Method and system for making ceramic articles with veins extending through the thickness
Patent Information
- Application Number
- PCT/IB2026/052884
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-03-26
- Filing Date
- 2026-03-25
- Publication Date
- 2026-10-01
Smart Images
Figure IB2026052884_01102026_PF_FP_ABST
Abstract
Description
[0001] TRANSLATION (RULE 12.3) 25 MAR. 2026
[0002] "METHOD AND SYSTEM FOR MAKING CERAMIC ARTICLES WITH VEINS EXTENDING THROUGH THE THICKNESS"
[0003] Cross-Reference to Related Applications
[0004] This Patent Application claims priority from Italian Patent Application No.
[0005] 102025000006174 filed on March 26, 2025, the entire disclosure of which is incorporated herein by reference.
[0006] Technical Field
[0007] The invention relates to a method for producing ceramic articles, in particular ceramic slabs or tiles, having veins imitating natural stone veins, wood veins or customised decorations and extending through the thickness of the ceramic articles, and to a corresponding plant for making ceramic articles.
[0008] In particular, the invention advantageously, though not exclusively applies to the production of ceramic articles with veins extending through the thickness of the ceramic articles, to which explicit reference will be made in the description below without because of this lacking generality.
[0009] Background
[0010] In the field of the production of ceramic articles, in particular slabs or tiles, it is known to use plants for manufacturing ceramic articles that imitate natural stones, such as marble and / or granite. These ceramic articles have inner veins distributed randomly throughout the thickness of the products.
[0011] In particular, a plant for manufacturing ceramic articles comprises a conveyor unit, a feeding unit for feeding semi-dry ceramic powders, i.e. with a moisture content of about 5-6%, onto an area of the conveyor unit while the conveyor unit conveys the received ceramic powders so as to form a layer of ceramic powders, a continuous compacting unit for compacting the layer of ceramic powders while it is conveyed by the conveyor unit so as to obtain a compacted layer of ceramic powders, a cutting unit for transversely cutting the compacted layer of ceramic powders while it is conveyed by the conveyor unit so as to obtain slabs, and a firing kiln for sintering the compacted layer of ceramic powders of the slabs so as to obtain the ceramic articles.
[0012] Plants are known, which are aimed at producing slabs with a random distribution of ceramic powders of different colours also using continuous compacting units, which comprise a conveyor unit for substantially continuously transporting the powder material along a given path through a work station, in the area of which a compactingdevice is located, which is designed, through the cooperation of pressure rollers, to compact the powder material so as to obtain a layer of compacted powder.
[0013] It is also known to create a graphic decoration over the layer of compacted ceramic powder, for example by digital printing, so as to make the finished article visually more similar to a natural product.
[0014] However, systems available so far for manufacturing ceramic articles from ceramic powders of different types have several drawbacks. These include the following. The distribution of the powders takes place randomly and, therefore, in an inherently non-reproducible manner. The veins, which are made in the thickness of the articles (and, therefore, can be visible looking at the edge of the articles themselves) are very rarely arranged in a coordinated position with respect to the surface decorations obtained by printing. The appearance of the product is significantly affected, making the lack of similarity relative to a natural product (for example, marble) much more evident.
[0015] Summary
[0016] The object of the invention is to provide a method and a corresponding plant for producing ceramic articles having veins that imitate veins of natural stones, said method not suffering from the aforementioned drawbacks and, at the same time, being able to be carried out in a straightforward and low-cost manner.
[0017] According to the invention, there are provided a method and a plant for making ceramic articles, in particular ceramic slabs or tiles, having veins imitating veins of natural stones and extending through the thickness of the ceramic articles, as defined in the appended independent claims.
[0018] Further advantageous features of the invention are disclosed in the dependent claims.
[0019] The appended claims describe preferred embodiments of the invention to be considered as an integral part of the description.
[0020] Brief Description of the Drawings
[0021] The invention will now be described with reference to the accompanying drawings showing a non-limiting embodiment thereof, wherein:
[0022] - figure 1 is a schematic side view of a plant for making ceramic articles according to the invention;
[0023] - figure 2 is a first horizontal section view of a part of a ceramic powder feeding unit of the plant of figure 1;
[0024] - figure 3 is a schematic side view of a different embodiment of a part of theplant for making ceramic articles according to the invention;
[0025] - figure 4 shows a detail of the embodiment of figure 3;
[0026] - figure 5 shows a flowchart describing part of the method for producing ceramic articles according to the invention;
[0027] - figures 6 to 8 show respective sub -flowcharts of some steps of the flowchart of figure 5;
[0028] - figure 9 shows an example of an input digital image relating to a natural stone having veins and processed by the method of the invention;
[0029] - figures 10 to 15 show an intermediate result of the method of the invention at the end of some steps of the flowchart of figure 5;
[0030] - figure 16 shows a flowchart describing part of the method for producing ceramic articles according to a further embodiment of the invention; and
[0031] - figures 17 and 18 show respective sub -flowcharts of some steps of the flowchart of figure 16.
