Computer-implemented method for measuring the elongation of an object and associated device

Abstract

The present disclosure relates to a computer-implemented method for measuring an elongation of an object (100), the method comprising the following steps:acquiring an image (E1) comprising an object (100) in a deformed position and an imprint (101) connected to the object (100), the imprint (101) being non-deformable relative to the object (100), the object (100) comprising a reference mark (102); andfrom the acquired image and dimensions of the object (100) at rest, determining (E2) an elongation () of the object (100) by tracking the reference mark (102) of the object (100) relative to the imprint (101), the dimensions of the object (100) at rest being defined relative to the reference mark (102) and relative to the imprint (101). Abstract Figure: Figure 1

Description

Title of the invention: Computer-implemented method for measuring the elongation of an object and associated device Technical field

[0001] The present disclosure relates to computer methods for image analysis, in particular image analysis methods for monitoring deformation of a support garment. STATE OF THE ART

[0002] Certain computer image analysis methods make it possible to measure a deformation of an object. To do this, a visible marker in an image is associated with the object, for example by printing the marker on the object. Indeed, since the marker deforms at the same time as the object, it makes it possible to follow its deformation. These methods could be used in particular in the medical field in order to monitor treatment for a chronic illness causing swelling of a limb, for example edema. In this case, to treat the illness, a patient wears a compression garment that covers the affected limb. However, to monitor the illness, a doctor measures a change in the circumference of the affected limb with a tape measure, a scanner or by water displacement. Thus, to implement this measurement, the tools used by the doctor could be substituted by monitoring the marker that deforms at the same time as the compression garment.

[0003] However, these methods have limitations in terms of accuracy. Indeed, the tracking can be distorted depending on the point of view from which the image was taken. Indeed, when the camera lens is not in front of the pattern, there can be a perspective effect that distorts the pattern in the image. Therefore, the distortion caused by the perspective effect can significantly affect the accuracy of the measurements. In addition, all the landmarks must be captured by a sequence of images, which requires moving the camera around the object to scan its surface or rotating and tilting the object in front of a fixed camera. These methods also require either the presence of a second person (as is the case for 3D scanners) or a specific and calculated positioning of the camera relative to the object.

[0004] They are then unsuitable for medical use. Indeed, certain pathologies require regular monitoring and justify that it be carried out by the patient himself, that is to say without systematically going to a health establishment, in order to improve the treatment of the pathology but also the costs generated by this care.

[0005] Some solutions which aim to choose a pattern less sensitive to this disturbance or to use an image processing method in order to re-establish the image in perspective in the measurement plane have been proposed. However, these solutions only make it possible to limit this effect without eliminating it and partially address the aforementioned drawbacks.

[0006] Another solution would be to keep the camera lens facing the pattern during image acquisition. However, such a solution may prove difficult to implement when the object is moving during acquisition, particularly when the user of the compression garment moves his leg or arm. This solution lacks precision and may prove incompatible in the case of treating pathologies such as edema. Furthermore, it does not address all of the aforementioned drawbacks.

[0007] Another solution consists of using a property of the cross-ratio (better known by the English acronym "cross-ratio"). Indeed, it is a mathematical equation which is invariant in projection. That is to say that a cross-ratio of four points aligned in a perspective image is equal to a cross-ratio of the same four points aligned in an image without perspective effect. However, although this method is effective for fixed objects, it becomes difficult to apply in the context of an object which is deformed due to the variations in lengths between these four points which are induced by the deformations of the object. Indeed, these variations add unknowns in the equality and make its resolution difficult. GENERAL STATEMENT

[0008] An aim of the invention is therefore to improve the monitoring by image analysis of the deformation of the object.

[0009] To this end, according to a first aspect of the present disclosure, a computer-implemented method is proposed for measuring an elongation of an object, the method comprising the following steps: acquiring an image comprising an object in a deformed position and an imprint connected to the object, the imprint being non-deformable relative to the object, the object comprising a reference mark; and from the acquired image and the dimensions of the object at rest, determination of an elongation of the object by tracking the reference frame of the object in relation to the imprint, the dimensions of the object at rest being defined in relation to the reference frame and in relation to the imprint.

