PROCEDURE OF EVALUATION OF THE CLARITY OF A MATERIAL, ESPECIALLY NON-HOMOGENEOUS

Abstract

The invention relates to a method for evaluating the clarity of a material, such as a lignocellulosic material, comprising the steps of: - providing a sample of said material, - partially backlighting the sample by a light source so as to form on said sample an unlit portion, an illuminated portion and an edge separating the unlit portion and the illuminated portion, - acquiring a first digital image, by an image sensor, of the partially backlit sample, characterized in that it further comprises the steps of: - generating a digital image corrected by a function for homogenizing the brightness level of the illuminated portion of said first digital image, and - determining a rise distance from the corrected digital image to evaluate the clarity of said material. Figure for abstract: Fig.1

Description

Title of the invention: METHOD FOR EVALUATING THE CLARITY OF A MATERIAL, IN PARTICULAR A NON-HOMOGENOUS MATERIAL TECHNICAL FIELD OF THE INVENTION

[0001] The present invention relates to the field of optical characterization of materials intended for example to be used in luminous display devices.

[0002] According to a preferred application, the invention is described below with reference to the optical characterization of so-called "non-homogeneous" materials, such as lignocellulosic materials. It should however be noted that the invention is applicable to any other material, in particular homogeneous materials with a diffusive or non-diffusive character. STATE OF THE ART

[0003] In various industrial fields, particularly the automotive industry, the readability of human-machine interfaces (light pictograms, screens, etc.) is of capital importance, particularly when it comes to signaling an alert.

[0004] Manufacturers of such interfaces pay particular attention to the sharpness (resolution for screens) of the information displayed and ensure that it is readable in all circumstances. This involves, among other things, controlling the sharpness of this information, which typically defines patterns or images.

[0005] In the case of screens, the resolution of the information displayed is mainly linked to the size of the pixels used. For luminous display devices, such as luminous pictograms, the sharpness is rather linked to the architecture of the device, in particular to the materials used and to the arrangement of the layers constituting the device.

[0006] The use of materials with a diffusive nature, particularly for aesthetic or camouflage purposes, to cover luminous pictograms may degrade the clarity of these pictograms. The edges of the pictogram may in particular appear blurred due to the diffusive nature of the material, which impairs its readability.

[0007] Known methods make it possible to evaluate sharpness by evaluating clarity. Clarity corresponds to the notion of "clarity" in Anglo-Saxon language and qualitatively defines the sharpness of light information, such as a pattern or an image, when passing through a material.

[0008] A material with low clarity will tend to degrade the sharpness of the display device, while a material with high clarity will have minimal or no impact on sharpness.

[0009] Among the known clarity assessment methods, the so-called rise distance method is conventionally used to provide a quantitative assessment of the capacity of a system, such as a camera or a camera, to reproduce the fine details and edges of a digital image. This method is also used to quantitatively assess the clarity of a material.

[0010] In this context, the rise distance method requires the provision of a digital image 11, as illustrated in [Fig.l], representing a material backlit by a light source and a mask partially obscuring the light source such that the digital image 11 comprises an unlit portion 12, an illuminated portion 13 and an edge 14 separating the unlit portion 12 and the illuminated portion 13.

[0011] The digital image typically consists of a grid of pixels aligned in columns and rows. The edge 14 is generally aligned with the columns or with the rows to facilitate the determination of an edge spread function, as illustrated in [Fig.2] in graphical form.

[0012] The edge spread function here represents the light intensity of a line of the digital image 11 (referenced II-II in [Fig.l]) as a function of a distance, which is generally expressed in millimeters.

[0013] The rise distance Dm is the distance over which the edge spread function passes from a predefined lower threshold of light intensity Sinf, typically equal to 10% of the maximum light intensity on the digital image 11, to a predefined upper threshold of light intensity Ssup, typically equal to 90% of the maximum light intensity on this digital image 11.

[0014] In practice, a rise distance is determined for each row (or column depending on the edge inclination) of the digital image, and an average of these rise distances is calculated. It should be noted that the greater the rise distance, the less clear the display through the material will be.

