Method for generating a data set for identifying wood-based panels and method for identifying wood-based panels

Abstract

The invention relates to a method for generating a data set for identifying wood-based panels, wherein: at least two wood-based panels, in particular a plurality of wood-based panels, form a panel package; at least one digital image of a part of the panel package with sections of the side edges of the wood-based panels is recorded; image sections are determined from the digital image with the sections of the individual side edges; a comparison data set for each image section is determined using the structure of the wood-based elements; the comparison data set of each wood-based panel is stored; and at least one piece of information regarding the respective wood-based panel and one piece of information regarding the respective panel package are linked to the comparison data set of each wood-based panel and stored in a database. The invention further relates to a method for identifying a wood-based panel.

Description

[0001] September 2, 2025

[0002] Method for generating a data set for identifying wood-based panels and method for identifying wood-based panels

[0003] The invention relates to a method for generating a data set for identifying wood-based panels and a method for identifying wood-based panels.

[0004] Wood-based panels, especially particleboard, OSB (oriented strand board), or fiberboard such as HDF (high-density fiberboard) or MDF (medium-density fiberboard), are manufactured on an industrial scale. Manufacturing parameters such as the type of processing and origin of the wood used, including wood-based elements like chips or fibers and their size distribution, the type of binders and any additives used, and the process temperatures and pressures are critical for the quality of the wood-based panels. Logistical information such as the date of manufacture, place of manufacture, transport routes, and storage facilities also plays a role.

[0005] Due to the increasing interconnectedness of distribution channels and the increasing demands on quality and origin certificates, it is desirable to be able to assign the manufacturing parameters and logistical information to each individual wood-based panel.

[0006] From the prior art of WO 2022 / 223563 Al, it is known to use an algorithm, which is also at least partially executed by artificial intelligence, to utilize the structure of a side edge of a wood-based panel for its unique identification. For this purpose, the structure of the side edge is captured and evaluated using a digital image based on the geometry, topography and / or the distribution of the wood-based panel elements such as chips or fibers.

[0007] The resulting data set is stored as a reference record in a database. Then, if the edge of the same wood-based panel is photographed at a later time, a comparison can be made with a test record derived from this image. If the comparison shows a match between the test record and one of the stored reference records, the wood-based panel can be identified. Thus, beyond identification, further characterization information can be assigned to the identified wood-based panel, including its properties, manufacturing data, and other information.

[0008] Creating a digital image of the edge of wood-based panels during the manufacturing process requires holding the panel in position relative to the camera or at least guiding or aligning it precisely. Furthermore, areas where the wood-based panels are handled are more prone to interfering contamination, which can impair image quality and lead to inaccurate results.

[0009] Therefore, the present invention aims to further improve and simplify the identification of wood-based panels.

[0010] The previously identified problem is solved according to the invention by a method for generating a data set for identifying wood-based panels, in which at least two wood-based panels, in particular several wood-based panels, form a panel package, in which at least one digital image of a part of the panel package with sections of the side edges of the wood-based panels is taken,

[0011] SE / sw 240939WO

[0012] September 2, 2025, in which image sections from the digital image are determined with the sections of the individual side edges, in which a comparison data set for each image section is determined based on the structure of the wood-based material elements, in which the comparison data set of each wood-based panel is stored, and in which at least one piece of information about the respective wood-based panel and the respective panel package is linked to the comparison data set of each wood-based panel and stored in a database.

[0013] A panel package consists of at least two, and in particular several, wood-based panels. The panels, which are usually stacked horizontally, define a horizontal and a vertical direction. A panel package typically contains 10 to 30 panels, depending on their thickness and the order quantity. This is because the wood-based panels can have different thicknesses and / or qualities. For transport, the wood-based panels are usually bundled into panel packages and secured to form a transportable unit using straps and / or other packaging materials.

[0014] Then, at least one digital image is captured of a portion of the panel stack, including sections of the side edges of the stacked wood-based panels. This single digital image is then analyzed by extracting image sections containing the individual side edge segments. For each image section, a comparison dataset is generated based on the structure of the wood-based elements using at least one algorithm.