[0010] Thus, by defining an undeformable reference from which the object stretches, we change the mode of deformation of the object to a mode in which an equality of the cross-ratios becomes an equation which no longer includes only one unknown. Consequently, it is possible to determine the elongation from this equation that does not depend on the perspective effect. Thus, an error factor in the measurement of the elongation is removed, which has the effect of improving the monitoring by image analysis of the deformation of the object.

[0011] The method also makes it possible to choose the points which define the deformable segments in the cross-ratio at any location of the object visible on the image.

[0012] It also allows the measurement to be carried out with a minimum of data. In fact, the measurement is implemented only from an image of a portion of the object.

[0013] It may be provided that the method comprises a determination of a segment in the image, the determination of the elongation of the object taking place in the segment.

[0014] The object can therefore be meshed to take into account the variations in elongation within the object. Thus, the precision is further improved, as is the determination of the dimension of the object as well as its volume.

[0015] It may further be provided that the imprint comprises two markers, the segment being defined between the two markers.

[0016] It may also be provided that the two markers are fiduciary markers, barcodes, or QR codes.

[0017] Thus, it is possible to distinguish the segments in a simple and reliable manner in the context of a computer implementation.

[0018] It may be provided that the method comprises a step of measuring, from the elongation, a dimension of the object.

[0019] It is therefore possible to measure a dimension of the object for a set of states between a resting state and a breaking state. In addition, it is possible to carry out this measurement precisely, that is to say that the precision of the measurement is of the order of the pixel. In addition, the repeatability of the measurement is improved. The measurement error in the image is also between 2% and 4%. The repeatability is also improved. In the case where the object is a compression garment, it is also possible for the user to carry out the measurement himself, without for example the intervention of a doctor. Therefore, the treatment of a pathology such as an edema is improved by regular monitoring of the patient made possible thanks to the method.

[0020] It may also be provided that the method comprises the following steps: from the elongation, measurement of a circumference of the object; and from the measurement, determination of a volume of the object.

[0021] It is therefore possible to measure a volume of the object for a set of states between a resting state and a breaking state. In addition, it is possible to carry out this measurement precisely, that is to say that the precision of the measurement is of the order of the pixel. The measurement error in the image is also between 2% and 4%. In addition, the repeatability of the measurement is improved. In addition, it is possible to carry out the measurement with a minimum of data. Indeed, the volume is calculated from a portion of the object visible in the image instead of performing a complete scan of the object to determine its volume.

[0022] It may also be provided to determine from the measurement of the circumference of the object a pressure field of the object.

[0023] Thus, thanks to the pressure field it is for example possible to evaluate the effectiveness of the support garment, in particular for the treatment of a pathology such as edema, in real time and in a dynamic environment.

[0024] It may also be provided to determine a normalized pressure field, for example expressed between zero as the lowest measured value and one as the highest measured value.

[0025] This solution is particularly advantageous when it is difficult, or even impossible, to determine the value of the pressure in Pascal, or even in mmHg (millimeter of mercury), which is a unit commonly used for compression garments, but one still wishes to have an idea concerning the pressure distribution within the garment, for example to check whether the stocking applies strong pressure to the distal end and progressively weaker pressure towards the proximal end.

[0026] It may be provided that the imprint comprises two distinctly connected patterns on the object.

[0027] In this case, the use of two prints makes it possible to position the points aligned on the prints. Thus, the length of a single segment is unknown when using the cross-ratio, which further improves the method.

[0028] It may be provided that the method is implemented by a terminal and in which the object is a support garment, the mark is formed of two lines and the imprint is formed of a strip.