[0015] The material represented by the digital image 11 of [Fig.l] is a non-homogeneous material, here a lignocellulosic material exhibiting local variations in transmittance or diffusion. Due to these variations, the brightness level varies within the illuminated portion 13, which generates noise on the edge spread function of certain lines. The edge spread functions can then vary greatly from one line to another, in particular when the brightness level varies near the edge 14.

[0016] When the edge spread functions are disturbed, such as that shown in [Fig.2], the light intensity thresholds Sinf and Ssup may be reached at random points, which does not allow the clarity of the material to be reliably assessed. The rise distance Dm determined on the basis of these edge spread functions 14 are not representative of the actual material clarity.

[0017] Thus, although the climb distance method is relatively efficient for homogeneous materials, such as glass, it does not give reliable results for non-homogeneous materials, such as lignocellulosic materials. Statement of the invention

[0018] The present invention aims to remedy all or part of the drawbacks of the state of the art cited above.

[0019] To this end, the present invention relates, according to a first aspect, to a method for evaluating the clarity of a material, such as a lignocellulosic material, comprising the steps of: - provision of a sample of said material, - partial backlighting of the sample by a light source so as to form on said sample an unlit part, an illuminated part and an edge separating the unlit part and the illuminated part, - acquisition of a first digital image, by an image sensor, of the partially backlit sample. The method further comprises the steps of: - generation of a digital image corrected by a function for homogenizing the brightness level of the illuminated part of said first digital image, and - determination of a rise distance from the corrected digital image to assess the clarity of said material.

[0020] “Sample” means a portion of material whose clarity is to be evaluated or a sub-part of this portion.

[0021] “Partial backlighting” means backlighting a portion of the sample such that a brightness contrast between the unlit portion and the lit portion formed on the sample is sufficient to be detected by the image sensor.

[0022] The term “brightness level homogenization function” means an image processing or post-processing adapted to standardize the brightness level over at least one area of a digital image.

[0023] In the evaluation method according to the invention, the rise distance method is not applied directly to a digital image representing a partially backlit sample (the first digital image), but to a (corrected) digital image generated from said first digital image and in which the brightness level of the illuminated part is homogenized.

[0024] By homogenizing the brightness level in the illuminated part, it is possible to attenuate possible local variations in the transmittance or diffusion of the material while retaining the clarity information at the edge. In this way, the method makes it possible to evaluate the clarity of a material, including a non- homogeneous, taking advantage of both the ease of implementation and the reliability of the climb distance method.

[0025] The material may be a non-homogeneous material. In particular, the material may be a lignocellulosic material. Such a lignocellulosic material may, for example, be partially delignified and impregnated with a resin. The material may also be wood, in particular natural wood, artificial wood or plastic-coated wood. The material may also be a plastic material, for example a plastic film. The plastic film may be coated with a printing layer, for example a printing layer giving a wood appearance. The material may also be glass, in particular diffusing glass, such as frosted glass or with colloids. The material may be paper. The material may be a composite material, for example glass-reinforced plastic. The material may be formed by a combination of several of the materials mentioned, in particular by an assembly of layers of different materials.

[0026] In a preferred embodiment, the step of generating a corrected digital image comprises a sub-step of fully backlighting the sample and a sub-step of acquiring a second digital image, by the image sensor, of the fully backlit sample, the corrected digital image being generated from the first digital image and the second digital image. The use of a second digital image, in addition to the first digital image, to generate the corrected digital image makes it possible to improve the accuracy of the results obtained by the ascent distance method.

[0027] The function of homogenizing the brightness level of the illuminated part may consist of dividing a characteristic information of each pixel at a given position of the first digital image by said information of each pixel at said given position of the second digital image, said characteristic information being representative of the brightness. The determination of information representative of the brightness for each pixel of the first and second digital images as well as the division of this information are relatively simple to implement and do not require heavy processing. The corrected digital image can thus be generated fairly quickly, of the order of a second or even less. In addition, this method offers satisfactory homogenization of the brightness level of the illuminated part, without disturbing the clarity information at the edge.

[0028] The characteristic information of the pixel representing the brightness can be the luminance. Luminance measures the light intensity perceived by the human eye on a given surface. It allows the clarity of the material to be rigorously evaluated.