[0015] The wood-based panels can be, in particular, chipboard, OSB (oriented strand board - OSB) or fiberboard such as HDF (high density fiber - HDF) or MDF (medium density fiber - MDF).

[0016] SE / sw 240939WO

[0017] September 2, 2025 For the later use of the comparison datasets, it is necessary that the position of the section of the wood-based panel's side edge used to create the comparison dataset is known or identifiable on the panel pack. This section could, for example, be identified by the type of packaging. In particular, straps used to secure the wood-based panels can serve as a marker for determining the relevant section. Alternatively, a digital image can be taken from all sides, but this significantly increases the effort involved.

[0018] The structure is determined based on the geometry and / or distribution of the wood-based material elements. The structure is defined, for example, by the form properties of the edge, such as the properties of the wood-based material elements themselves, including their size, their individual and / or collective orientation, size variability, the existence and / or number of different wood-based material element types, the dimensions of the gaps between wood-based material elements, and the presence and distribution of unfilled spaces between them. It is not only possible to distinguish individual wood-based material elements, but rather to consider the overall appearance of the edge as the structure. Alternatively or additionally to the form properties, the structure can also be determined by the color and / or brightness of the edge.This can be the base color and base brightness of the side edge, but also the variability of the color and brightness, the color of the wood-based material elements or the additives.

[0019] This method exploits the fact that the visible structure of the wood-based material, particularly on a side edge, consisting of the wood-based material elements and optionally also of the resins and possibly additives that bind them, can be influenced by both different starting materials and by random variations in the properties of the wood-based material that are not controlled during the manufacturing process.

[0020] SE / sw 240939WO

[0021] The diverse structure of the product, as of September 2, 2025, allows it to be used as an identification feature. Therefore, without the need for labels, printed codes, or other markings, individual sections of the side edges can be used as identification features.

[0022] The wood-based material elements can be, in particular, chips or fibers. Accordingly, the comparison dataset can be determined for each image section based on the structure, especially the geometry and / or distribution, of the chips or fibers. For example, the comparison dataset can be determined based on the orientation and / or size distribution of the chips or fibers.

[0023] The comparison data set is preferably determined as a so-called hash value. After processing the digital image section, the hash value assigns a unique alphanumeric value to its content. Search algorithms can then use this alphanumeric value to identify the content of another digital image at a later time.

[0024] Furthermore, the comparative data record of each wood-based panel is stored, and at least one piece of information about the respective wood-based panel and its respective panel package is linked to this comparative data record and stored in a database. For example, an identification number of the panel package can be stored as information, so that the panel package can be identified by identifying the wood-based panel. And the other wood-based panels that are also stacked in the same package can then be identified using the panel package identification. This is because the specific package in which the wood-based panels will be stacked is determined during the production process.

[0025] Further information for characterizing the wood-based panel may include: - Material information on the type of wood or wood species used in the wood-based materials,

[0026] SE / sw 240939WO

[0027] September 2, 2025 - Material information on the formulation of the binders and other additives used,

[0028] - technical data sheets and laboratory protocols,

[0029] - Manufacturing parameters such as temperature, pressure and time profile of the pressing process,

[0030] - Production date, production time, production plant and production location,

[0031] - Dimensions or format, in particular external dimensions and layer thicknesses,

[0032] - Type of applied coating or decoration,

[0033] - Certifications of the board, for example, certification for the wood used according to the FSC and PEFC standards for sustainable forestry and corresponding ISO standards,

[0034] - Certifications of the board regarding the emission of volatile substances according to TSCA Title VI, CARB2 and CE,

[0035] - Storage locations and storage times, as well as transport routes and transport times,

[0036] - Authorization information for subsequent retrieval of the data records

[0037] By analyzing a digital image of a package of panels, the wood-based panels contained within the package are identified. The individual edge segments are separated from the image sections and used for the unique identification of the wood-based panels in the image. Therefore, it is sufficient to create a digital image of an entire package of panels or a section of a package and generate a hash value for each individual panel from this image.