[0029] Indeed, the method is suitable for a medical application, for example in monitoring pathologies such as edema, lymphedema, deep vein thrombosis or any other pathology causing swelling or atrophy of a limb. In this case, the patient wears a compression garment on the arm or leg to treat these pathologies. He can then measure with a terminal, for example his mobile phone, his compression garment independently, without intervention from a doctor, to monitor the evolution of his pathology. Consequently, the garment is quickly and easily analyzed by the phone which is capable of implementing the method. Thus, the method is implemented without the use of expensive and uncommon equipment and it makes it possible to make the monitoring of pathologies by a patient accessible.The measurement is precise, that is to say that the precision of the measurement is of the order of the pixel, unlike a measurement taken by a tape measure whose precision is of the order of ten millimeters, or even a few centimeters. The measurement error in . The image is also between 2% and 4%. The repeatability of the measurement is also improved. In addition, the process improves the speed of the measurement compared to other existing methods. In fact, the time saving is estimated at 10 minutes on average compared to the use of a tape measure. In addition, the process improves the safety of the measurement because it is implemented without contacting a measuring tool with the arm or leg. It is also possible for the user to carry out the measurement themselves, for example without the intervention of a doctor. Therefore, the treatment of a pathology such as edema is improved by regular monitoring of the patient made possible thanks to the process.

[0030] It may further be provided that the strip is between the two lines, or the two lines are adjacent.

[0031] Thus, it is possible to apply the method to any markers present on the image.

[0032] It may also be provided that the imprint is formed of a first strip and a second strip, a first line being taken on one end of the first strip and a second line being taken on one end of the second strip opposite the end of the first strip.

[0033] It may also be provided that the strip comprises markers, the markers being aligned on the strip.

[0034] Thus, the measurement accuracy can also be further improved in the case of a medical application.

[0035] According to a second aspect of the present disclosure, there is provided an assembly comprising: an object comprising a marker; and an imprint connected to the object, the imprint being non-deformable relative to the object, the assembly being configured to allow determination of an elongation of the object by tracking a reference point of the object relative to the imprint.

[0036] According to a third aspect of the present disclosure, there is provided a device for measuring a deformation of an object, the device comprising: an assembly comprising: an object comprising a reference mark; and an imprint connected to the object, the imprint being non-deformable relative to the object, the assembly being configured to determine an elongation of the object by tracking a reference mark of the object relative to the imprint; and a terminal configured to: acquire an image comprising the assembly; and from the acquired image and dimensions of the object at rest, determine an al lengthening of the object by tracking the object's reference frame relative to the imprint, the dimensions of the object at rest being defined relative to the reference frame and relative to the imprint. DESCRIPTION OF FIGURES

[0037] Other characteristics, aims and advantages will emerge from the following description, which is purely illustrative and non-limiting, and which must be read in conjunction with the appended drawings in which:

[0038] [Fig.l] schematically represents a device for measuring a deformation of a support garment worn by a user;

[0039] [Fig.2a], [Fig.2b], [Fig.2c] and [Fig.2d] schematically illustrate several geometric configurations of the imprint and the marker;

[0040] [Fig.3a], [Fig.3b], [Fig.3c] and [Fig.3d] schematically illustrate markers respectively on the imprint of [Fig.2a], [Fig.2b], [Fig.2c] and [Fig.2d];

[0041] [Fig.4] illustrates several types of markers;

[0042] [Fig.5] schematically illustrates a mode of implementation of a method of measurement of a deformation of an object;

[0043] [Fig.6a], [Fig.6b], [Fig.6c] and [Fig.6d] illustrate four aligned points po located on a footprint and a marker in the case of an object at rest; and

[0044] [Fig.7a], [Fig.7b], [Fig.7c] and [Fig.7d] illustrate four aligned points positioned on the imprint and the reference mark in the case where the object of [Fig.6a], [Fig.6b], [Fig.6c] and [Fig.6d] is deformed. DETAILED DESCRIPTION

[0045] A device 1 for measuring an elongation of a support garment 100 is illustrated as an example in [Fig.l].

[0046] The compression garment 100 is configured to be worn by a user 2 on a limb, for example an arm 21 and / or a leg 22, for example by fitting. It is suitable for a medical application, for example in monitoring pathologies such as edema, lymphedema, deep vein thrombosis or any other pathology causing swelling or atrophy of a limb, thanks to the compression force exerted by the garment on the limb affected by the pathology. The compression garment 100 is of substantially cylindrical geometry and is formed of a flexible fabric in order to be able to stretch the garment during fitting and to compress the limb when the garment is fitted.

