DEVICE FOR DETECTING A DEFECT ON A SURFACE USING MULTIDIRECTIONAL LIGHTING

Description

[0001] The present invention relates to a device for detecting a defect on a surface.

[0002] A current device essentially consists of an optical device capable of acquiring an image or film of the surface. This image or film is then viewed by an operator who indicates the defects he has identified.

[0003] However, the operator may not spot a defect due to inattention. In addition, the criteria for judging a defect are subjective, meaning that the definition of what a defect is may vary depending on the operator.

[0004] A common solution to these problems is to have the film or image viewed by multiple operators. Additionally, surface image examples are used to demonstrate what is a defect and what is not.

[0005] However, the examples shown are neither exhaustive nor reproducible.

[0006] In the event of a defect, experts are consulted. There is then a lot of back and forth between experts, as the experts do not necessarily agree with each other.

[0007] Analyzing a surface is likely to be long and complex.

[0008] JP 2012 047673 A describes a system for inspecting a surface. Its illumination unit comprises several light sources and a hood on which the light sources are arranged. The hood is composed of a hemispherical element that covers the inspected object. An illumination control unit sequentially turns on the light sources in a predetermined order. The imaging control unit causes the image sensor to capture an image in synchronization with the moment when the illumination control unit turns on a light source. Thus, if the number of an image is known, it is possible to identify the illumination direction of the inspected object.

[0009] WO 02 / 090952 A1 describes a method for examining a surface in which the light which successively strikes the surface at different angles of incidence and which is reflected on said surface is measured by an image sensor.

[0010] DE 35 40 288 A1 describes a weld inspection system in which light is irradiated onto a welded part at different angles of incidence by a light emitter in order to collect information indicating the three-dimensional configuration of the welded part and to determine whether the configuration is acceptable or not.

[0011] The invention aims in particular to remedy this drawback by providing a device allowing simple and reliable detection of a defect on a surface.

[0012] To this end, the invention relates in particular to a device for detecting a defect on a surface according to claim 1.

[0013] The detection device according to the invention may further comprise characteristics according to claims 2 to 10.

[0014] The invention will be better understood by reading the following description, given solely by way of example and with reference to the appended figures, among which: there figure 1 is a schematic sectional view of an exemplary device according to an embodiment of the invention, the figure 2 is a schematic view of the device of the figure 1 , there figure 3 is a diagram of steps of an embodiment of the method implementing a device of the invention, the figure 4 is an example of a surface on which the detection device is used. figure 1 , there figure 5 is an example of the evolution of three parameters on a line of the surface of the figure 4 , there figure 6 is an example of defect detection on the surface of the figure 4 , and the figure 7 is an example of parameters at a set of points with a defect and at a set of points without a defect.

[0015] An example of a device 10 for detecting a defect in a surface 12 is shown in figures 1 And 2 .

[0016] The detection device 10 comprises an optical device 14, a lighting device 15 comprising a first number of light source(s) 16 and an electronic calculation device 18.

[0017] The optical device 14 has an optical axis aligned with a given optical direction O. It has an acquisition field.

[0018] The optical device 14 is capable of acquiring an image of the surface 12 along the given optical direction O. The acquisition field coincides with the surface 12 to be inspected, so that the acquired image represents the surface 12.

[0019] The optical device comprises, for example, at input an optic adapting the image acquired by the optical device, for example, the acquisition field.

[0020] For each point on the surface, an observation direction B, B' is defined between the optical device 14 and the point on the surface 12.

[0021] The acquisition field is fixed here.

[0022] The optical direction O is typically substantially perpendicular to the surface 12. Alternatively, the optical direction O is not perpendicular to the surface 12.

[0023] The optical device 14 is, for example, a camera.

[0024] The optical device 14 is, more particularly, a charge transfer device, more commonly called a CCD sensor from the English “ charge-couple device ", or a CMOS sensor from the English " complementary métal oxide semi-conductor » .

[0025] Alternatively, the acquisition field coincides with only a part of the surface 12 and the detection device 10 is able to be moved relative to the surface 12 so as to be able to acquire the entire surface 12 in several parts.