[0032] Description of Embodiments
[0033] In figure 1, number 1 generically indicates, as a whole, a plant for making ceramic articles, in particular ceramic slabs or tiles, having veins that imitate veins of natural stones and extend through the thickness of the ceramic articles.
[0034] The system 1 comprises a conveyor unit 2 for conveying ceramic powders from an input station 3, in which the ceramic powders are loaded, to an output station 4, from which the ceramic articles exit, according to an advancement direction 5 going from the input station 3 to the output station 4. In the example shown, the conveyor unit 2 comprises at least a first conveyor belt 6, which extends from the input station 3.
[0035] The plant 1 comprises a feeding unit 7 arranged in the area of the input station 3 for feeding ceramic powders of respective different types onto an area 8 of the conveyor unit 2 while the conveyor unit 2 conveys, according to the advancement direction 5, the ceramic powders received so as to form a layer of ceramic powders 9. In the example shown in figure 1, the feeding assembly 1 is shown along a vertical section plane substantially perpendicular to the conveyor belt 6 and the area 8 is located on a portion of the conveyor belt 6.
[0036] With reference also to figure 2, which shows the feeding unit 7 along a horizontal section plane, therefore substantially parallel to the conveyor belt 6, the feeding unit 7 comprises a distribution device 10 for distributing the ceramic powders in the area 8 of the conveyor unit 2 in a variable manner along a transverse direction 11(figure 2), which is transverse to the advancement direction 5, while the conveyor unit 2 conveys the ceramic powders received in the advancement direction 5 so as to define veins in the layer of ceramic powders 9.
[0037] The feeding unit 7 comprises a plurality of containing chambers 12, and in particular a number N1 of containers, and in particular a plurality of containing chambers 12, each designed to contain a respective ceramic powder of a certain type. For example, the ceramic powders contained in the containing chambers 12 have respective different colours. In the example shown, N1 is equal to 4.
[0038] The distribution device 10 comprises a plurality of discharge openings 13 arranged in succession along the transverse direction 11 (figure 2) and a corresponding plurality of actuators 14, each designed to establish a communication between a respective discharge opening 13 and any one of the containing chambers 12. The feeding unit 7 is substantially known and will not be described in further detail.
[0039] The plant 1 further comprises a compacting unit 15 for compacting the layer of ceramic powders 9 while it is conveyed by the conveyor unit 2 so as to obtain a compacted layer of ceramic powders 16.
[0040] In particular, the compaction unit 15 comprises a pressure belt 17, which converges towards the conveyor belt 6 in the advancement direction 5. In this way, the pressure belt 17 cooperates with the conveyor belt 6 to exert a pressure from the top to the bottom, which gradually increases in the advancement direction 5, upon the layer of ceramic powders 9 so as to compact it. The conveyor belt 6 and the pressure belt 17 are mainly made of metal, and in particular steel, so that they cannot be substantially deformed while pressure is exerted upon the layer of ceramic powders 9.
[0041] The plant 1 further comprises a cutting unit 18 for transversely cutting the compacted layer of ceramic powders 16 while it is conveyed by the conveyor unit 2, so as to obtain ceramic articles, for example slabs 19, each having a portion of the compacted layer of ceramic powders 16.
[0042] In particular, the conveyor unit 2 comprises a further conveyor belt 20, which is located in the area of the cutting unit 18, which comprises at least one cutting blade 21 arranged above the conveyor belt 20 and designed to come into contact with the compacted layer of ceramic powders 16 so as to cut it parallel to the transverse direction 11 and obtain the slabs 19.