[0029] The step of generating a corrected digital image may further comprise a sub-step of digitally aligning the first digital image and the second digital image. The spatial alignment between the first digital image and the The second digital image helps to limit noise that could be generated by a slight shift between these digital images. Digital alignment can be performed with an image alignment algorithm. Furthermore, digital alignment can only be implemented if a mechanical alignment is not correctly performed, for example if there is a shift greater than one pixel.

[0030] The method may further comprise, after the determining step, a step of correcting the rise distance by subtracting a reference rise distance from the rise distance determined in the determining step. The reference rise distance may be determined from a first reference digital image representing the light source partially illuminating so as to form a non-illuminated portion, an illuminated portion and an edge separating the non-illuminated portion from the illuminated portion. Since the image sensor used is not perfect, it may happen that the rise distance determined in the determining step is overestimated. The step of correcting the rise distance thus makes it possible to overcome inaccuracies due to the physical limits of the image sensor. The value obtained makes it possible to evaluate the clarity of the material even more precisely.

[0031] The first digital image may comprise several distinct edges, the determining step being implemented for each edge and the evaluation method further comprising, after the determining step, a step of calculating an average of the rise distances measured for each edge. Having several distinct edges makes it possible to evaluate the clarity at different locations on the sample and to estimate, via the averaging step, an average value of clarity of the material. It is then possible to globally evaluate the clarity of a material.

[0032] The invention also relates, according to a second aspect, to a use of the method as described above for controlling the clarity of a display device comprising a material backlit by a light source and a mask located between the material and the light source so as to partially obscure the light source. The material may be non-homogeneous, such as a lignocellulosic material.

[0033] The invention finally relates, according to a third aspect, to a device for evaluating the clarity of a material. This evaluation device has advantages similar to those described previously in relation to the evaluation method. The device comprises a sample of said material, a light source configured to backlight, in particular partially and / or totally, the sample, an image sensor configured to provide digital images of said sample, and a digital processing device configured to implement the evaluation method according to the first aspect of the invention, from digital images acquired by the image sensor.

[0034]

[0035]

[0036]

[0037]

[0038] The sample of the material may be a sample of non-homogeneous material, such as a lignocellulosic material, and preferably wood. The sample may have a general plate shape with two opposite main faces. The light source may comprise a flat surface that diffuses light evenly. In this case, one of the main faces of the material sample may be pressed against the flat surface of the light source. This allows the sample to be backlit evenly and allows for easy positioning, since the light source can thus act as a support. BRIEF DESCRIPTION OF THE FIGURES Other advantages, aims and particular characteristics of the present invention will emerge from the following non-limiting description of at least one particular embodiment of the devices and methods which are the subject of the present invention, with reference to the appended drawings, in which: • [Fig.l] illustrates a digital image representing, in a manner known in the prior art, a sample of a non-homogeneous material partially backlit by a light source so as to form on the sample an unlit part, an illuminated part and an edge separating the unlit part and the illuminated part; • [Fig.2] represents, in a manner known in the prior art, a spreading function of the edge of the first digital at the level of a line ILII represented in [Fig.l]; • [Fig. 3] represents, in a three-dimensional schematic view, an embodiment of a device for evaluating the clarity of a material; • [Fig.4] is an exploded view of a control assembly of the device of [Fig.3]; • [Fig.5] represents, in the form of a block diagram, the steps of an embodiment of a method for evaluating the clarity of a material; • [Fig.6] illustrates a second digital image representing the sample of [Fig.l] fully backlit by the light source; • [Fig.7] illustrates a corrected digital image generated from the first digital image of [Fig.l]; and • [Fig.8] represents a spread function of an edge of the digital image corrected at the level of a line VIII-VIII represented in [Fig.7]. DETAILED DESCRIPTION OF THE INVENTION Figures 3 and 4 represent a device 1 for evaluating the clarity of a material according to an embodiment of the invention. The device 1 mainly comprises a control assembly 2 comprising a sample 4 of a material as well as a digital image sensor 3 configured to acquire images of the control assembly 2 or of an area of this assembly.

[0039] The device 1 optionally comprises a digital processing device 20 configured to digitally process images acquired by the image sensor 3, according to a method described in the remainder of the description.

