enable the system to act autonomously to continue measuring until the end of the experiment, for example due to sample failure or removal of the sample by the user.

[0025] The system may further comprise an accelerometer or a magnetometer, and the controller may be arranged to determine the orientation of the first camera and / or the second camera based on an output of the accelerometer or the magnetometer. The accelerometer may determine the orientation of the deformation monitoring system relative to a horizontal plane, while the magnetometer may determine the orientation of the deformation monitoring system relative to magnetic north. Thus, the deformation monitoring system may be used to measure objects in an environment and compare the deformation to a known reference frame.

[0026] The deformation monitoring system may further comprise an alignment system arranged to project alignment marks near the object. The alignment system may comprise a laser arranged to project the alignment marks near the object, i.e. generally towards the object. The alignment marks may represent the limits of the measurement volume of the first camera and the second camera and / or the centre of the measurement volume. The alignment system may be fixed to the cameras to move with the cameras and may thus help a user to set up the deformation monitoring system to ensure that the measurement points are within the measurement volume and will be in the images acquired by the cameras.

[0027] The controller may be arranged to perform a verification function, the verification function comprising: identifying a verification object; measuring a distance between measurement points on the verification object; and comparing the measured distance with a known value of the distance between the measurement points on the verification object. The controller may be arranged to repeat the measurement and comparison steps of the verification function until the location of at least one of the measurement locations cannot be determined. The verification function may improve the accuracy and reliability of the results and may be performed in an automated manner, thereby reducing the burden on the user. Repeating the measurements may further improve the verification function by allowing the user to freely move the verification object for a more complete verification process.

[0028] The controller may be arranged to record at least one of the images from the first camera and the second camera, the locations of the measurement points and the determined deformations.The recording function may allow the system to be used as a standalone system and transfer the data to a separate system at a later time.

[0029] The controller is arranged to identify an identifier from the image from the first camera and / or the second camera. The identifier may be unique. The identifier may be on or associated with the object. The identifier may be alphanumeric or a barcode or QR code. Data from the identifier may be acquired and may be recorded with the measurement data and / or the image. This may assist a user in data processing at a later time.

[0030] According to a second aspect of the present invention, a method for monitoring deformation of an object is provided, the method comprising: providing an object having a first measurement point and a second measurement point; imaging the first measurement point and the second measurement point with a first camera; imaging the first measurement point and the second measurement point with a second camera; determining the locations of the first measurement point and the second measurement point based on the imaging of the first camera and the second camera; deforming the object to produce a deformed object having deformation measurement points; imaging the first deformation measurement point and the second deformation measurement point with the first camera; imaging the first deformation measurement point and the second deformation measurement point with the second camera; determining the locations of the first deformation measurement point and the second deformation measurement point based on the imaging of the first camera and the second camera; and determining the deformation of the object based on the locations of the first measurement point and the second measurement point and the locations of the first deformation measurement point and the second deformation measurement point.

[0031] In this way, a non-contact solution for measuring the deformation of an object is provided.

[0032] The method may also include drawing or attaching a symbol representing the measurement point on the object. This may allow the controller to more easily determine the measurement point.

[0033] Imaging the first measurement point and the second measurement point with the first camera may include taking a single image, and / or imaging the first measurement point and the second measurement point with the second camera may include taking a single image. Typically, the cameras may be arranged to provide images, wherein each image contains all measurement points. In particular, where there are two cameras, each of the two cameras may be arranged such that each measurement point is visible in each image from the two cameras.

[0034] The method may include positioning the cameras so that the side of their image sensors with the most pixels is aligned with the long axis of the test object.

[0035] The method may include positioning one of the cameras so that its optical axis is perpendicular to a plane that intersects the first measurement point and the second measurement point. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A test device comprising a deformation monitoring system according to the invention is shown;

[0037] Figure 2a and 2bis a schematic diagram of a deformation monitoring system according to the present invention;

[0038] Figure 3a shows a test specimen for use with a deformation monitoring system according to the present invention;

[0039] Figure 3b A verification object for use with a deformation monitoring system according to the invention is shown;

[0040] Figures 4a to 4g Different possible appearances of the measurement points are shown;

[0041] Figure 5 is a flow chart describing a method for monitoring deformation according to the present invention; and

[0042] Figure 6 is a flow chart illustrating an alternative method of monitoring deformation according to the present invention. DETAILED DESCRIPTION

[0043] The device of the invention can measure the position, movement and deformation of an object at multiple moments. At each moment, it measures the position of one or more measurement points on the object. These positions are measured in three dimensions in a coordinate system local to the device. This coordinate system is called the natural coordinate system. Based on the measured positions of the measurement points, the device can calculate measurement data at each moment, such as:

[0044] The location of each measurement point;

[0045] The position of a measurement point relative to another measurement point;

[0046] The distance between a pair of measurement points;

[0047] The position and / or orientation of a line passing through two or more measurement points;

[0048] The position and / or orientation of a plane passing through three or more measurement points;

[0049] the position and / or orientation of a local coordinate system, wherein the positions of at least three measurement locations within the local coordinate system are known. The local coordinate system may be a local coordinate system of the object;

[0050] any change from one moment to another; and

[0051] • The position, motion and / or deformation of an object can be easily inferred from one or more of the above.