[0038] If information about a wood-based panel needs to be obtained at a later time, a digital image of its side edge or a section thereof can be used to determine an associated test data record, such as a corresponding hash value. Based on this test data record, the data linked to the identified wood-based panel can then be retrieved from the database.

[0039] In a preferred embodiment of the method, the image sections are combined with the sections of the horizontally running side edges of the vertically superimposed

[0040] SE / sw 240939WO

[0041] Wood-based panels ordered on September 2, 2025, were detected in the digital image by means of horizontally running differences in the image data.

[0042] The separation of individual layers of wood-based panels in the digital image can therefore be achieved through visually perceptible differences across the respective panel thicknesses. In particleboard, such visually perceptible differences arise from the varying porosity of the core and surface layers. However, separation could also be performed based on other properties, such as determining the different densities of the core and surface layers, or by visually capturing the cross-sections of coatings applied to the top and bottom surfaces of the panels. The individual wood-based panels can be positioned close together or spaced apart.

[0043] For transporting larger quantities of wood-based panels, at least two packages of panels are stacked on top of each other to form a panel stack. Several packages of panels can therefore be stacked on top of each other to create panel stacks that can be moved as a whole, for example, to transport them from a warehouse to a truck or a railway car. In the described process, the information for identifying the panel stack is linked to the comparison data set of the wood-based panels and additionally stored in the database.

[0044] In a further preferred method, the digital image of the panel package or stack is captured immediately before packaging. At this point, the relevant information about the individual wood-based panels, as well as the information necessary for further handling of the panel package, is available from the panel manufacturer's MES (Manufacturing Execution System). Each wood-based panel can thus be linked to data about the panel itself and to data about the panel package in which it will be transported.

[0045] SE / sw 240939WO

[0046] September 2, 2025. By separating the wood-based panels within a panel package or stack, the number of panels per package can be determined using the digital image. This number can then be used to verify the specified quantity. If the determined number does not match the specified number in the panel package, implausible discrepancies exist, requiring manual intervention. This allows for a plausibility check to be performed even after the panel package or stack has been packaged.

[0047] The problem outlined above is also solved by a method for identifying a wood-based panel, in which at least two wood-based panels, in particular several wood-based panels, form a panel package, in which at least one digital image of a part of the panel package with sections of the side edges of the wood-based panels is taken, in which at least one image section with the section of a side edge of at least one wood-based panel is determined from the digital image, in which the image section is determined based on the structure of the

[0048] For wood-based panel elements, a test data set is determined, in which the test data set is compared with the comparison data sets stored in a database, and in which, if a comparison data set matches the test data set, the associated wood-based panel is identified.

[0049] The wood-based panels can be, in particular, chipboard, OSB (oriented strand board - OSB) or fiberboard such as HDF (high density fiber - HDF) or MDF (medium density fiber - MDF).

[0050] SE / sw 240939WO

[0051] September 2, 2025 The structure of the wood-based material elements is analyzed in the same way as when generating the comparison data set, based on the geometry and / or distribution of the wood-based material elements.

[0052] The wood-based material elements can be, in particular, chips or fibers. Accordingly, the test data set can be determined for each image section based on the structure, especially the geometry and / or distribution, of the chips or fibers. For example, the determination of the comparison data set can be based on the orientation and / or size distribution of the chips or fibers.

[0053] Furthermore, the test data set is again preferably a hash value.

[0054] The described procedure preferably takes place after the wood-based panels or panel packages have been delivered to an intermediary or an end customer. For this purpose, a mobile camera device, in particular a smartphone, is preferably used, whereby the recorded digital image is displayed directly to the user for evaluation.

[0055] In a preferred embodiment of the method, the distance between a camera device and the panel stack is determined, the thickness of at least one wood-based panel is determined from the distance and the optical properties of the camera device, and the thickness of the wood-based panel is used as a selection criterion for the comparison step with the comparison data sets.

[0056] The characteristics of the camera system include the focal length, the size of the optical sensor, and the size of the sensor's pixels, from which, at a given distance to an object, the dimensions, in this case the thickness of a photographed wood-based panel, can be determined from the pixel values ​​of the digital image.