[0043] According to a further embodiment, the cutting unit 18 comprises at least two further cutting blades 22, which are arranged on opposite sides of the conveyor belt 20and are designed to cut the compacted layer of ceramic powders so as to define side edges of the slabs 19 (substantially parallel to the advancement direction 5). According to a further embodiment, the cutting blades 22 are at least three so as to divide the slab 19 into two or more longitudinal portions. The plant 1 further comprises a firing kiln 23 located in the area of the output station 4 for sintering the ceramic powders of the slabs 19 so as to obtain the ceramic articles. The conveyor unit 2 comprises a series of conveyor rollers 24 for conveying the slabs 19 from the cutting unit 18 to the firing kiln 23.
[0044] Advantageously, though not necessarily, the plant 1 comprises a printing device 25, which is designed to create a graphic decoration over the slabs 19 conveyed by the conveyor unit 2. In particular, the printing device 25 is located between the cutting unit 18 and the firing kiln 23 so that the graphic decoration is created on the slabs 19 conveyed by the conveyor rollers 24.
[0045] According to a further embodiment, the plant 1 comprises a dryer 26 located downstream of the cutting unit 18, and in particular between the cutting unit 18 and the printing device 25, for drying the slabs 19 before the graphic decoration is created.
[0046] The plant 1 comprises a digital human-machine interface 27 for acquiring data and commands from an operator and an electronic control unit 28, which is designed to dialogue with the human-machine interface 27 and to control the feeding unit 7, and in particular the actuators 14, the compacting unit 15, the cutting unit 18 and, if present, the printing device 25 so as to obtain the ceramic articles according to a continuous process. In other words, the plant 1 has a continuous operation.
[0047] According to a further embodiment which is not shown herein, the plant 1 is devoid of the cutting unit 18 and comprises a different compacting unit configured to confine the layer of ceramic powders 9 longitudinally and transversely while it is compacted so as to obtain portions of compacted ceramic powder which will generate respective ceramic articles. Thus, the electronic control unit 28 suitably controls the feeding unit 7 and the different compacting unit so as to obtain the ceramic articles.
[0048] According to a further embodiment shown in figures 3 and 4, which illustrates part of the plant 1 and in which the corresponding elements are indicated with the same numbers and abbreviations as in figure 1, the plant 1 comprises, instead of the feeding unit 7, a different feeding unit 29 comprising, in turn, a plurality of distribution devices 30, each designed to distribute a respective ceramic powder of a certain type in the area 8 of the conveyor unit 2 in a variable manner along a transverse direction 11, while theconveyor unit 2 conveys the ceramic powders received in the advancement direction 5. In particular, each distribution device 30 is designed to distribute in the zone 8, parallel to the transverse direction 11, a strip of ceramic powder 31 of the respective type of ceramic powder.
[0049] With particular reference to figure 4, each distribution device 30 comprises a respective container, and in particular a containing hopper 32, for containing the respective ceramic powder of a certain type. The containing hopper 32 comprises an output mouth 33, whose longitudinal dimension extends parallel to the transverse direction 11.
[0050] Each distribution device 30 further comprises a respective plurality of discharge elements 34 arranged in succession along the respective output mouth 33 and parallel to one another so as to define a build-up plane 35, which is inclined at an angle a with respect to a horizontal plane and is designed to receive the ceramic powder exiting from the output mouth 33. Each discharge element 34 defines, together with a wall 36 of the output mouth 33, a respective discharge opening 37. The discharge openings 37 are thus arranged in succession along the transverse direction 11.
[0051] Each distribution device 30 further comprises a respective plurality of actuators 38, each designed to operate a respective discharge element 34 so as to cause the ceramic powder accumulated on the build-up plane 35 to fall onto the area 8 of the conveyor unit 2, through the respective discharge opening 37.
[0052] According to a particular embodiment, each actuator 38 is designed to move the respective discharge element 34 so as to increase its angle with respect to the horizontal plane relative to the aforementioned angle a to the point of overcoming the static friction between the ceramic powder and the material of the discharge element 34 so as to cause the fall of the ceramic powder accumulated on the build-up plane 35.
[0053] According to a further particular embodiment, each actuator 38 comprises at least one vibrating element, which can be caused to vibrate in order to transfer a vibration to the respective discharge element 34 so as to cause the fall of the ceramic powder accumulated on the build-up plane 35. For example, each discharge element 34 comprises a blade inclined at the aforementioned angle a with respect to the horizontal plane and the vibrating element is designed to transmit a vibration to the blade.
[0054] The electronic control unit 28 is designed to control the feeding unit 29, and in particular the actuators 38.
[0055] The control unit 28 has a digital memory of its own and the human-machineinterface 27 substantially is of a known type and allows the operator to enter data into the control unit 28, and in particular to load an input digital image, based on which binary control masks for the distribution device 10 or the distribution devices 30 are determined, as better explained below.