[0047] The measuring device 1 is also suitable for other compression garments, for example a t-shirt or trousers whose shape fits several parts of the user 2. Furthermore, the device 1 is not limited by a type of fabric used (it can be a unidirectional or bidirectional textile fabric), nor by a fabric material (fiberglass, Kevlar, cotton, etc.). The device 1 is also suitable for other types of objects that have a Young's modulus, therefore that obey Hooke's law. It may be an object comprising an inflatable material, such as a balloon, a parachute or an inflatable structure. The device 1 may also be suitable for measuring an elongation of the skin of the user 2 in the context of monitoring an evolution of dimensions or volume of a limb, for example in the case of a potential inflammation of the tissues of a limb or monitoring a recovery of muscle mass of an atrophied or injured limb. It may also be suitable for measuring an elongation of a civil construction, a sail or a material by tensile testing in general.

[0048] The device 1 comprises an assembly 10 configured to follow the elongation of the support garment 100.

[0049] The assembly 10 comprises: - the 100 support garment; and - an imprint, formed of two strips 101a, 101b in this example, connected to the support garment 100.

[0050] The imprint can be connected to the compression garment by weaving. It is also possible to connect the imprint to the garment by other methods. For example, by laser beam by applying glue to the imprint and heating the imprint positioned on the garment by air convection. Other heat sources can be used, such as laser. The imprint can also be connected by screen printing.

[0051] The imprint is distinct from the garment 100. The imprint can be distinguished from the garment by its color. In the example in [Fig.l], the imprint is black and is distinguished from the white compression garment. Of course, the imprint can be any color, it can even be the same color as the garment but of a different shade, for example by different gray levels. Other visual properties of the garment can also be used to distinguish the imprint from the garment, by using filters or different types of light, for example by highlighting the imprint under ultraviolet or infrared light. It is also possible to make it distinct by varying the contrast, the imprint protruding from the garment.

[0052] The imprint is non-deformable relative to the garment 100. In other words, during an elongation of the garment, the imprint remains fixed while a fabric of the garment stretches. To do this, a Young's modulus of the imprint is greater than that of the measured object. It can be greater by up to a factor of ten compared to that of the material of the measured object and, in the example of the compression garment 100, by up to a factor of eight compared to the fabric of the garment. For example, the imprint can be in nylon, whose Young's modulus is between 2 GPa and 4 GPa, or polyester, whose Young's modulus is between 3 GPa and 5 GPa, while the compression garment is made of elastane, whose Young's modulus reaches 0.6 GPa. In the case where the Young's modulus of the print is close to the Young's modulus of the object, for example when the print is also in a fabric with a Young's modulus close to that of the garment, it is possible to stiffen the print by increasing a density of the fabric by a factor of ten compared to a density of the fabric of the garment, for example by embroidery. It is also possible to stiffen the fabric print by using a different weave than that used for the compression garment.According to another example, in the case of an object comprising an alloy or a ceramic, it is possible to stiffen the impression by subjecting it to heat treatment using quenching or tempering followed optionally by annealing.

[0053] Thus, in this example, when the compression garment 100 deforms, it is possible to see the bands 101a, 101b move apart or move closer together relative to a position of the bands 101a, 101b in a resting state of the compression garment 100. For example, when a volume of the user's 2 limb increases over time, the garment undergoes an elongation which will move the bands apart. It is then possible to visually follow a change in the volume of the limb relative to a resting position, for example when the garment is worn for the first time, by tracking a movement of the bands 101a, 101b on the garment 100.

[0054] The assembly 10 according to several embodiments has been illustrated in detail in [Fig.2a], [Fig.2b], [Fig.2c] and [Fig.2d]. In these figures, a portion of the support garment 100 is shown.