[0057] SE / sw 240939WO

[0058] September 2, 2025 To determine the distance, a LIDAR (Light Detection And Ranging) sensor can be used together with the camera setup to measure the distance between the camera setup and the wood-based panel.

[0059] Determining the thickness of the wood-based panel reduces the computational effort required to compare the test data set with the comparison data sets stored in a database, because the search can be limited to comparison data sets on wood-based panels with the already determined thickness.

[0060] Preferably, the assigned panel package is identified together with the wood-based panel. This allows all other wood-based panels to be specified and identified via their stored position within the panel package.

[0061] Furthermore, it is possible to identify multiple wood-based panels, in particular all identifiable wood-based panels from at least one panel package. This allows these wood-based panels to be identified directly, rather than indirectly via their stored association with the panel package.

[0062] Furthermore, if at least two packages of panels form a stack, the stack can also be identified by identifying one of the wood-based panels in one of the packages. This is because the comparison data set can also store information about which stack of panels each wood-based panel is contained in.

[0063] After packaging the panel pack or stack, the side edges of the wood-based panels are usually at least partially obscured for some of them. However, knowing the structure of the panel pack or stack, it is then sufficient to determine the test data set or hash value of a single wood-based panel from the entire stack to...

[0064] SE / sw 240939WO

[0065] to be able to determine on September 2, 2025, which other wood-based panels are present in this stack.

[0066] This eliminates the need for a shipping label for each record package, which would otherwise have to be permanently attached. At the same time, the problem of lost shipping labels is solved, and it also becomes possible to automatically determine the identity of a record package. Determining the identity of a record package is generally time-consuming with shipping labels, as these are not always automatically readable or are often attached to different parts of the record packages – depending on customer requirements.

[0067] This is particularly advantageous for automated storage and handling systems, as well as for the operators of transport vehicles. The operators can then concentrate more on the safe handling of the heavy stacks of panels. For this purpose, the handling vehicles can be equipped with at least one suitable camera that digitally records the known location or corner of the panel stack and sends the captured digital images for analysis.

[0068] Automated package recognition can also be used to implement developments known as the "Internet of Things" or "IoT." This allows for the automated provision of information, such as the precise type and characteristics of the packaging required for the package. Subsequently, control systems can be used to automatically determine whether the packaging matches the intended configuration. For processors of wood-based panels, this information can include details about the type of wood-based panels contained in the package, as well as whether any of the panels have a quality defect that must be taken into account during processing.

[0069] SE / sw 240939WO

[0070] September 2, 2025 In addition to separating wood-based panels from a stack, it is also possible to separate different stacks that are in a picture from each other.

[0071] To monitor the loading of stacks of panels onto a truck or railcar, it is sufficient to take an image showing at least a section of one edge of each stack, which is then used to create inspection records. With sufficient image resolution, it is possible to determine how many stacks are loaded, which stacks are loaded, and therefore which wood-based panels are contained within the stacks and thus on the truck or railcar. This information can then be used for automated delivery note generation at the wood-based panel manufacturer or for automated goods receipt recording at the wood-based panel processor.

[0072] To monitor the inventory of wood-based panels in a warehouse, it is sufficient to take a photograph of each stack of panels. Creating a list of all package numbers is then no longer necessary.

[0073] In a further embodiment of the described method, the distance between a camera device and the panel package is determined, the thickness of at least one wood-based panel is determined from the distance and the optical properties of the camera device, and the thickness of the wood-based panel is used as a selection criterion for the comparison step with the comparison data sets.

[0074] The invention will now be explained using exemplary embodiments with reference to the drawing. The drawing shows...

[0075] SE / sw 240939WO

[0076] September 2, 2025 Fig. 1-4 a schematic representation and tables for an embodiment of a method according to the invention with application to a section of a side edge of a wood-based panel and evaluation of geometric parameters of the chip distribution,

[0077] Fig. 5 shows a schematic representation of a further embodiment of a method according to the invention using four sections of a side edge of a wood-based panel.

[0078] Fig. 6 shows a schematic representation of a further embodiment of a method according to the invention, applied to a section on a side edge of a wood-based panel.