[0040] The control assembly 2 comprises a light source 5 configured to backlight the sample 4. The light source 5 is located opposite the image sensor 3 relative to the sample 4. Thus, the light emitted by the light source 5 passes through the sample 4 before being received by the image sensor 3. The sample 4 has a general plate shape whose thickness is such that at least part of the light emitted by the light source 5 is transmitted by the sample 4. The maximum thickness of the sample 4 depends in particular on the level of light transmission and diffusion in the material.

[0041] The light source 5 is located on the side of a first main face 6 of the sample 4 in order to illuminate by transmission a second main face 7, opposite the first main face 6. The light source 5 here has a flat surface which is illuminated homogeneously. This makes it possible to ensure that any variations in brightness on the second main face 7 result only from local variations in the transmittance or diffusion within the material of the sample 4.

[0042] For example, the light source 5 is an LED panel. The homogeneity of the light source 5 can be further improved by covering its flat surface with a diffusing film, such as a translucent layer of plastic material (PMMA, PET).

[0043] In the illustrated embodiment, the device 1 comprises a mask 8 intended to be arranged between the sample 4 and the light source 5. The mask 8 serves to partially obscure the light source 5 (i.e. only a part of the surface of the light source) in order to form, on the second main face 7, an unilluminated (obscured) part in which at least a part of the light rays emitted by the light source 5 is absorbed by the mask 8, an illuminated (non-obscured) part in which the light rays are transmitted by the sample 4, and an edge separating the unilluminated part and the illuminated part.

[0044] As visible in [Fig.4], the mask 8 comprises at least one occultation zone and at least one transmission zone. The mask here comprises six transmission zones bearing the reference 8a. Each transmission zone 8a is separated from an occultation zone by at least one edge which is, preferably, rectilinear. Each transmission zone here has a square shape, thus forming four edges.

[0045] The occultation zone may be opaque so as not to transmit any rays light emitted by the light source or generally opaque so as to transmit a portion of the light rays, which portion is relatively small compared to the portion of the light rays transmitted by the transmission zone. Each transmission zone 8a may be transparent or semi-transparent.

[0046] In the example illustrated, the mask 8 is produced by printing on a transparent or translucent plate, so as to define at least one occultation zone and at least one transmission zone. The transmission zone is for example devoid of a printing layer. The printing can be by screen printing, by inkjet (with one or more layers), by physical deposition of a material, for example chromium, in the vapor phase or by microlithography.

[0047] The mask 8 is preferably located as close as possible to the first main face 6 of the sample 4. The mask 8 may for example be in contact with the first main face 6 of the sample 4. This makes it possible to limit the diffusion of the light rays in the sample 4, and to display on the second main face 7 an edge that is as sharp as possible.

[0048] As the mask 8 is generally only used to evaluate the clarity of the material of the sample 4, the device 1 comprises a temporary holding means 10 configured to temporarily hold the mask 8 in contact with or as close as possible to the first main face 6 of the sample 4. The mask 8 can thus be removed after having been positioned to obscure the light source 5. This in particular avoids permanently fixing a mask on a sample whose clarity would prove not to be suitable for a display device.

[0049] It is also preferable that the position of the sample 4 relative to the image sensor 3 is maintained, for example when the mask 8 is positioned or removed, between different acquisitions by the image sensor 3. The device 1 comprises in this respect a means for maintaining the sample 4 in position relative to the image sensor 3, such as a lateral edge against which the sample 4 comes to bear, or an adhesive which fixes the sample 4 to a support fixed relative to the image sensor 3. There may nevertheless appear a slight displacement of the sample 4 relative to the image sensor 3.

[0050] To prevent such a movement from affecting the evaluation of the clarity of the material, the evaluation device 1 preferably comprises an alignment mark 9 placed on the sample 4 to be detected by the image sensor 3. The alignment mark 9 comprises, for example, two distinct opaque patterns fixed on the second main face 7 of the sample 4 so as to both be detected by the image sensor 3. The patterns can be of different shapes and sizes, and are here represented in the form of squares arranged at opposite corners of the sample 4. The alignment mark 9 can be fixed by gluing or by clamping, in particular at by means of known clamps or fasteners.

[0051] As seen in [Fig.4], the alignment mark 9 preferably comprises a transparent support on which the patterns are arranged. This makes it easier to place the alignment mark 9 on the sample 4.