[0052] Figure 1A test apparatus 10 is shown that includes a test sample 12 that can be deformed by movement of a crosshead 18 that can apply a compressive or tensile force on the test sample 12. The test sample 12 has two measurement points: a first measurement point 14 and a second measurement point 16. Although the measurement points 14, 16 are shown as circles drawn on the surface of the test sample 12, it should be understood that alternative measurement points may be used, such as edges, corners, or holes in the test sample 12, and other symbols such as crosses and spots may be drawn on the surface. In some cases, features of the test sample 12 (e.g., edges and corners) that are not themselves measurement points may be used in determining the locations of the measurement points. These features may be referred to as indirect measurement points. The deformation monitoring system 100 is Figure 1 The arrangement may be used as an extensometer, arranged to observe the test sample 12 as it deforms due to movement of the crosshead 18.

[0053] Figure 2a and 2b is a schematic diagram showing part of the deformation monitoring system 100 .

[0054] The controller 102 is arranged to control the functions of the deformation monitoring system and may take any suitable form, such as a general purpose computer, one or more data processors with appropriate memory and I / O means, and / or an ASIC.

[0055] The deformation monitoring system 100 also has at least two cameras 104a, 104b, and may have one or more lights 104c. The lights and cameras may be collectively referred to as a camera and lighting assembly 104. The term camera as used herein refers to any image capture device that produces an image that can be analyzed to find measurement points. Similarly, the term "light" refers to any device that can emit electromagnetic radiation to improve the image captured by the camera.

[0056] One or both of the two cameras can utilize a monochrome sensor with 1440×1080 pixels. The camera(s) may also or alternatively be arranged to output a strobe output signal that is effective during image exposure and receive a trigger input signal that can be used to trigger the sensor to capture an image. The strobe output signal can be used to actuate a strobe light so as to synchronize camera exposure with light activation. A suitable image sensor is a Sony IMX 273 sensor capable of capturing a full frame at 276Hz. The sensor can be coupled with an 8mm focal length lens, a 470nm bandpass filter, and / or a polarization filter. An image sensor with a higher pixel resolution can be used to increase the measurement resolution of the device. An image sensor capable of capturing at a higher frame rate can be used to achieve a higher measurement data rate. Lenses with different focal lengths can be used to achieve different sizes, shapes, or positions of the measurement space.

[0057] Furthermore, the cameras may be coupled such that the strobe output of one camera may be received as a trigger input signal for the other camera(s) so that all cameras may synchronously capture images at the same time.

[0058] Light 104c is arranged to illuminate a measurement point on the test sample, such as Figure 1 The cameras 104a, 104b are positioned to observe measurement points on the test sample.

[0059] Lamp 104c can have an array of 32 LEDs that emit primarily blue light. A suitable LED is a Cree XP-E2 blue LED. Each LED can be coupled to a lens to produce a beam angle that uniformly illuminates the measurement volume of the device. The LEDs can be coupled to a polarizer element so that the emitted light is primarily polarized in a single direction. The LEDs can be divided into two or more groups, such as two groups of 16, so that the intensity of each group can be different, thereby achieving uniform illumination of all aspects of the test sample.

[0060] In the case where the light can be a strobe light and can be synchronized with the camera, the strobe outputs of the cameras can be combined by a logical "OR" function so that the light can be active as long as either camera is acquiring an image. The strobe signals can be connected so that the LED is only turned on when the combined strobe signal is active. This can have the effect of synchronizing the strobing of the light with the exposure cycle of the camera. The camera can be operated at a capture rate greater than 80Hz in order to avoid visible flickering of the light, which could otherwise be harmful to people with photosensitivity to strobe lights.

[0061] The cameras 104a, 104b are mounted so that they are held in fixed registration with each other by a mounting bracket or the like so that their extrinsic calibration parameters do not change after the calibration process has been performed. Prior to performing the calibration process, the focal length and focus distance of the camera lenses are fixed so that after performing the calibration process, the intrinsic parameters of the cameras do not change. In other embodiments, the cameras may simply have a known orientation and / or position.

[0062] The cameras 104a, 104b each have a field of view which is a frustum extending from the lens of the camera. Where the two frustums overlap, measurements can be made using the two cameras. The overlapping space can be referred to as the measurement space.

[0063] Some or all of the components of the system may be mounted on or in a common housing, which may also include a power source, such as a battery or a wired electrical input connector.

[0064] Each camera captures a sequence of two or more images and sends them to the controller. Image capture can be synchronized between all cameras so that all cameras capture images at the same moment in time. This ensures that the 3D measurements made from the images are not adversely affected by any movement of the test specimen relative to the cameras.

[0065] In order to keep the measurement latency of the device as low as possible, each image can be provided to the controller as soon as possible after it is captured. If this requires, for example, transferring the image data over a bus, the bus should have high bandwidth and low latency. In one example, the high bandwidth bus can be a MIPI-CSI bus with one or more lanes running at more than 1 Gbit / s.