[0079] Figs. 7-9 Illustrations to explain a method according to the invention for generating a data set for identifying wood-based panels and

[0080] Figs. 10-12 Illustrations to explain a method according to the invention for identifying a wood-based panel.

[0081] In the following description of the various embodiments according to the invention, components and elements with the same function and mode of operation are provided with the same reference numerals, even if the components and elements may differ in their dimensions or shape in the various embodiments.

[0082] First, the method known from WO 2022 / 223563 Al for generating comparison data sets and test data sets is explained with reference to Figs. 1 to 6, before embodiments of the methods according to the invention are explained.

[0083] SE / sw 240939WO

[0084] September 2, 2025 Figures 1 to 4 show a first embodiment of a method for identifying wood-based panels.

[0085] Fig. 1 shows a side view photograph of a section of a side edge 2 of a wood-based panel. A right-angled corner 4 is located at the right edge of the view, which can be used as a reference for a dimension "da" to determine the position of a section 6. Section 6 is shown with a dashed guideline, which is not, however, on the side edge itself. The side edge 2 is not printed or otherwise marked. Fig. 2 shows section 6 in a separate view.

[0086] In a further step of the process, a digital image is captured and analyzed as a comparison image for section 6 of side edge 2 using a camera (not shown). Conventional image analysis software such as OpenCV, scikit-image, Google Cloud Vision API, Microsoft Computer Vision API, vlfeat, dlib, Matlab Image Processing Toolbox, etc., is used to detect a number of individual wood chips. The outlines of the detected chips are partially determined by the software based on predefined criteria such as size or area. This is illustrated by a long chip in Figures 1 and 2, whose area is marked by a constant gray color.

[0087] As shown in Fig. 2 for the selected chip, the detected chips are measured according to length dl and angle a of the orientation of the chips with respect to the direction of extension (horizontal in Fig. 2), the corresponding data for N=20 chips are partially listed in a table in Fig. 3.

[0088] Fig. 4 shows the graphical representation of the data according to the table in Fig. 3, where the length dl in the x-axis direction and the angle a relative to the x-axis are plotted as vectors. The corresponding vectors vi, V2, vs, and V20 are shown in Fig. 4. The vectors are projected onto the x-axis, taking into account the cosine value of the angle a and calculating the absolute value.

[0089] SE / sw 240939WO

[0090] On September 2, 2025 (amount), the values ​​Amount dx in the right-hand column of the table in Fig. 3 will then result.

[0091] One advantage of calculating the absolute value of dx(N) is that the value is the same regardless of the orientation of the wood-based panel when the image is taken.

[0092] Subsequently, the sum of the absolute values ​​dx over N=1 to 20 yields a specific value for section 6, which enables the identification of the wood-based panel. Thus, in the presented and described method, a comparative data set, here the sum of the values ​​N=1 to 20 over the absolute values ​​dx(N) as well as the individual values ​​dx(N) from the comparison image, is determined and stored based on the geometry of the wood-based elements of the chips.

[0093] At least at a later point in time after production, for example during storage in the manufacturing plant, during transport and further storage and / or during the use of the wood-based panels for further processing, for example as furniture panels, a digital image is taken again as a test image for the specified section of the side edge of a wood-based panel.

[0094] The digital test image is subjected to the same analysis as previously explained with reference to Figures 1 to 4, and a test data set is determined from the test image based on the geometry of the wood-based panel elements (chips, fibers) in the same way as for the comparison data sets. The test data set is then compared with the comparison data sets for the numerous wood-based panels stored in the database. If the test data set matches a comparison data set, the tested wood-based panel is identified as the wood-based panel for which the comparison data set was generated.

[0095] SE / sw 240939WO

[0096] September 2, 2025. The test image is preferably taken with the same camera and under identical lighting conditions. However, the method can also be robust enough to withstand fluctuations that, for example, a permanently installed camera takes the comparison image after the wood-based panel has been manufactured, while the test image can be taken at a later time with a mobile camera, in particular also with a mobile phone camera.