[0055] In the embodiment illustrated in [Fig.2a], the garment comprises a marker, formed by two lines 102 distinct from the strip 101. A strip 101 between the two lines is connected to the compression garment 100 instead of two strips as illustrated in [Fig.l]. In the embodiments illustrated respectively in [Fig.2b] and [Fig.2c], the two lines 102 are adjacent on the garment and they follow each other. They can also be positioned in front of the strip 101 or behind the strip 101. The marker can be produced by contrast ink which does not affect the elasticity properties of the garment, for example by tracing the lines 102 on the garment 101. Thus, the visual monitoring of the deformation is done by monitoring a displacement of the lines 102 relative to the strip 101.

[0056] The mark may also be a grid drawn on all or part of the garment 101. Of course, the mark may be any other geometric pattern that extends over all or part of the garment. Of course, the imprint may also be any other geometric pattern. For example, a cross, a triangle, a square. The imprint and the mark may also have different thicknesses and dimensions. on the object.

[0057] The marker can also be any element projecting from the garment or from an object in general. Indeed, in the embodiment illustrated in [Fig.2d] which is similar to that of [Fig.l], the marker is formed of two lines 102 facing each other and taken on one end of each of the bands 101a, 101b. Thus, the visual tracking is done in a manner similar to that described above in the example illustrated in [Fig.l].

[0058] Consequently, according to the aforementioned specific cases, the mark can be distinct from the imprint just as it can be confused with the imprint without departing from the scope of the present disclosure.

[0059] It may be provided that the marker is computer-generated, for example by superimposing the marker on an image of the object and the print obtained by image acquisition means. The marker may also be a combination of a portion visible on the garment and a portion generated by computer.

[0060] It may also be provided to combine several patterns of prints and markers for the same object. For example, on the same compression garment, a first portion of the garment may correspond to the embodiment of [Fig.2a] and a second portion of the garment may correspond to the embodiment of [Fig.2d].

[0061] The imprint may also comprise markers. In the embodiments illustrated in [Fig.3a], [Fig.3b] and [Fig.3c], the markers are placed on the strip 101. In the embodiment illustrated in [Fig.3d], the markers are positioned on one of the strips 101a, 101b. They may also be positioned on both strips 101a, 101b. The markers are geometrically identical in pairs and delimit a portion of the strip. In these examples, the markers are QR codes. The markers may also be other fiduciary markers, a set of examples of which are illustrated in a non-limiting manner in [Fig.4].

[0062] The device 1 further comprises a terminal 11 configured to obtain an image of the assembly and to measure the elongation of the compression garment 100 from the image. In the example illustrated in [Fig.l], the terminal 11 is a mobile phone. It can also be a computer, or an image acquisition device such as a camera or a video device. It can also be any other means capable of acquiring images. The terminal can further be connected to a database 12 remote from the terminal 11. The terminal can connect to the database 12 and exchange data with the database 12 by WiFi connection or by mobile telephone network (4G or 5G). The terminal can also be configured to display data on a screen, for example the deformed compression garment 100.

[0063] To measure the elongation of the support garment, the terminal is configured to implement a measurement method in which, with reference to [Fig.5], the following steps are implemented.

[0064] The method is implemented as an example in the case of a compression garment. Of course, the measurement method can be implemented for any object that obeys Hooke's law. Furthermore, the method is implemented by a terminal. However, this is a non-limiting example and the method is suitable for being implemented by computer in general.

[0065] During a step E0, an authentication of a serial number of the garment 100 is implemented by querying the database 12.

[0066] If the authentication is valid, an image comprising a portion of the support garment 100 in a deformed position and the print is acquired during a step E1, preferably in an adequate resolution.

[0067] If the authentication is not valid, the database 12 sends an error message to the terminal 11.

[0068] During a step Eli, the image is improved by applying at least one of the following operations: correction of the image resolution; filtering; smoothing; image thresholding (better known by the English acronym “thresholding”); noise removal; histograms; histogram modifications; linear transformations; non-linear transformations; contrast modification; image restoration to remove degradations.

[0069] During a step E12, a verification of the validity of the markers 103 is implemented. For example, the terminal connected to the database 12 can verify a correspondence of the markers 103 with marker models pre-recorded in the database 12.