[0066] Digital cameras contain image sensors that consist of a two-dimensional rectangular array of pixels, where the array typically has more pixels on one side than on the other. Since test specimens are often long and thin, it is advantageous to position the camera so that the side of the image sensor with the most pixels is aligned with the long axis of the test specimen. This maximizes the number of pixels along the length of the specimen, helping to maximize the measurement resolution of the device.

[0067] If the measurement location is generally contained within a plane (including along a line), or within multiple parallel planes, it may be advantageous to position one of the cameras so that its optical axis is perpendicular to that plane. This may make it easier to automatically detect the measurement location in images captured from that camera. It may also be easier for an operator to interpret images captured from that camera.

[0068] The cameras may be fitted with optical bandpass filters so that the light detected by the image sensor within each camera is substantially monochromatic. Without such filters, any changes in the wavelength of the light illuminating the sample may result in an apparent movement of the measurement site, resulting in inaccurate measurement data, due to chromatic aberrations of the lenses. Such changes in the wavelength of the light illuminating the test sample will typically occur if the ambient lighting changes (e.g., artificial lighting is turned on / off, curtains / blinds are opened / closed, a cloud passes in front of the sun, etc.), or if the temperature of the test sample changes, such that the amount and / or wavelength of light emitted by the test sample changes.

[0069] The lamp may be arranged to produce substantially monochromatic light. This may be used in place of, or in conjunction with, an optical bandpass filter for use with a camera to reduce the effects of chromatic aberration of the camera lens. If used with an optical bandpass filter, the nominal wavelength of the monochromatic light may be matched to the nominal bandpass wavelength of the optical filter.

[0070] The lamps may be arranged to produce polarised light and the cameras may be fitted with polarising filters. In this case, the polarising filters are aligned so that the light they pass is polarised in a direction substantially orthogonal to the polarisation direction of the lamp light. This reduces glare in the image caused by specular reflections of light from shiny test specimens, windows, transparent safety screens etc. Such glare could otherwise interfere with the appearance of the measurement location in the image, adversely affecting the accuracy of the measurement data.

[0071] The lights may be strobed so that they are on when any camera is exposing an image, but they may be off when no camera is exposing an image. Thus, the proportion of time that the lights are on, also known as the "duty cycle," may be less than 100%. A duty cycle of less than 100% reduces the power consumption of the device and / or allows the lights to be overdriven (operated at an intensity greater than their nominal rating). Overdriving the lights allows the camera exposure time to be reduced, thereby reducing measurement delays for the device and reducing measurement inaccuracies that might otherwise result from motion blur.

[0072] In the illustrated embodiment, the controller 102 is arranged to perform the following functions:

[0073] Camera & lighting control and image capture capabilities

[0074] Automatic operation function

[0075] Direction detection function

[0076] Measurement location detection function

[0077] Measurement function

[0078] Recording function

[0079] Verification function

[0080] Test sample label reading function

[0081] It should be understood that the controller 102 may have any one or more of these functions in appropriate combinations. For example, the controller may have the functions of controlling the camera and light, automatic operation, direction detection, measurement location detection, and measurement. However, any of the other three functions may also be performed by the controller, or any of the first five functions may be omitted from the controller.

[0082] To perform these functions, the controller 102 may include a processor, electronic memory, and input and output connections. Each function of the controller will be described in detail in the following sections.

[0083] The camera & light control and image capture functions of the controller configure each camera to capture a sequence of two or more images. The captured images are grouped into image sets, where each image set contains one image from each of at least two cameras. If the camera and lighting assembly has more than two cameras, ideally, an image set contains one image from each camera. In this way, the sequence of images captured by the cameras becomes an image set sequence. If the image capture is synchronized between all cameras, all images in an image set will be captured at the same moment in time.

[0084] As images are captured and grouped into image sets, the image sets are passed to other elements of the controller (e.g., those portions that perform measurement functions). To keep the device's measurement latency as low as possible, each image set is sent to other portions of the controller as quickly as possible after the image is received from the camera.

[0085] The camera and light control and image capture functions of the controller may also configure the cameras and lights appropriately, including but not limited to camera exposure time, camera image gain, camera frame rate, and light intensity. Camera and light settings that affect image exposure (e.g., camera exposure time, camera image gain, and light intensity) may be automatically changed so that optimal image exposure is achieved even if the appearance of the measurement location changes over time or between test samples. Image exposure is optimal when the image area corresponding to the measurement location is exposed so that contrast is maximized, but without any overexposed pixels (i.e., no pixel in the area has an intensity value that is consistent with the maximum possible intensity value for the pixel). Achieving optimal image exposure may improve the signal-to-noise ratio of the image, thereby helping to maximize the accuracy and resolution of the measurement data.

[0086] The automatic operation function of the controller can automatically control all parts of the device so that it does not require any interaction from the operator to perform measurements on a batch of one or more test samples. In summary, it automatically determines when the test sample is placed within the measurement space of the device by automatically identifying the measurement locations on the test sample. It then determines when to start measuring the locations of these measurement locations and continues to measure them until the device automatically determines that the testing of the test sample is complete, at which point it stops the measurement process and prepares to detect the next test sample.