[0052] In a variant not illustrated, the positioning mark consists of two distinct patterns fixed directly, for example by gluing, on the second main face of the sample. In this variant, the patterns can be opaque or generally opaque stickers.

[0053] The digital processing device 20 is configured to digitally align digital images acquired by the image sensor 3 from the alignment mark 9. The image sensor 3 is in fact configured to acquire a digital image of the second main face 7 of the sample 4.

[0054] The digital processing device 20 is configured to implement, from digital images acquired by the image sensor 3, a method for evaluating the clarity of the material.

[0055] A particular embodiment of the method 100 for evaluating the clarity of the material of the sample 4 is shown in [Fig. 5]. The evaluation method 100 firstly comprises a step 110 of providing the sample 4.

[0056] The evaluation method 100 comprises a step of partial backlighting 120 of the sample 4 by the light source 5. To backlight the sample 4, the latter is first positioned and held in a predefined position relative to the light source 5 by the temporary holding means 10. When the light source comprises an illuminated flat surface, the sample 4 is for example held in abutment against this surface.

[0057] The partial backlighting step 120 is implemented by partially obscuring the light source 5 with a mask 8 so as to form on the sample 4, in particular on its second main face 7, a non-illuminated (masked) part, a lit (non-masked) part and an edge separating the non-illuminated part and the lit part.

[0058] The occultation is implemented by positioning the mask 8 between the sample 4 and the light source 5. To do this, the sample 4 is for example moved away from the light source 5. Once the mask 8 is correctly positioned, it is held in a fixed position relative to the sample 4. The sample 4 is for example brought closer to the light source 5 via the temporary holding means 10 until the mask 8 is clamped between the sample 4 and the light source 5.

[0059] At the end of the partial backlighting step 120, the evaluation method 100 comprises a step 130 of acquisition, via the image sensor 3, of a first digital image of the partially backlit sample 4.

[0060] The first digital image 11 acquired in the acquisition step 130 is illustrated in [Fig.l].

[0061] The sample 4 is, in this figure, a wooden plate whose thickness is such that the sample is translucent. The sample or material may for example be transparent, semi-transparent or translucent. The intensity of the light source 5 may be adjusted according to the type of material or the thickness of the plate, so that the image sensor 3 detects sufficient light.

[0062] The first digital image 11 represents the second main face 7 of the sample and comprises an unilluminated part 12, an illuminated part 13 and an edge 14 separating the unilluminated part 12 and the illuminated part 13.

[0063] In practice, the first digital image represents several distinct edges. A thresholding of this first digital image makes it possible to identify and label the unlit part(s) and the lit part(s), then to define a sub-image for each of the edges. Each sub-image is preferably centered on a respective edge and is then used in the remainder of the method.

[0064] The unlit portion 12 shown in the first digital image 11 is homogeneous and generally has a black or close to black color. The lit portion shown in the first digital image 11 is not homogeneous here and has color variations representative of brightness variations. These variations result for example from local variations in transmittance or diffusion within the material of the sample 4. Since these variations are detrimental to the evaluation of the clarity of the material, it is appropriate to avoid them while retaining the information of the material linked to the clarity, in particular the brightness level along the edge 14.

[0065] To do this, the evaluation method 100 comprises a step 140 of generation, by the digital processing device 20, of a digital image corrected by a function of homogenization of the brightness level of the illuminated part of said first digital image 11. The corrected digital image 16 is illustrated in [Fig.7].

[0066] The corrected digital image 16 comprises a dark portion 17 corresponding to the unlit portion 12 of the first digital image 11, a light portion 18 corresponding to the lit portion 13 of the first digital image 11 and an edge 19 corresponding to the edge 14 of the first digital image 11. Unlike the lit portion 13 of the first digital image 11, the light portion 18 of the corrected digital image 16 does not exhibit any variation in the brightness level. In other words, the light portion 18 of the corrected digital image 16 is homogeneous, which makes it possible to evaluate the clarity reliably, in particular using the rise distance method.

[0067] In a preferred embodiment of the invention, the acquisition step 140 comprises a sub-step of total backlighting 141 of the sample 4. The step of re full backlighting 141 is implemented similarly to the partial backlighting step 120 described previously.

[0068] The total backlighting step 141 is implemented by removing the mask 8 so that the light source 5 is no longer obscured by the latter. The removal of the mask 8 can be implemented in the same way as for its positioning for the occultation described previously.