[0070] During a step E13, a segment in the image is determined. The segment can be determined by detecting two markers 103. In the case where there are at least four markers 103, several segments are determined. Each segment can also be determined by processing the image. The image segmentation methods are known to a person skilled in the art and will not be detailed further.

[0071] During a step E14, the image is cropped at the segment level. For example, the image is cropped by rotation and translation so that it is defined between the markers 103. In the case where there are at least four segments, depending on the precision required by the method, the image may be cropped at the two markers at the ends of the image. Consequently, all the markers are visible. According to another embodiment, the image may be divided into several parts each comprising one or more segments according to the desired precision and each part is then cropped.

[0072] A detection of contours in the image is implemented during a step E15. For example, the contours corresponding to the imprint and the markers 103 are determined. The contours are highlighted in the image by removing the colors background in the contours or by other means, for example by highlighting or by changing the color of the contours or by transforming them into a binary image, in this case in black and white.

[0073] During a step E16, the imprint, the markers 103 as well as the reference points 102 are identified in the image.

[0074] During a step E17, a virtual line 104 and intersection points pv pv p^ and p"^ of the virtual lines with the bands 101a, 101b and the reference mark 102 are added by superposition in the image. It may also be provided to superimpose several virtual lines 104a, 104b, 104c and intersection points for each of the lines.

[0075] During a step E2, an elongation of the garment 100 by tracking the reference mark 102 relative to the imprint is determined from the image and the dimensions of the garment at rest.

[0076] Indeed, when the garment deforms, the mark moves away from the imprint. In the embodiment illustrated in [Fig.6a] and [Fig.7a], the lines 102 move away on either side of the strip 101 during the deformation. In the embodiments illustrated respectively in [Fig.6b] with [Fig.7b], and in [Fig.6c] with [Fig.7c], the lines 102 each move away from the strip 101. In the embodiment illustrated in [Fig.6d] and [Fig.7d], the lines 102 on each of the strips 101a, 101b move away from each other.

[0077] The preceding examples show a mode of deformation of the garment in which the garment elongates relative to the strips. By analogy, the preceding examples also show a mode of deformation in which the garment contracts relative to the strips. In the first case, the markers move away from the strip or move away from each other while in the second case, they move towards the strip or move towards each other.

[0078] Thus, the Applicant found that it was possible to deduce an expression for the elongation from an equality between a cross-ratio of four points aligned in the image and a cross-ratio of four points aligned in an image without perspective effect. Indeed, in these deformation modes, the equality only includes a single unknown which corresponds to the elongation.

[0079] An example of solving the equality which makes it possible to obtain the elongation is presented below in the most unfavorable case, that is to say that in which the band 101 is between the lines 102.

[0080] In the resolution example, the equality is stated as follows:

[0081] Pp'i vp\p'3 v —-—- P2P3 PP. P2P3 PP3

[0082] With pr p^ pv p4 the intersection points in pixels taken in the image having undergone a perspective effect and p', p2, p... p points of intersection with the strip 101 and the lines 102 of the deformed garment without perspective effect. Indeed, the points p", p", Py P4 are a reconfiguration of the points Pv pv py P4 under the perspective effect.

[0083] The lengths P}P3, P^Py P^P^ and P\P3 are determined from Chasles relations in the following manner.

[0084] p\p3 = p \p2 + P2P3

[0085] P2P^P2P3 + P3P^

[0086] p\p4 = p\p2 + p2p3+P^

[0087] Now, by making an assumption of a uniformly distributed elongation in the garment which is valid because the garment obeys Hooke's law, the lengths P4P2 and P3P^ are equal. Moreover, as the strip 101 is non-deformable relative to the reference frame, here in this case the lines 102, then the length P2P3 is equal to the length P^Py Consequently, the lengths P\Py P2Py P2P^ and P1P3 are expressed in the following manner.

[0088] P'^ = k ■ P4P2 + P2P3

[0089] P2P4 = P2P3 + k ■ P3P4

[0090] p\p4 = 2k ■ p^p2 + P2P3

[0091] With k the elongation which is greater than 1 in the case of an elongation of the garment 100 and which is less than 1 in the case of a retraction of the garment 100.