[0026] Each acquired image consists of a matrix of pixels of dimension i*j, with i the number of horizontal lines and j the number of vertical lines of the matrix, with at least one of the two parameters (i,j) greater than 1. The pixel is, for example, a rectangle of size noted h*L, with h its height and L its width. The height h is equal to the width L. Alternatively, the height is different from the width

[0027] We subsequently call a pixel of an acquired image or a part of the surface acquired in a pixel indifferently by "point of the surface".

[0028] The light sources 16 are capable of illuminating the surface 12.

[0029] For each point on the surface, the light sources 16 each have a given lighting direction E, E' connecting the light source to the point on the surface 12. The lighting direction is different for each light source.

[0030] The light sources 16 are typically arranged in a hemisphere 20 surrounding the surface 12.

[0031] The hemisphere has a diameter D less than or equal to 50 mm (millimeters), preferably less than or equal to 20 mm.

[0032] The diameter D is, moreover, strictly greater than 5 mm.

[0033] The first number of light sources 16 is between 32 and 128, more particularly equal to 96.

[0034] The light sources 16 are, for example, light-emitting diodes (LEDs).

[0035] They can be turned off or on. They are configured to be turned on independently of each other.

[0036] The 16 light sources have identical light intensity.

[0037] Alternatively, the lighting device 15 comprises one or more possibly non-identical lighting devices.

[0038] The electronic computing device 18 is, for example, a computer, a calculator, a computing module, at least one programmable logic component such as FPGAs (from the English Field-Programmable Gate Array ) or at least a dedicated integrated circuit such as ASICs (from the English Application-Specific Integrated Circuit ) .

[0039] The electronic computing device 18 is, for example, connected to the optical device 14 and to the lighting device 15.

[0040] The connection is, for example, made by cables.

[0041] Alternatively, the connection between the electronic computing device 18 and the optical device 14 and / or the lighting device 15 is achieved by wireless technology. The wireless technology is, for example, according to a Wifi protocol corresponding to a set of wireless communication protocols governed by the standards of the IEEE 802.11 group (ISO / IEC 8802-11).

[0042] The electronic computing device 18 is capable of switching on or off each of the light sources 16 independently of one another.

[0043] The electronic calculation device 18 is, furthermore, provided to trigger the acquisition of at least one image of the surface by the optical device 14.

[0044] It is thus configured to cause the acquisition of a plurality of images of the surface by the optical device 14, each image being acquired according to the given optical direction O and with a single light source switched on or a defined combination of single sources switched on. This source or this combination is different for each image. Each image corresponds to a different lighting.

[0045] For each point of the surface, the electronic calculation device 18 is configured to calculate from the acquired images a plurality of parameters, the parameters comprising coefficients of an equation characterizing the response of said point of the surface as a function of the lighting direction and an observation direction, then to deduce from the calculated parameters if the surface has a defect at said point.

[0046] This last point is detailed below in relation to the process of detecting a fault.

[0047] The device 10 here comprises a tool 22 comprising the optical device 14 and the lighting device 15.

[0048] In the example shown, the electronic calculation device 18 is placed outside the tool 22.

[0049] In the case of a connection of the electronic computing device 18 with the optical device 14 and / or the lighting device 15 by cable, the electronic computing device 18 is, for example, placed on a movable support 23. The support 23 comprises, for example, casters.

[0050] The tool 22 has dimensions such that it is included in a cylinder having a length equal to 200 mm, more particularly equal to 150 mm, and a diameter equal to 100 mm, more particularly equal to 75 mm.

[0051] Tool 22 has a mass less than 3 kg (kilograms), more specifically less than 2 kg.

[0052] The tool comprises a handle 24 and a head 26 secured to each other. The tool 22 is, for example, made up of the handle 24 and the head 26.

[0053] The handle 24 and the head 26 are immobile relative to each other. For example, they are made of one material.