[0069] The evaluation method 100 then comprises a sub-step 142 of acquisition, by the image sensor 3, of a second digital image 15, illustrated in [Fig.6], representing the sample 4 fully backlit.

[0070] The second digital image 15 represents the second main face 7 of the sample 4 and comprises a part (on the right) which is identical to the illuminated part 13 of the first digital image 11.

[0071] The sub-step of acquiring the second digital image can be implemented before the step of acquiring the first digital image. The evaluation method can then comprise, between the sub-step of acquiring the second digital image and the step of acquiring the first digital image, a step of positioning the mask instead of the step of removing the mask.

[0072] The first digital image 11 and the second digital image 15 are each made up of a grid of pixels aligned in columns and rows, the number of pixels depending on the resolution of the image sensor 3. Each pixel comprises information which is representative of the brightness, that is to say varying according to the light transmitted by the material. The information representative of the brightness is preferably the luminance. In this case, the image sensor 3 is calibrated so as to be able to determine the luminance. Of course, any other information proportional to the luminance, or behaving in the same way can be envisaged.

[0073] The generation step 140 further comprises, for each pixel at a given identical position (defined for example by a pair of x and y coordinates) of the first digital image 11 and of the second digital image 15, a sub-step 144 of determining characteristic information of the pixel representative of the brightness.

[0074] The characteristic information associated with a pixel located in the unlit part 12 of the first digital image has, for example, a low value, in particular close to zero. Indeed, the brightness is low or even zero in this part since the light rays are blocked by the mask 8. Conversely, the characteristic information associated with a pixel located in the lit part 13 of the first digital image or with a pixel of the second digital image 15 has, for example, a value greater than zero.

[0075] At the end of the determination sub-step 144, the evaluation method 100 comprises a sub-step 145 of calculating a value of at least one pixel of the corrected digital image 16 by dividing the characteristic information of each pixel at said given position (x, y) of the first digital image 11 by the characteristic information of each pixel at said given position (x, y) of the second digital image 15. In particular, the luminance value for each pixel of the first digital image 11 is denoted Em(x,y), and Ev(x,y) for each pixel of the second digital image 15. The value of the pixels of the corrected digital image 16, denoted Rf(x,y), is generated according to the equation: Rf(x,y) = Em(x,y) / Ev(x,y).

[0076] By dividing the characteristic information associated with a pixel located in the unlit portion 12 of the first digital image 11 by the characteristic information associated with a pixel located at an identical position on the second digital image 15, the value obtained will be close to or equal to zero. Indeed, the information associated with the pixel of the second digital image 15 necessarily has a higher value than the characteristic information associated with the pixel of the first digital image 11, due to the absence of the mask 8. This digital processing thus makes it possible to reconstruct the unlit portion 12 of the first digital image 11 on the corrected digital image 16.

[0077] Furthermore, by dividing the characteristic information associated with a pixel located in the illuminated part 13 of the first digital image 11 by the characteristic information associated with a pixel located at an identical position on the second digital image 15, the value obtained will be close to or equal to one. Indeed, since the second digital image 15 represents the illuminated part 13 of the first digital image 11, the characteristic information associated with each pixel for a given position is identical, provided that the images are correctly aligned. Thus, this determination sub-step 172 makes it possible to overcome the inhomogeneities of the illuminated part 13 of the first digital image 11, while retaining the information on the clarity of the material.

[0078] If the sample 4 is not positioned in the same place on the first digital image 11 and on the second digital image 15, the value obtained may be different or far from one, which may harm the homogeneity of the light part 18 on the corrected digital image 16. When the first digital image 11 and the second digital image 15 are not aligned, for example when there is an offset of at least one pixel, the evaluation method preferably comprises a sub-step 143 of digital alignment of these digital images 11, 15. This sub-step of digital alignment 143 is implemented before the determination sub-step 144.

[0079] The digital alignment 143 is for example implemented using an image alignment algorithm, such as the ORB method (for “Oriented FAST and Rotated BRIEF” in English, with FAST for “Features from Accelerated Segment Test” and BRIEF for “Binary Robust Independent Elementary Features”). The alignment algorithm 143 then uses the alignment marker as a key point to align the first digital image 11 and the second digital image 15 with respect to each other. Of course, the digital alignment can be implemented by any other known technique, and in particular the method known by the English expression “feature based image alignment”.