[0087] You can Figure 5300 in the flowchart. In step 302, the initial state performs the camera & light control and image capture functions. After receiving a set of captured images, the automatic operation function detects the measurement location visible in the captured image (within the measurement space of the device) using the measurement location detection function in step 304. It will be determined in step 306 whether the measurement location is successfully detected, and based on this determination, the controller can proceed to the measurement state of step 308, otherwise it will capture the next set of available images and try to detect the measurement location again. In some cases, the controller can determine whether the measurement point is stationary, and further images can be acquired until it is determined that the measurement point is stationary. Based on determining that the measurement point in the image acquired in step 302 is stationary, the controller can move to the measurement state at step 308.

[0088] Once in the measurement state 308, the controller will repeatedly perform the measurement function on one image set after another in steps 308 and 310 until it is automatically determined that the measurement process is complete in step 312. The measurement process can be determined to be complete based on, for example, but not limited to, the following criteria:

[0089] The measurement location is no longer within the measurement space;

[0090] The 3D position of the measurement location can no longer be measured. This can happen, for example, if the measurement location is not visible to more than two cameras; or

[0091] The test specimen has ceased to deform.

[0092] When the measurement process is determined to be complete, it returns to the initial state 302 where camera & lighting control and image capture are performed.

[0093] The direction detection function determines a transformation that can be used within the measurement function of the controller to transform the measurement data from a natural coordinate system to a measurement coordinate system. The natural coordinate system can be centered on the device and can be the coordinate system in which the location of the measurement point is initially determined. The measurement coordinate system can be a coordinate system that is more relevant to the user, such as a coordinate system aligned with features of the measurement sample and / or with the North Pole and / or with a vertical axis.

[0094] The transformation may be an identity transformation, in which case the measured coordinate system is the same as the natural coordinate system. Alternatively, one or more sensors may be used to detect the orientation of the device relative to an external reference, such as vertical, horizontal, or magnetic north. Such sensors may include one or more accelerometers or magnetometers. Given the detected orientation of the device, it is easy to determine a transformation that transforms the measurement data from the natural coordinate system to a measured coordinate system that is defined relative to one or more external references (e.g., horizontal, vertical, or magnetic north).

[0095] The measurement location detection function of the controller is arranged to automatically find all measurement locations visible in the images of the image set. For each measurement location found, its position in each image in which it is visible is determined. The position in the image can be represented as a coordinate in the two-dimensional image sensor array of the camera.

[0096] If the measurement locations have a known appearance, or if markers with a known appearance have been applied to the measurement locations, they can be found and their image locations determined by techniques well known to those skilled in the art of image processing (e.g., edge detection, template matching, generalized Hough transform, Harris corner detector, etc.). For example, if the measurement locations appear as circles, the Hough transform can be used to find the circles and determine the image coordinates of the center of each circle. Such techniques can be used to easily find measurement locations with a variety of appearances, including but not limited to circular spots, rings, lines, crosshairs, bullseyes, checkers, and "spot" patterns.

[0097] Alternatively or additionally, if there are other identifiable features on the test sample in addition to the plotted measurement points, and the position of the measurement location relative to these identifiable features is known, the measurement location can be found by first finding the other identifiable features and then inferring the position of the measurement location based on the position of the identifiable features. Examples of such identifiable features include edges, corners, holes, etc. Such identifiable features can be easily found using techniques well known to those skilled in the art of image processing (e.g., Canny edge detector, Harris corner detector, generalized Hough transform, etc.).

[0098] Figure 3a A test sample 200 is shown with known measurement points 202a, 202b spaced a distance D1 apart and located on a centerline 208, equidistant from a centerpoint 206. The centerline 208 and centerpoint 206 can be easily calculated from the detected positions of corner features 204a-d on the sample, and thus the positions of the measurement points 202a, 202b can be inferred from the detected positions of the corner features 204a-d.

[0099] In some test scenarios, it may be advantageous to check whether a measurement point has been stationary for a predetermined amount of time before deciding that the measurement point has been successfully found. For example, this can prevent the device from inappropriately entering a measurement state when a test specimen is being loaded into a universal testing machine (UTM) before a test is performed.

[0100] A range of different shapes and symbols can be applied to the test specimen to represent the measurement points. Examples of this are shown in Figure 4a to Figure 4g Shown in. Figure 4a Solid circles are shown, Figure 4b A hollow circle or ring is shown, Figure 4c The dashed line is shown, Figure 4d shows a cross, Figure 4e shows a ring with a solid circle inside, Figure 4f shows two squares touching diagonally at a corner, also called a checkerboard, Figure 4g A speckle pattern is shown. Any of these shapes can be used to represent the measurement location.