[0092] The lengths P\P2, P2P3 and P3^4 may be predetermined in an image comprising the garment 100 at rest, the strip 101 and the lines 102 previously acquired. They may also be predetermined by prior measurements by measuring means such as a tape measure.

[0093] Thus, there remains only one unknown which corresponds to the elongation k which is expressed, from the cross-ratio, in the following manner which is a second degree equation.

[0094] Ak2 + Bk + C = 0

[0095] With A, B and C real-valued coefficients.

[0096] Since all the lengths in the cross-ratio are known, then the coefficients A, B and C are also known. Thus, by solving the second degree equation, it is possible to express the elongation k as a function of the coefficients A, B and C. It is therefore possible to obtain a value of the elongation k.

[0097] According to a numerical example, it is considered that the measured lengths P\P2, P7P3 and P3P4 are as follows:

[0098] PXP2 = P3P4 = 5 mm

[0099] P2P3 = 4 mm

[0100] the expressions of the lengths PXPV PzPy P2P4 and P\P3 become in the example:

[0101] PiP3 = 5-Æ + 4

[0102] P2P4 = 4 + 5-k

[0103] = 2 ■ 5 ■ / < + 4 = 10 ■ / < + 4

[0104] with P\PV P2Py P2Pi and P^ in mm-

[0105] Furthermore, in the numerical example, it is considered that the lengths P\P2, PP:: and P^p". are as follows: A 4-

[0106] P^P^ 30 pixels

[0107] P'2P'3 = 10 pixels

[0108] p'3p'4 = 20 pixels

[0109] Thus, the expression of the equality of the cross-ratio in a developed form is the next.

[0110] 10 30+10+20 4 1OA+4

[0111] The following second degree equation is deduced from this.

[0112] 25-^-40-^-16 = 0

[0113] After solving the second degree equation, the elongation k is equal to 1.93. The elongation is much greater than 1 and therefore corresponds to an elongation of the garment 100.

[0114] Of course, a similar reasoning can be applied for the embodiments illustrated in Figure 2b, Figure 2c and Figure 2d. The reasoning can also be applied beyond the compression garment 100, for any object that obeys Hooke's law. Indeed, the elongation k is expressed from an analysis of an image, predetermined measurements and the expression of the cross-ratio. In addition, the expression of the cross-ratio and the hypothesis of uniform deformation remain valid as long as the object obeys Hooke's law. Consequently, the reasoning remains valid for any set that includes an object, a reference frame and an imprint that is non-deformable relative to the reference frame.

[0115] In a case where several lines 102 are superimposed in the image, a plurality of intermediate elongations k, are determined according to the same reasoning then the elongation k is obtained by taking an average of the plurality of intermediate elongations k,. In the case of several segments, the elongation k is determined for each segment.

[0116] During a step E3, a circumference 301 of the deformed garment 100 is determined from the elongation k and a circumference of the garment at rest predetermined in an image of the garment 100 at rest or by measuring means such as a tape measure. Indeed, by assuming that the elongation is distributed uniformly in the garment, the circumference of the deformed garment is proportional to the circumference of the garment at rest. It is therefore possible to obtain the circumference of the deformed garment by multiplying the circumference of the garment at rest by the elongation. In the numerical example, it is considered that the circumference of the garment at rest is 100 mm, then the circumference of the deformed garment is 193 mm. In the case where several segments are in the image, then a circumference of the deformed garment 100 is determined for each segment.

[0117] It is thus possible to repeat steps E1 to E3 from a first end of the garment to a second opposite end of the support garment to obtain a surface 30 of the support garment deformed by concatenation of the circumferences 301.

[0118] During a step E4, a volume 401 of the deformed garment 100 is determined from the circumference 301.

[0119] A pressure field 402 applied to the deformed garment 100 is also determined from the circumference 301 independently of the determined volume 401.