[0080] At the end of the step 140 of generating the corrected digital image 16, the evaluation method 100 comprises a step 150 of determining the rise distance Dm2 from the corrected digital image 16. The rise distance Dm2 is determined by establishing a spreading function of the edge 19 of the corrected digital image 16.

[0081] [Fig.8] shows an edge spread function along line VIII-VIII in [Fig.7]. Unlike the edge spread function 14 shown in [Fig.2], the edge spread function 19 is not noisy and corresponds to a spread function of a homogeneous material. The rise distance Dm2 can thus be reliably measured along the entire length of the edge 19, which makes it possible to assess the clarity of the material along the edge 19.

[0082] In order to evaluate the clarity of the material more globally, for example over a wider area of the sample 4, the rise distance can be determined at different locations on the sample 4. In particular, the rise distance can be measured for several distinct edges of the mask located at different positions on the sample 4. This makes it possible to evaluate the clarity at different positions of the sample 4.

[0083] After determining the rise distances at different locations of the sample, the evaluation method may comprise a step of calculating the average of the rise distances measured for each edge, in order to obtain an average rise distance used to evaluate the overall clarity of the sample material.

[0084] To further improve the accuracy of the edge rise distance or the average rise distance, the evaluation method 100 preferably comprises a step 160 of correcting the rise distance determined directly from the corrected digital image 16.

[0085] The rise distance may in fact be overestimated due to measurement noise from the image sensor 3, which is generally not perfect. The rise distance is then likely to vary from one image sensor to another.

[0086] To overcome this noise, a reference rise distance is determined from a corrected reference digital image, obtained in the same way as the corrected digital image 16 but in which the sample 4 has been removed. In particular, the corrected reference digital image is generated from a first reference digital image, representing the illuminated light source (without sample) and partially obscured by the mask so as to form on this digital image an illuminated part, an unilluminated part and an edge, and from a second digital image representing the unobscured light source.

[0087] In theory, in the case of a perfect image sensor, the reference rise distance should be zero since, in the first reference digital image, there is no material other than the mask between the light source and the image sensor. Since the image sensor 3 is generally not perfect, the reference rise distance makes it possible to measure the noise coming from this sensor.

[0088] To overcome this, and thus correct the climb distance measured in the determination step 150, it is appropriate to subtract the reference climb distance from this climb distance.

[0089] It may also be envisaged, in order to obtain a corrected value of the average climb distance of sample 4, to subtract an average reference climb distance from the average climb distance.

[0090] The clarity evaluation method according to the invention is thus particularly well suited to reliably and precisely evaluating the clarity of non-homogeneous materials, such as lignocellulosic materials and in particular wood.

[0091] The clarity assessment makes it possible to assign each material or each sample a particular use based on its clarity. This also makes it possible to evaluate the performance of the material, particularly for the production of luminous display devices. This also makes it possible to reliably and precisely characterize optically non-homogeneous materials, for example those of a diffusive nature, as well as luminous display devices camouflaged by such materials.

[0092] The evaluation method can also be used to control the clarity of a display device comprising a non-homogeneous material, backlit by a light source and a mask located between the material and the light source so as to partially obscure the light source. The light source of the luminous display device is used to backlight the material whose clarity is to be evaluated. It is of course preferable that the luminous display device is designed so that the mask can be removed or positioned easily.

[0093] Variants not illustrated are described below.

[0094] In one variant, the mask is produced directly on the sample, in particular by printing.

[0095] In another variant, the mask consists of a substantially opaque plate in which openings are formed forming the transmission zones. The mask may comprise a plate of vinyl, metal, plastic or any other opaque or generally opaque material. The openings may be made by cutting, by laser scraping, by engraving, in particular laser.

[0096] In one variant, the light source is configured to partially backlight the sample and thus form on the sample an unlit portion, an illuminated portion and an edge separating the unlit portion and the illuminated portion.