[0097] The database is generally stored on a server that is connected to the computing units for creating comparison data sets and the computing units for creating test data sets via a local network of the manufacturing site and / or via the Internet, enabling data exchange.

[0098] Fig. 5 shows a further embodiment of a method for identifying wood-based panels, in which, in comparison to the example according to Figs. 1 to 4, a total of four sections 6a, 6b, 6c and 6d are defined, which are spaced apart from the corner 4 at distances dai, da2, das and da4.

[0099] Subsequently, comparison images are taken for the four sections 6a, 6b, 6c and 6d along side edge 2, and a sum is calculated for each section 6a, 6b, 6c and 6d in the manner previously explained. a , b, c and d were formed. The numerical values ​​of the four sums a , b, c and d are then written in the order of the sections into a 4-dimensional vector V, which forms the comparison data set.

[0100] In a subsequent inspection of a wood-based panel, the four sections positioned in the specified segments dai, das, das, and da4 along a given side edge are photographed in the same manner and evaluated as described previously, generating a vector V as a test data set. This test data set is then compared with the reference data sets stored in the database. If there is a match between the test data set and one of the reference data sets, the tested panel is then marked as approved.

[0101] SE / sw 240939WO

[0102] September 2, 2025: Wood-based panel identified as the wood-based panel for which the comparison data set was generated.

[0103] The comparison data set and the test data set are thus a sequence of structural values. a , b, c and d were determined for a given proportion of the wood-based material elements.

[0104] The digital images, i.e., the comparison images and the test images, can be captured with conventional cameras, such as CCD cameras, digital single-lens reflex cameras (DSLRs), mirrorless system cameras (DSLMs), or cameras from mobile devices like cell phones or tablets. A digital image can also be acquired with a spectral camera that determines spectral data for each pixel. The image can be stored in the RGB color space with values ​​for red (R), green (G), and blue (B), or in the L*a*b* color space with pixel values ​​L*, a*, or b* for brightness (L*j), red-green color intensity (a*j), and blue-yellow color intensity (b*j), or in other color spaces.

[0105] Fig. 6 shows an embodiment in which the side edge 2 of the wood-based panel is recorded in the area of ​​a corner 4 for structural analysis. The position of the corner 4 can be easily determined visually and can therefore be readily used as a measure for defining the area of ​​the side edges to be analyzed.

[0106] Figures 7 to 9 show an embodiment of a method according to the invention for generating a data set for identifying wood-based panels.

[0107] Figure 7 shows part of a production area in a manufacturing plant for wood-based panels 12. To the left of a roller door 14 is the production line for the wood-based panels 12, at the end of which multiple wood-based panels 12 are stacked to form panel packs 16. The panel packs 16

[0108] SE / sw 240939WO

[0109] 2. September 2025 are transported by a conveyor belt 18, which transports the plate packages 16 in Fig. 7 from left to right in several stages one after the other.

[0110] The horizontally stacked wood-based panels 12 define the horizontal and vertical directions used in the following description. The relevant side edges 20 of the wood-based panels 12 can then be seen in the side views.

[0111] Fig. 7 shows a complete stack of panels 16, the movement of which has been paused for a short time. Behind the stack of panels 16 is a camera device 22, the setup and position of which relative to the stack of panels are shown in Fig. 8. This camera device 22 captures at least one digital image of a portion of the stack of panels 16, including sections of the side edges 20 of the stacked wood-based panels 12.

[0112] After the creation of at least one digital image, the stack of plates 16 is transported further for processing in the form of packaging and storage.

[0113] Fig. 8 shows the camera device 22 in a perspective view with a lighting screen 24 and a centrally arranged digital camera 26, which has an optical sensor with a resolution of at least 8 megapixels, preferably at least 12 megapixels and more preferably at least 24 megapixels.

[0114] Fig. 9 shows the arrangement from a different view, in which the area of ​​the side edges 20 captured by the digital camera 26 can be seen.

[0115] The dashed lines in Figs. 8 and 9 each mark the observation area of ​​the side edges 20 of the stacked wood-based panels 12. A respective digital image extends over the dashed lines.

[0116] SE / sw 240939WO

[0117] September 2, 2025 beyond the boundaries in order to fully capture the part of the stack of plates 16 to be mapped.