[0054] Alternatively, the handle 24 and the head 26 are integral in a connection zone, the handle 24 being able to be moved relative to the head 26 according to a pivot connection or a ball joint connection.

[0055] The handle 24 is capable of being grasped by a user.

[0056] It extends mainly in a longitudinal direction, for example, aligned with the optical direction O.

[0057] The handle 24 is substantially cylindrical.

[0058] The head 26 here comprises the optical device 14 and the lighting device 15.

[0059] The head 26 defines a concavity 28.

[0060] The concavity 28 is open, that is to say it communicates with the exterior of the tool 22.

[0061] The concavity 28 is open opposite the handle 24.

[0062] The concavity defines an internal shape corresponding to the hemisphere 20 of the lighting device 15.

[0063] The head 26 is connected to the handle 24 at a point aligned in the longitudinal direction of the handle 24 with the top of the hemisphere 20. More particularly, the handle 24 is aligned in its longitudinal direction with the top of the hemisphere 20.

[0064] The light sources 16 are capable of illuminating the concavity 28.

[0065] The optical device 14 is placed at the top of the hemisphere 20.

[0066] The optical axis of the optical device 14 extends into the concavity 28.

[0067] The head 26 is capable of being applied in contact with the surface 12. More particularly, it comprises a support surface 30 capable of being placed in contact with the surface 12, so that the surface 12 is capable of being illuminated with the lighting device 15 and acquired with the optical device 14.

[0068] The bearing surface 30 is adapted so that the concavity 28 of the head 26 forms a closed space when the bearing surface 30 is placed in contact with the surface 12.

[0069] The bearing surface 30 extends around a circle defining an edge of the concavity 28.

[0070] The bearing surface 30 is included in a plane substantially perpendicular to the optical direction O of the optical device 14.

[0071] The bearing surface 30 is a ring whose internal radius corresponds to a radius of the hemisphere 20 of the lighting device 15.

[0072] The tool 22 further comprises at least one adapter piece having a generally cylindrical shape extending between a first and a second end surface. The first end surface is adapted to cooperate with the bearing surface 30. The second end surface is adapted to cooperate with the surface 12 on which the tool is intended to be applied. In particular, if the surface 12 is curved, the second end surface is a ring having a curvature similar to the surface. The adapter piece is, for example, opaque, so that the light is maintained in the space defined between the light sources 16, the surface 12 and the adapter piece.

[0073] Alternatively, the electronic calculation device 18 is included in the tool 22, more particularly in the handle 24.

[0074] A method of detecting a defect in a surface will now be described with regard to the figure 2 .

[0075] The method is implemented here by the device described previously.

[0076] The process includes the following steps: acquisition 100, calculation of parameters 102, and deduction of the presence of a defect 104.

[0077] During the acquisition 100, the device 10 is placed in contact with the surface 12, more particularly at the level of the support surface 30 of the device 10.

[0078] If the surface 12 is curved, an adaptation part is advantageously used between the tool and the surface 12 to allow better adaptation of the tool to the surface.

[0079] During the acquisition 100, a plurality of images 200 of the surface is acquired by the optical device 14. Such an image is represented in figure 3 .

[0080] Each image 200 is acquired with illumination of the surface 12 according to a given illumination direction E, E' for each point of the surface and with the optical axis of the optical device 14 according to a given optical direction O.

[0081] The 200 images are acquired with different lighting directions E, E' or combinations of lighting directions.

[0082] In the embodiment shown, the lighting directions E, E' differ for each acquired image and the optical direction O is invariable and substantially perpendicular to the surface 12.

[0083] For each acquired image, a single light source or a defined combination of single sources 16 of the illumination device is illuminated, the single light source or combination being different for each acquired image.

[0084] All light sources or combinations are turned on in turn, so as to acquire one image per light source or combination. Alternatively, some sources or combinations are not turned on.

[0085] Then, during the calculation of parameters 102, for each point 202 of the surface, that is to say here for each pixel of the acquired images, a plurality of parameters are calculated from the acquired images.

[0086] The parameters are calculated from the coefficients of an equation characterizing the response of said point on the surface as a function of the lighting direction E, E' and an observation direction B, B'.