[0101] The measurement function of the controller receives a sequence of image sets, and for each image set, it generates measurement data such as those described above. In addition to or in place of any of the aforementioned measurement data, other measurement data may be generated. Such other measurement data may include one or more of the following:

[0102] Axial strain between two measurement locations located along the longitudinal axis of the test specimen

[0103] Transverse strain between two measurement locations located across the transverse axis of the test specimen

[0104] Strain in any direction, such as 45 degrees to the longitudinal axis

[0105] Displacement of the test specimen

[0106] Rotation of the test specimen (around any axis)

[0107] Bending of the sample

[0108] Distance between two measurement locations

[0109] Displacement of one or more measurement locations

[0110] The following steps outline the methodology used to generate the measurement data:

[0111] 1) Determine at least one three-dimensional position (in a natural coordinate system) of each measurement location at a time corresponding to the time when an image in the image set was captured.

[0112] 2) Calculating measurement data based on the determined three-dimensional position(s) of the measurement location(s), for example by comparing the relative positions of the measurement location(s) at different times.

[0113] 3) Using the transformation determined by the direction detection function, transform the measurement data into the measurement coordinate system.

[0114] 4) Output the measurement data through the measurement data output system.

[0115] 5) If the controller includes a recording function, the image set and measurement data may be transferred to the portion of the controller that provides the recording function.

[0116] The recording function includes a storage element, such as a disk drive, onto which various types of data are recorded. The recorded data may include one or more of the following:

[0117] A sequence of one or more images from one or more cameras;

[0118] a sequence of one or more sets of images; and

[0119] A sequence of one or more measured data values.

[0120] Storage devices may be volatile or non-volatile.

[0121] The controller's verification feature allows the operator to confirm that the device is measuring accurately. To use this feature, the operator will Figure 3b The verification object shown is placed within the measurement volume of the device. The verification object 250 has two or more verification marks 252a, 252b located at known positions relative to each other. For example, the verification object 250 has two verification marks at a known distance D2 of 100.000 mm apart.

[0122] The verification element of the controller measures the distance D2 between the two verification marks 252a, 252b. The measured distance D2 is then compared with the known distance between the verification marks to calculate an error value. In order to determine whether the device is measuring accurately, the error value is compared with a predefined acceptable error value range. If the error is within the acceptable range, the device is considered to be measuring accurately.

[0123] In one approach, a controller receives a sequence of image sets, and for each image set, it performs the following steps:

[0124] 1) Detect the location of the verification mark. This can be achieved using the verification object detection function which is essentially the same as the measurement location detection function of the controller.

[0125] 2) Determine the three-dimensional position of each verification mark. This can be achieved using a verification function that is essentially the same as the measurement function of the controller.

[0126] 3) Compare the measured relative positions of the verification marks to the known relative positions to calculate an error value. For example, if the verification marks are known to be 100.00mm apart and the measured spacing is 100.010mm, the error is calculated to be 0.010mm. The error value can also be expressed as a relative error by dividing by the known distance. Relative errors can be expressed as a percentage. In this example, an error of 0.010mm can be expressed as a relative error of 0.01%.

[0127] 4) Compare the error value with a predefined range, such as + / -0.1 mm or + / -0.1% relative error. If the error value is within the acceptable range, the device is considered to be operating accurately.

[0128] 5) Report to the operator whether the device is operating accurately. Error values ​​and / or relative errors can also be reported to the operator.

[0129] The system can utilize Figure 6 The control scheme shown in FIG. operates with a large degree of autonomy, as described in more detail below. Once in the verification state, it will repeatedly perform the verification function on one image set after another until it autonomously determines that the verification process is complete. The verification process can be determined to be complete based on criteria such as, but not limited to, the following:

[0130] Verify that the mark is no longer within the measurement space

[0131] The 3D position of the verification mark can no longer be measured. This can happen, for example, if the measurement location is not visible to more than two cameras. When the verification process is judged to be complete, it returns to the initial state of performing camera & light control and image capture.

[0132] The controller may also have a sample tag reading function. The sample tag reading function may automatically read information encoded in a tag associated with a test sample. The tag may consist of a handwritten or machine-printed alphanumeric sequence or other encoding, such as a barcode or QR code, that encodes information associated with the test being performed, such as a unique serial number, batch, or test identifier. The tag may be attached elsewhere within the frustum of at least one of the test sample or the camera.

[0133] Any information read from the sample tags can be added to the measurement data.

[0134] In one embodiment, the label is read using readily available methods and tools for optical character recognition (OCR), barcode reading, or QR code reading.

[0135] The deformation monitoring system 100 also has a measurement data output system 110. This provides an interface for sending measurement data to one or more external systems. The measurement data may also include information read from a sample tag. The measurement data may be sent in the form of an analog signal or a digital data stream. The interface may consist of more than one channel so that multiple measurement data values ​​can be sent simultaneously. The interface may employ a wired or wireless connection.

[0136] The deformation monitoring system 100 may have an alignment system 106 that helps an operator position the test specimen and / or the device so that the measurement point is within the measurement volume of the device. This may be achieved by making various aspects of the measurement volume visible to the operator, where these aspects may include:

[0137] ·Measure the extent of a space, such as the top, bottom, or corners of a space

[0138] ·Measure points or lines in space, such as center points or center lines

[0139] In one embodiment, these aspects are made visible to the operator and aligned with one or more aspects of the measurement space by projecting a visible marker (such as a dot or line) to the operator. Such a marker may be projected by a laser line or laser dot projector.