[0118] The digital camera 26 can capture stacks of plates 16 of, for example, up to 150 cm in height, whereby the captured images have sufficient resolution despite the necessary distance to the stack of plates to enable a reliable evaluation afterwards.

[0119] After the digital image has been captured, the stack of plates 16 in Fig. 7 is then transported further to the right.

[0120] The image analysis is performed using an evaluation unit not shown in the figures. Using evaluation software, image sections from the digital image are determined based on the segments of the individual side edges 20.

[0121] In the present embodiment, only an edge region of a stack of panels 16 is used for the identification of the wood-based panels 12. It is advantageous to use a digital camera 26 with an image sensor that has a high aspect ratio. With the same resolution of the camera, a high density of image information can thus be provided over a narrow but tall area for calculating a comparison data set in the form of a hash value.

[0122] Furthermore, it has proven advantageous to illuminate the stacks of plates 16 homogeneously for photographing them. Therefore, an LED light panel with a height approximately equal to the height of the plate stacks to be photographed is used as a light shield 24. A recess is provided in the center of the light shield 24 in which the digital camera 26 is positioned.

[0123] SE / sw 240939WO

[0124] September 2, 2025. Using the evaluation software, a comparison data set for each image section is determined from the digital image based on the structure of the wood-based material elements, as previously explained in connection with Figures 1 to 6. The comparison data sets are calculated as hash values.

[0125] Each comparison data record is unique for each of the image sections of the wood-based panels 12 and is stored for each wood-based panel 12. At least one piece of information about the respective wood-based panel and the respective panel package is linked to the comparison data record of each wood-based panel 12 and stored in a database. Each panel package 16 receives a unique identification number during production, even before stacking. Additionally, the wood-based panels 12 to be produced, or those just produced, are assigned to a specific panel package 16 during production. If the database system stores information about which wood-based panels 12 comprise a panel package 16, this information can be accessed later.

[0126] The image sections containing the sections to be evaluated on the horizontally extending side edges 20 of the vertically stacked wood-based panels 12 are identified in the digital image by means of horizontally extending differences in the image data. These differences are based on varying densities and / or porosities of the core and surface layers of the individual wood-based panels 12, or on coatings on the top and bottom surfaces of the panels. The evaluation software uses an automatic selection tool, similar to those found in image editing software such as Photoshop.

[0127] In the production of wood-based panels 12, it is common practice to arrange several panel packages 16 on top of each other for transport, thus forming a panel stack 30, which is shown in Fig. 10.

[0128] SE / sw 240939WO

[0129] September 2, 2025 In this application of the procedure, the information for identifying the stack of panels 30, usually an identification number, is linked to the comparison data record and also stored in the database. This allows a single wood-based panel 12 to be assigned not only to a package of panels 16, but also to a stack of panels 30.

[0130] Figures 10 to 12 illustrate a method according to the invention for identifying a wood-based panel 12.

[0131] Fig. 10 shows a stack of panels 30 with a total of four panel packages 16, each containing a plurality of wood-based panels 12. At the top edge of each panel package 16, a horizontal section of packaging 32 can be seen, which laterally covers several wood-based panels 12.

[0132] Also shown is a user's mobile phone 34, whose camera captures a digital image of part of one of the panel packages 16, showing sections of the side edges 20 of the stacked wood-based panels 12. Using a program installed on the mobile phone 34, the wood-based panels 12 of the panel package 16 are separated. The program utilizes the differences in the layer structure of the individual wood-based panels 12, which are also clearly visible to the naked eye. With appropriate image analysis and image processing software, the interfaces between the individual wood-based panels 12 can be highlighted.

[0133] Fig. 11 shows how the user selects one of the recorded wood-based panels, so that a section of a side edge 20 of a wood-based panel 12 is determined from the digital image and thus the image section of interest is defined.

[0134] For this image section, a test data set, in particular a hash value, is then generated in the manner described above based on the structure of the wood-based material elements.