[0087] For each point 202 of the surface 200, the calculation of parameters 102 here comprises the following steps: definition of a quantity 106, choice of model 108, and calculation of coefficients 110.

[0088] When defining a quantity 106, a quantity evolving according to the acquired images is defined.

[0089] For example, the magnitude here is the gray intensity of the surface point, the gray intensity being variable according to the acquired images.

[0090] In parallel, a model representing the evolution of the quantity is chosen 108. The model depends on the lighting direction(s) E, E' and / or the observation direction B, B'. The model includes coefficients.

[0091] Here the pattern depends only on the lighting direction. It has, for example, the following form: G ( E ) = a 0 × D u 2< + a 1 × D v 2< + a 2 × D u × D v + a 3 × D u + a 4 × D v + a5 , with G the magnitude, E the lighting direction, a 0 , a 1 , a 2 , a 3 a 4 and a 5 the coefficients and D u and D v being coordinates of the vector connecting the lighting source to the point on the surface.

[0092] The coefficients a 0 , a 1 , a 2 , a 3 a 4 and a 5 are here undetermined.

[0093] Then, when calculating the coefficients 110, an application of the magnitude to the model is carried out so as to calculate values ​​of a 0 , a 1 , a 2 , a 3 a 4 and a 5 .

[0094] The parameters here are equal to the coefficients a 0 , a 1 , a 2 , a 3 a 4 and a 5 thus calculated.

[0095] For each point on the surface, the same quantity and the same model are defined. Thus, for each point on the surface, we obtain parameters a 0 , a 1 , a 2 , a 3 a 4 and a 5 . We obtain, for example, a map of the surface representing the values ​​of the parameters for each point on the surface.

[0096] On the figure 5 , the evolution of three parameters is represented along the line 204 of points of the surface represented in the figure 4 .

[0097] Alternatively, the parameters are other values ​​calculated from the coefficients.

[0098] When deducing the presence of defect 104, it is deduced from the calculated parameters whether the surface has a defect at said point.

[0099] On the figure 7 , for example, the parameters are represented with the curve C 1 at a point not presenting a defect and with the curve C 2 at a point presenting a defect.

[0100] The result obtained in step 104 for each point is, for example, a boolean representing binary whether a fault is detected at said point.

[0101] On the figure 6 , the boolean is represented along line 204: if the boolean is equal to 0, no fault is detected; if the boolean is equal to 1, a fault is detected.

[0102] So, on the figure 6 , a single 300 defect is detected and extends over several adjacent points.

[0103] One or more criteria are used to infer the presence of a defect from the parameters. These criteria are described below and are used alone or in combination to detect a defect. For example, a defect is detected when at least two of the criteria are met.

[0104] A first criterion is that a fault is detected at said point when at least one parameter at said point is not included in an interval.

[0105] More specifically, the interval is centered around the average value of said parameter over all points on the surface.

[0106] Alternatively, the interval is a given interval.

[0107] Alternatively, a fault is detected at said point when at least a given number of parameters at said point are not within a respective interval.

[0108] A second criterion is that a defect is detected at said point when at least one parameter at said point differs by a value chosen as a detection threshold compared to the background noise of said parameter at an adjacent point, the surface not presenting a defect at the adjacent point.

[0109] Thus, step by step, the surface is analyzed.

[0110] It is then appropriate to define at least one first point on the surface for which a defect is not present.

[0111] Alternatively, the electronic calculation device 18 considers that a location on the surface is free of defects and, step by step, detects defects in the surface. If the result is deemed unacceptable, for example, if a majority of the surface is detected as having a defect, then the step is repeated by considering that another location, different from previously, is free of defects.

[0112] A third criterion is based on the geometry of a signal.

[0113] The signal here is the evolution of one or more parameters on a defined set of points. The points of the set are here adjacent. The set of points is, for example, an area centered on a central point.