[0140] The deformation monitoring system 100 may have a user interface 108 including buttons, consoles, voice recognition, gesture recognition, indicators, display screens, touch screens, audio, voice output, etc. These elements interact with the controller to allow the operator to perform one or more of the following operations:

[0141] Monitor the status of a device, such as on / off, measuring / not measuring, etc.

[0142] Check whether the device is measuring accurately

[0143] Monitor measurement data values

[0144] Configure or control various aspects of the device

[0145] One or more elements of the user interface may be provided remotely by a separate device such as a computer, tablet or smartphone. In this case, the apparatus may provide an interface that allows connection between the controller and the remote elements of the user interface. The interface may be wired or wireless, such as Wi-Fi, Ethernet, Bluetooth or a cellular or satellite data connection.

[0146] Figure 6 A control method 350 for providing automated functionality to the deformation monitoring system 100 is shown. The system may begin by operating a camera and lighting device to acquire an image at step 352. The controller then detects a measurement point 354 in the image. At step 356, it is determined whether the controller has found the measurement point. If the measurement point has been identified, the controller may begin measuring parameters, such as the distance between the measurement points, at step 358. The system may then repeatedly capture images and measure characteristics of the image at steps 360 and 358 until it is determined at step 362 that the measurement is complete.

[0147] If the system cannot detect the measurement point at step 354, the system may enter a verification procedure and may attempt to detect a verification object at step 364, such as Figure 3b 250 shown in FIG. 3. If the system determines that the verification object is within the captured image, a decision is made at step 366 to begin the verification function of the controller, which has been described above. The verification function is performed at step 368 and may include taking a plurality of images using the camera and lighting system at step 370. Once the verification is complete, as determined at step 372, the system may return to the initial step and may be ready to receive a sample for testing.

[0148] Therefore, the deformation monitoring system according to the embodiment of the present invention has the function of solving one or more, and in some embodiments, all of the following technical problems:

[0149] Inaccurate measurement data

[0150] oIf the object is shiny (e.g. metal), reflections will be picked up by the camera and interfere with the measurement

[0151] o If the object is hot (such as part of a jet engine), it may glow red, yellow, or even white-hot. This glow can interfere with the measurement

[0152] oIf an object is behind a window (e.g. a security screen), the camera may pick up reflections from the window and interfere with the measurement

[0153] o Out-of-plane motion (movement of the object towards or away from the device) can result in inaccurate measurement data. For example, consider a device with a single camera. If the object moves closer to the camera, it will appear larger in the image (due to perspective principles). However, if the object remains the same distance from the camera, but is heated and expands (due to thermal expansion), it will also appear larger in the image. Such a device cannot determine whether the object has actually deformed or has moved towards / away from the camera. Therefore, if there is motion towards or away from the camera, the device cannot accurately measure deformation.

[0154] oIt cannot be easily checked whether the device is measuring accurately.

[0155] Complex and slow to use

[0156] o Specific markers (e.g. circular spot markers) must be applied at the desired measurement locations on the object. These markers define the measurement locations and are designed to be easily detected in the camera image. Applying these markers can be a time-consuming and complex process.

[0157] o User input is required to configure and operate the equipment, such as manually selecting measurement locations, manually starting and stopping measurement functions, etc. This may require an interactive user interface (e.g., a GUI on a touch screen), thereby increasing complexity. In the case of batch testing (i.e., sequential testing of multiple objects), the requirement for operator interaction can significantly increase the time to test each object.

[0158] o Provide a device that does not require a precise distance ("operating distance") or precise alignment with the point(s) on the object being measured.

[0159] oIn most test scenarios, it is advantageous to measure in a coordinate system, such as the local coordinate system of the object being measured. However, many devices measure in the local coordinate system of the device. Further steps must be taken to transform the measurement data to the desired coordinate system. These steps add complexity and are time consuming, making the device complex and slow to use.

[0160] oWhen testing multiple objects (e.g. batch testing), records must be kept describing which set of measurement data relates to which object. This is usually achieved by tagging the object with a unique number or code. This tag must then be entered into the system that captures the measurement data. This task is usually performed by an operator reading the tag and then manually entering it through the device's user interface. This process is time consuming and prone to errors.

[0161] Not suitable for closed-loop control

[0162] oSome test scenarios require that the device forms part of a closed-loop control system, so that the loading force applied by the test bench varies automatically depending on the currently measured deformation. For the control loop to operate successfully, the measurement data generated by the device must have a sufficiently high data rate and the measurement delay must be low enough and constant enough. Some devices are not able to generate measurement data at a sufficient data rate or with an acceptable delay and are therefore not suitable for use in closed-loop control systems.

[0163] In some cases, the camera may acquire an image of only one measurement point. In this case, the deformation may be measured using a known reference (e.g., a fixed point) to which the test specimen may be anchored.

[0164] In addition, the deformation measurement system can have more than two cameras. A greater number of cameras can improve the reliability of the system and allow a larger measurement space and allow a larger range of data to be acquired.

[0165] A deformation measurement system having more than two cameras or arranged to operate with only one measurement point may have substantially similar features to the deformation measurement system described above.