[0135] SE / sw 240939WO

[0136] The test data set will be determined on September 2, 2025. The test data set will then be compared with the comparison data sets stored in the database, and if a comparison data set matches the test data set, the wood-based panel assigned to the image section will be identified.

[0137] By selecting a wood-based panel 12 using the touch function on the mobile phone, it is marked and a connection is established with the database in which all panel data with hash values ​​of all wood-based panels are stored. Depending on requirements, specific data can then be returned for the selected wood-based panel 12, the panel package 16 and / or the panel stack 30.

[0138] In the upper area of ​​the display of the mobile phone 34 shown in Fig. 11, which is shown enlarged in Fig. 12, data for the identified wood-based panel 12 are displayed. In this case, this includes the stack number 4708242011***2***25***25 and the panel thickness of 18.00 mm. Additionally, data relating to the database query process is displayed.

[0139] Thus, the assigned panel package 16 is identified together with the wood-based panel 12, so that all wood-based panels 12 of the panel package 16 are also identified. For this purpose, the wood-based panels 12 of the associated panel package 16 are determined. This also allows wood-based panels 12 that are obscured by the horizontal section of the packaging 32 to be identified.

[0140] The camera of the mobile phone 34 can also have an integrated LiDAR sensor, where L1DAR stands for Light Detection and Ranging. The LiDAR sensor makes it possible to determine the distance between the camera and the side edge 2. From this, the thickness of the wood-based panel can be calculated. This has the advantage that the determined hash value only needs to be compared with the hash values ​​for the corresponding panel thickness stored in the database. This reduces the time required for comparing the hash values.

[0141] SE / sw 240939WO

[0142] September 2, 2025

Claims

September 2, 2025 Patent claims 1. Method for generating a data set for identifying wood-based panels, wherein at least two wood-based panels, in particular several wood-based panels, form a panel package, wherein at least one digital image of a part of the panel package with sections of the side edges of the wood-based panels is taken, wherein image sections from the digital image with the sections of the individual side edges are determined, wherein a comparison data set for each image section is determined based on the structure of the wood-based elements, wherein the comparison data set of each wood-based panel is stored, and wherein at least one piece of information about the respective wood-based panel and about the respective panel package is linked with the comparison data set of each wood-based panel and stored in a database.

2. Method according to claim 1, wherein the image sections with the sections of the side edges of the vertically arranged wood-based panels are recognized in the digital image by means of horizontally running differences in the image data.

3. Method according to claim 2, wherein different porosities in middle and top layers, different densities of the middle and top layers and cross-sections of coatings applied to the top and bottom surfaces of the plates are used as differences in the image data.

4. Method according to one of claims 1 to 3, wherein at least two plate packages are arranged one above the other and form a plate stack, and wherein the information for identifying the plate stack is linked to the comparison data set and stored in the database.

5. Method according to any one of claims 1 to 4, wherein the digital image of the disk package or disk stack is taken immediately before packaging the disk package or disk stack.

6. Method for identifying a wood-based panel, wherein at least two wood-based panels, in particular several wood-based panels, form a panel package, wherein at least one digital image of a part of the panel package with sections of the side edges of the wood-based panels is taken, wherein at least one image section with the section of a side edge of at least one wood-based panel is determined from the digital image, wherein a test data set is determined for the image section based on the structure of the wood-based elements, wherein the test data set is compared with the comparison data sets stored in a database, and wherein, if a comparison data set matches the test data set, the associated wood-based panel is identified.

7. Method according to claim 6, wherein the distance between a camera device and the panel stack is determined, wherein the thickness of at least one wood-based panel is determined from the distance and from the optical properties of the camera device and SE / sw 240939WO September 2, 2025 where the thickness of the wood-based panel is used as a selection criterion for the comparison step with the comparison data sets.

8. Method according to claim 6 or 7, wherein the associated panel package is identified together with the wood-based panel.

9. Method according to one of claims 6 to 8, wherein several wood-based panels, in particular all wood-based panels of the at least one panel package, are identified.

10. Method according to any one of claims 6 to 9, wherein, when at least two plate packages form a plate stack, the plate stack is also identified. SE / sw 240939WO 2. September 2025

Images