[0114] Patterns are defined corresponding to cases in which a defect is detected and / or in which no defect is detected. The signal shape is compared to the patterns, so as to decide whether a defect is detected. The defect is detected at the center point, in an area including the center point and included in the point set or in the point set.

[0115] Thus, for each point on the surface, it is deduced from the parameters whether a defect is present or not.

[0116] This can, for example, be represented on a binary map of the surface in order to easily visualize defects.

[0117] Alternatively, the light sources 16 do not have identical light intensity. A calibration of the device is then carried out.

[0118] Advantageously, in addition to deducing the presence of a defect, the method comprises a step of categorizing the nature of the defect.

[0119] Depending on the criteria previously described, the nature of a defect can be defined, for example, between a crack, a scratch, a deformation or porosity.

[0120] Alternatively, the calculation of the parameters 102 is based on another mathematical model such as the bidirectional reflectance distribution function (or BRDF) or the discrete modal decomposition (or DMD) or the bidirectional texture function (or BTF).

[0121] Direct deduction from parameters calculated from acquired images makes it possible, in particular, to overcome the susceptibility to the presence of defects that varies between operators. The process thus makes it possible to obtain a single, objective result. Thus, the process is simplified and safe.

[0122] The tool has small dimensions. This makes it possible to inspect a surface with limited accessibility. This also reduces the difficulty of the user's work when using the detection device, as the tool is easily portable by the user, particularly thanks to its handle.

[0123] In addition, the detection device can be moved by a user or a robotic system. This makes it easy to carry out surface inspection.

Claims

1. A device for detecting a defect on a surface (12), the detection device (10) comprising: - an optical device (14) having an optical axis, the optical device (14) being able to acquire an image of the surface (12) along a given optical direction (O), - a lighting device (15) comprising a plurality of light sources (16), the light sources (16) being arranged in a hemisphere (20), each light source (16) having an off state and an on state, each light source (16) having a lighting direction (E, E'), - an electronic calculation device (18), the electronic calculation device (18) being able to control the optical device (14) and the lighting device (15) so that the optical device (14) acquires a plurality of images of the surface (12), different light sources (16) or different combinations of light sources (16) being switched on for each image, the hemisphere (20) having a diameter D smaller than or equal to 50 mm (millimeters), the lighting device (15) comprising a first number of light sources (16), the first number being comprised between 32 and 128, each light source (16) having a different lighting direction (E, E'), for each acquired image, a single light source or a defined combination of sources (16) being illuminated, the single light source or combination (16) being different for each acquired image (200), the light sources (16) having respective light intensities identical to one another, the electronic calculating device (18) being configured to calculate, for each point of the surface, from acquired images, a plurality of parameters, the parameters including coefficients of an equation characterizing the response of said point of the surface as a function of the lighting direction and an observation direction, then to deduce from the calculated parameters whether the surface has a defect at said point.

2. The detection device according to claim 1, characterized in that the optical device (14) is immobile relative to the hemisphere (20).

3. The detection device according to claim 1 or 2, comprising a bearing surface (30) able to be placed in contact with the surface (12).

4. The detection device according to claim 2 and 3, characterized in that the optical direction (O) of the optical device (14) is substantially perpendicular to a bearing surface (30) of the device (10) able to be placed in contact with the surface (12).

5. The detection device according to claim 3 or 4, characterized in that the bearing surface (30) is a ring, the inner radius of which corresponds to a radius of the hemisphere.

6. The detection device according to any one of claims 3 to 5, comprising an adapting piece having a first end able to be mounted on the bearing surface (30) and a second end able to cooperate with the surface (12).

7. The detection device according to any one of the preceding claims, comprising a tool comprising the optical device (14) and the lighting device (15).

8. The detection device according to claim 7, characterized in that the tool comprises a handle (24) able to be grasped by a user.

9. The detection device according to claim 7 or 8, characterized in that the tool comprises a head (26) defining a concavity (28) defining an inner surface corresponding to the hemisphere (20) of the lighting device (15).

10. The detection device according to claim 9 combined with claim 3, characterized in that the concavity (28) is open, the bearing surface (30) extending around an edge of the concavity (28).