[0166] Further specific examples of the present disclosure are set forth in the following clauses:

[0167] A. A method for determining a deformation of an object, the method comprising:

[0168] providing an object having a first measurement point and a second measurement point;

[0169] Imaging the first measurement point and the second measurement point using a first camera;

[0170] Imaging the first measurement point and the second measurement point with a second camera;

[0171] Determine the locations of the first measurement point and the second measurement point based on the images of the first camera and the second camera;

[0172] deforming the object to produce a deformed object having deformation measurement points;

[0173] Imaging the first deformation measurement point and the second deformation measurement point with a first camera;

[0174] Imaging the first deformation measurement point and the second deformation measurement point with a second camera;

[0175] Determining locations of a first deformation measurement point and a second deformation measurement point based on imaging by the first camera and the second camera; and

[0176] Based on the locations of the first and second measuring points and the locations of the first and second deformation measurement points, a deformation of the object is determined.

[0177] B. A method according to clause A, wherein the measurement points are symbols drawn onto the object.

[0178] C. A method according to clause A, wherein the measurement point is a hole, an edge or a corner of the measurement object.

[0179] D. A method as described in clause B, wherein determining the location of the measurement point includes determining the location of one or more holes, edges and / or corners of the object.

[0180] E. A method according to any preceding clause, wherein determining the locations of the first and second measurement points and / or the first and second deformation measurement points comprises using a Hough transform.

[0181] F. A method according to any preceding clause, wherein determining the locations of the first and second measurement points and / or the first and second deformation measurement points comprises determining the locations in an object-based coordinate system.

[0182] G. A method according to any preceding clause, further comprising illuminating the object with an artificial light source.

[0183] H. A method according to clause G, wherein the artificial light source produces monochromatic light.

[0184] I. A method according to clause H, wherein the first camera and / or the second camera comprises a bandpass filter arranged to allow light of a wavelength of the artificial light source to pass therethrough.

[0185] J. A method according to clause G, H or I, wherein the artificial light source produces polarized light.

[0186] K. A method according to clause J, wherein the first camera and / or the second camera includes a polarization filter arranged orthogonally to the polarization direction of the artificial light source.

[0187] L. A method according to any of clauses G to K, wherein the light source is a strobe light.

[0188] M. A method according to clause L, wherein the strobe light is synchronized with the first camera and the second camera so that the object is illuminated.

[0189] N. A method according to any preceding clause, wherein imaging the first measurement point and the second measurement point with the first camera comprises capturing a single image; and / or

[0190] wherein imaging the first measurement point and the second measurement point with the second camera comprises capturing a single image; and / or

[0191] Wherein, imaging the first deformation measurement point and the second deformation measurement point with the first camera comprises taking a single image; and / or

[0192] The imaging of the first deformation measurement point and the second deformation measurement point by using the second camera includes capturing a single image.

[0193] O. A method according to any preceding clause, wherein the first camera and the second camera are maintained in a fixed orientation.

[0194] P. A method according to any preceding clause, wherein the first camera and the second camera each have a focal length, and when imaging the measurement point and the deformation measurement point, the focal lengths are the same.

[0195] Q. A method according to any preceding clause, wherein the first camera and the second camera are synchronized so that imaging of the first measurement point and the second measurement point with the first camera and imaging of the first measurement point and the second measurement point with the second camera are simultaneous; and / or

[0196] The first camera and the second camera are synchronized, so that imaging of the first deformation measurement point and the second deformation measurement point by the first camera and imaging of the first deformation measurement point and the second deformation measurement point by the second camera are simultaneous.

[0197] R. A method according to any preceding clause, wherein the controller attempts to determine the locations of the first and second measurement points and, based on the degree of success of the determination, instructs the first and / or second cameras to acquire further images of the first and second measurement points.

[0198] S. A method according to any preceding clause, further comprising determining an orientation of the first camera and / or the second camera based on output of an accelerometer or a magnetometer.

[0199] T. A deformation monitoring system comprising:

[0200] a first image capture device arranged to capture a first image of the object;

[0201] a second image capture device arranged to capture a second image of the object, the second image capture device being at a known orientation relative to the first image capture device; and

[0202] A controller is arranged to cause the deformation monitoring system to perform the method of any one of clauses A to S.

Claims

1. A deformation monitoring system for measuring deformation of an object, the system comprising: a first camera arranged to capture a first image of the object; a second camera arranged to capture a second image of the object, the second camera being at a fixed orientation relative to the first camera; as well as A controller is arranged to cause the deformation monitoring system to perform the following steps: Imaging the object with the first camera at a first time; imaging the object with the second camera at the first time; Determine, based on the imaging of the first camera and the second camera at the first time, a first location of a first measurement point on the object at the first time and a first location of a second measurement point on the object at the first time; imaging the object with the first camera at a second time; imaging the object with the second camera at the second time; Determine, based on the imaging of the first camera and the second camera at the second time, a location of the first measurement point on the object at the second time and a location of the second measurement point on the object at the second time; as well as Based on the locations of the first measurement point and the second measurement point at the first time and the second time, a deformation of the object is determined.