[0120] For example, each point of the pressure field 402 can be determined by applying Laplace's law in which the pressure at each point of the deformed garment 100 is equal to the ratio between a tension of the garment, defined as the product of the Young's modulus of the garment, given by the manufacturer or obtained by tensile testing, and the deformation of the garment deduced from the measurements obtained by implementing steps E1 to E2, and the radius of the circumference obtained in step E3. Assuming a uniform distribution of the Young's modulus in the garment, it is further possible to determine a normalized pressure field, for example expressed between zero as the lowest measured value and one as the highest measured value, by choosing the Young's modulus as being equal to one because the tension is proportional to the elongation.

[0121] It is thus possible to repeat steps E1 to E4 from the first end of the garment to the second opposite end of the support garment to obtain a volume and a total pressure field 40 of the support garment deformed by concatenation of the volumes 401 and the pressure fields 402.

[0122] Furthermore, the circumferences 301, the surface area 30, the volumes 401, the pressure fields 402, the volume and the total pressure field 40 determined can furthermore be displayed on a screen of the terminal 11.

[0123] In order to improve the calculation speed, the database can be configured to measure the elongation of the support garment 100 from the image by implementing steps E1 to E2 and then transmitting the results to the terminal 11 for display.

[0124] Many modifications can be made to the device and the method without go beyond the scope of the presentation.

Claims

Claims

1. A computer-implemented method for measuring an elongation of an object (100), the method comprising the following steps: acquiring an image (El) comprising an object (100) in a deformed position and an imprint (101) connected to the object (100), the imprint (101) being non-deformable relative to the object (100), the object (100) comprising a reference mark (102); and from the acquired image and dimensions of the object (100) at rest, determining (E2) an elongation (k) of the object (100) by tracking the reference mark (102) of the object (100) relative to the imprint (101), the dimensions of the object (100) at rest being defined relative to the reference mark (102) and relative to the imprint (101).

2. Method according to claim 1, comprising a determination (El3) of a segment in the image, the determination of the elongation (k) of the object taking place in the segment.

3. The method of claim 2, wherein the fingerprint comprises two markers (103), the segment being defined between the two markers (103).

4. The method of claim 3, wherein the two markers (103) are fiduciary markers, barcodes, or QR codes.

5. Method according to one of claims 1 to 4, comprising a step of measuring (E3), from the elongation (k), a dimension of the object (100).

6. Method according to one of claims 1 to 5, comprising the following steps: from the elongation (k), measurement (E3) of a circumference of the object (100); and from the measurement, determination (E4) of a volume of the object (100).

7. Method according to one of claims 1 to 6, in which the imprint comprises two distinctly connected patterns on the object (100).

8. Method according to one of claims 1 to 6, the method being implemented by a terminal (11) and in which the object (100) is a support garment, the mark (102) is formed of two lines and the imprint (101) is formed of a strip.

9. A method according to claim 8, wherein the strip (101) is between the two lines (102), or the two lines (102) are adjacent.

10. A method according to claim 8, wherein the imprint (101) is formed of a first strip (101a) and a second strip (101b), a first line being taken on one end of the first strip and a second line being taken on one end of the second strip (101b) opposite the end of the first strip (101a).

11. A method according to one of claims 8 to 10, wherein the strip comprises markers (103), the markers being aligned on the strip.

12. Assembly (10) comprising: an object (100) comprising a marker (102); and an imprint (101) connected to the object (100), the imprint (101) being non-deformable relative to the object (100), the assembly being configured to allow determination of an elongation (Æ) of the object (100) by tracking a marker of the object (100) relative to the imprint.

13. Device (1) for measuring a deformation of an object, the device comprising: an assembly (10) comprising: an object (100) comprising a marker (102); and an imprint (101) connected to the object (100), the imprint (101) being non-deformable relative to the object (100), the assembly being configured to determine an elongation (k) of the object (100) by tracking a marker (102) of the object (100) relative to the imprint (101); and a terminal (11) configured to: acquire an image comprising the assembly (10); and from the acquired image and the dimensions of the object at rest, determining an elongation (^j of the object (100) by tracking the reference frame (102) of the object (100) relative to the imprint (101), the dimensions of the object (100) at rest being defined relative to the reference frame (102) and relative to the imprint (101).

Images