[0097] In this variant, it is no longer necessary to position a mask between the light source and the sample to partially backlight the sample. This variant therefore makes it possible to avoid using a mask. Of course, the light source is then configured to illuminate partially (i.e. only a part of the surface of the light source) or completely in order to completely backlight the sample. The light source can be a screen of the LCD, TFT, LED, OLED type, or any other similar technology, with a part of the screen which is lit and a part of the screen which is off.

[0098] Thus, the partial backlighting step of the evaluation method is implemented by partial illumination of the light source, for example by illuminating only a part of the light source. In other words, the unilluminated part, the illuminated part and the edge are formed directly by the light source, without the need for occultation of said light source by the mask. The same applies to the determination of the reference rise distance.

[0099] In this variant, the total backlighting step of the evaluation method is implemented by total illumination of the light source, without the need to remove the mask since the latter is not necessary for occultation.

[0100] In one variant, the light source may be a video projector.

[0101] In a variant, the step of generating the corrected digital image is implemented work by blurring at least the illuminated part of the first digital image 11. This has the effect of homogenizing the brightness level of this part. A brightness threshold can be established so as to discriminate between the unlit part and the lit part.

[0102] A blur function can thus be implemented in different ways. For example, the blur function can be implemented in particular with a sliding blur kernel (or convolution matrix) of variable shape and dimension or with different known blurring methods such as the mean, the median or the Gaussian. This homogenization function is relatively simple to implement, but does not offer as much precision as the function described in the context of the preferred embodiment.

[0103] Of course, the embodiments described above are in no way limiting.

Claims

Claims

1. Method for evaluating (100) the clarity of a material, such as a lignocellulosic material, comprising the steps of: - providing (110) a sample (4) of said material, - partially backlighting (120) the sample (4) by a light source (5) so as to form on said sample (4) an unlit portion (12), an illuminated portion (13) and an edge (14) separating the unlit portion (12) and the illuminated portion (13), - acquiring (130) a first digital image (11), by an image sensor (3), of the partially backlit sample (4), characterized in that it further comprises the steps of: - generating (140) a corrected digital image (16) by a function for homogenizing the brightness level of the illuminated portion of said first digital image (11), and - determining (150) of a rise distance (Dm2) from the corrected digital image (16) to assess the clarity of said material.

2. Method according to claim 1, in which the step of generating (140) a corrected digital image (16) comprises a sub-step of total backlighting (141) of the sample (4) and a sub-step of acquiring (142) a second digital image (15), by the image sensor (3), of the totally backlit sample (4), the corrected digital image (16) being generated from the first digital image (11) and the second digital image (15).

3. Method according to claim 2, in which the function of homogenizing the brightness level of the illuminated part consists of dividing characteristic information of each pixel at a given position of the first digital image (11) by said information of each pixel at said given position of the second digital image (15), said characteristic information being representative of the brightness.

4. Method according to claim 3, in which the characteristic information of the pixel representative of the brightness is the luminance.

5. A method according to any one of claims 2 to 4, wherein the generating step (170) further comprises a sub-step of digitally aligning (143) the first digital image (11) and the second digital image (15).

6. A method according to any one of claims 1 to 5, comprising in furthermore, after the determining step (150), a step of correcting (160) the rise distance by subtracting from the rise distance (Dm2) determined in the determining step (150) a reference rise distance, the reference rise distance being determined from a first reference digital image representing the light source (5) partially illuminating so as to form an illuminated part, an unilluminated part and an edge separating the unilluminated part from the illuminated part.

7. Method according to any one of claims 1 to 6, in which the first digital image (11) comprises several distinct edges (14), the determining step (150) being implemented for each edge (14), and the evaluation method (100) further comprising, after said determining step (150), a step of calculating the average of the ascent distances (Dm2) determined for each edge (14).

8. A method according to any one of claims 1 to 7, wherein said material is a non-homogeneous material, such as a lignocellulosic material, and preferably wood.

9. Use of the method according to any one of claims 1 to 8 for controlling the clarity of a display device comprising a non-homogeneous material backlit by a light source and a mask located between the material and the light source so as to partially obscure the light source.

10. Device for evaluating the clarity of a material, the device (1) comprising a sample (4) of said material, a light source (5) configured to backlight the sample (4), an image sensor (3) configured to provide digital images of said sample (4), and a digital processing device (20) configured to implement the evaluation method according to any one of claims 1 to 8, from digital images acquired by the image sensor (3).

Images