2. The deformation monitoring system according to claim 1, wherein: Determining the location of the measurement point includes identifying a symbol or feature on the object.

3. The deformation monitoring system according to claim 2, wherein: The controller is arranged to determine the locations of one or more holes, edges and / or corners of the object, and wherein the determination of the locations of the measurement points is based at least in part on the determined locations of the one or more holes, edges and / or corners.

4. A deformation monitoring system according to any one of the preceding claims, further comprising an artificial light source arranged to illuminate the object.

5. The deformation monitoring system according to claim 4, wherein: The artificial light source is arranged to generate monochromatic light.

6. The deformation monitoring system according to claim 5, wherein: The first camera and / or the second camera comprises a band pass filter arranged to allow light of a wavelength of the artificial light source to pass.

7. A deformation monitoring system according to claim 4, 5 or 6, wherein: The artificial light source is arranged to generate polarized light.

8. The deformation monitoring system according to claim 7, wherein: The first camera and / or the second camera comprises a polarization filter arranged orthogonally to a polarization direction of the artificial light source.

9. The deformation monitoring system according to any one of claims 4 to 8, wherein: The light source is a strobe light.

10. The deformation monitoring system according to claim 9, wherein: The strobe light is synchronized with the first camera and the second camera so as to illuminate the object during imaging of the object by the first camera and the second camera.

11. A deformation monitoring system according to any one of the preceding claims, wherein: The first camera and the second camera each have a focal length, and the controller is arranged such that the focal length is the same at the first time and the second time.

12. A deformation monitoring system according to any one of the preceding claims, wherein: The controller is arranged to attempt to determine the locations of the first and second measurement points at the first time and determine whether to instruct the first and / or second cameras to acquire at least one further image of the object based on a degree of success in determining the locations of the measurement points.

13. The deformation monitoring system according to claim 12, wherein: Based on a failure to determine the location of the measurement point at the first time, the controller is arranged to instruct the first camera and / or the second camera to acquire a further image of the object before determining the location of the measurement point.

14. The deformation monitoring system according to claim 12 or 13, wherein: The controller is arranged to attempt to determine the location of the measurement point in the image from the first camera and in the image from the second camera at the first time and to determine the degree of success separately for each image.

15. The deformation monitoring system according to claim 14, wherein: Based on a failure to determine the location of the measurement point in images from only one of the first camera and the second camera at the first time, the controller is arranged to instruct only the corresponding camera of the first camera and the second camera to acquire at least one additional image before determining the location of the measurement point.

16. A deformation monitoring system according to any one of claims 12 to 15, wherein: Based on successfully determining the location of the measurement point at the first time, the controller is arranged to stop instructing the first camera and / or the second camera to acquire further images of the object and start determining the location of the measurement point.

17. A deformation monitoring system according to any one of the preceding claims, wherein: The controller is arranged to cause the deformation monitoring system to acquire images of the measurement points at further times after the second time until the location of at least one of the measurement locations cannot be determined.

18. A deformation monitoring system according to any preceding claim, further comprising an accelerometer or a magnetometer, the controller being arranged to determine the orientation of the first camera and / or the second camera based on an output of the accelerometer or magnetometer.

19. A deformation monitoring system according to any preceding claim, further comprising an alignment system arranged to project alignment marks adjacent the object.

20. The deformation monitoring system according to claim 19, wherein: The alignment system comprises a laser arranged to project an alignment mark proximate to the object.

21. A deformation monitoring system according to claim 19 or 20, wherein: The alignment marks indicate limits of a measurement space of the first camera and the second camera and / or a center of the measurement space.

22. A deformation monitoring system according to any one of the preceding claims, wherein: The controller is arranged to perform a verification function comprising: Identify and verify objects; measuring distances between measurement points on the verification object; and The measured distance is compared to a known value of the distance between the measurement points on the verification object.

23. The deformation monitoring system according to claim 22, wherein: The controller is arranged to repeat the measuring and comparing steps of the verification function until the location of at least one of the measurement locations cannot be determined.

24. A deformation monitoring system according to any one of the preceding claims, wherein: The controller is arranged to record at least one of: images from the first camera and the second camera, the location of the measurement point, and Determined deformation.

25. A deformation monitoring system according to any one of the preceding claims, wherein: The controller is arranged to identify an identifier from images from the first camera and / or the second camera.

26. A method for monitoring deformation of an object, the method comprising: providing an object having a first measurement point and a second measurement point; Imaging the first measurement point and the second measurement point with a first camera; Imaging the first measurement point and the second measurement point with a second camera; Determine the locations of the first measurement point and the second measurement point based on the imaging of the first camera and the second camera; deforming the object to produce a deformed object having deformation measurement points; Imaging the first deformation measurement point and the second deformation measurement point with the first camera; Imaging the first deformation measurement point and the second deformation measurement point with the second camera; Determine the locations of the first deformation measurement point and the second deformation measurement point based on the imaging of the first camera and the second camera; as well as Based on the locations of the first and second measuring points and the locations of the first and second deformation measurement points, a deformation of the object is determined.

Images