[0056] In particular, the control unit 28 is configured to implement the method for producing ceramic articles according to the invention, which is described below.
[0057] The method according to the invention entails the following general steps:
[0058] - a feeding phase, during which at least two ceramic powders of respective different types are fed onto the area 8 of the conveyor unit 2 while the conveyor unit 2 conveys, according to the advancement direction 5, the ceramic powders received so as to form the layer of ceramic powders 9;
[0059] - a compacting phase, during which the layer of ceramic powders 9 is compacted while it is conveyed by the conveyor unit 2 so as to obtain the compacted layer of ceramic powders 16; and
[0060] - a cutting phase, during which the compacted layer of ceramic powders 16 is cut along the transverse direction 11 while it is conveyed by the conveyor unit 2 to obtain the ceramic articles.
[0061] In an embodiment described above, in the plant 1 lacks the cutting unit 18 and comprises a different compacting unit, the aforementioned cutting phase is absent and the compacting phase produces portions of compacted ceramic powder.
[0062] The feeding phase comprises a ceramic powder distribution phase, during which each of the ceramic powders is distributed in the area 8 of the conveyor unit 2 in a variable manner along the transverse direction 11, while the conveyor unit 2 conveys the ceramic powders received in the advancement direction 5 so as to define the veins in the layer of ceramic powders 9 and, therefore, in the compacted layer of ceramic powders 16.
[0063] The ceramic powders can be selected from among a number N1 of available ceramic powders, contained in the respective containing chambers 12.
[0064] The method according to the invention comprises the following further general steps:
[0065] - a data acquisition phase, during which a high-resolution input digital image relating to a natural stone having veins and a second number (M) between 2 and N1 are acquired;
[0066] - an image processing phase, during which the input digital image is processedin order to obtain a number M of binary control masks, each of which represents a respective type of veins, for example veins with a certain colour, and is associated with a respective ceramic powder, for example a ceramic powder having a colour similar to that of the veins.
[0067] Furthermore, the ceramic powder distribution phase entails controlling the distribution of each ceramic powder along the transverse direction 11 according to the respective binary control mask.
[0068] In particular, in the embodiment of the plant 1 shown in figures 1 and 2, the actuators 14 of the distribution device 10 are selectively controlled according to the binary control masks to establish a communication between each of the discharge openings 13 and one of the containing chambers 12 in order to vary the distribution of each ceramic powder along the transverse direction 11.
[0069] In the embodiment of the plant 1 shown in figures 3 and 4, the actuators 38 of each distribution device 30 are selectively controlled according to the binary control masks to selectively operate the discharge elements 34 and, hence, cause the ceramic powder to fall through the respective discharge openings 34 in order to vary the distribution of each ceramic powder along the transverse direction 11.
[0070] The flowchart of figure 5 shows the image processing phase in greater detail. Hence, with reference to figure 5, the image processing phase is preceded by the data acquisition phase (block 100 of figure 5), in which the input digital image and the number M are acquired, the latter defining the number of control masks to be obtained.
[0071] The input digital image is acquired in a certain colour space, i.e. a starting colour space, which typically is the RGB colour space. The image processing phase comprises a conversion of the input digital image from the starting colour space to the Lab or CIELAB or CIE 1976 colour space (L*, a*, b*) (block 101). The digital image converted into the LAB space allows for more precise evaluations in terms of colorimetric distance in subsequent processing. In other words, in the Lab colour space, the colour of each pixel has a unique identification.
[0072] Preferably, though not necessarily, the converted digital image is blurred (block 102).
[0073] At this point, a plurality, equal to the number M, of reference colour tones are selected from among the colours of the plurality of available ceramic powders (block 103). According to an embodiment, the selection of the reference colour tones is carried out manually by the operator by using the human-machine interface 27, and inparticular, for each binary control mask to be obtained, by selecting one or more pixels on the input digital image and by acquiring the colour thereof. According to an alternative embodiment, the selection of the reference colour tones is performed automatically by means of a clustering algorithm of the input digital image.
[0074] For each reference colour tone (namely, for each control binary mask to be obtained), a respective raw binary mask is generated, which has pixels of a first value only in correspondence with those certain pixels of the converted digital image that are within a first threshold DCTH1 of colorimetric distance from the reference colour tone (block 104).
[0075] Said first value of the pixels of the raw binary masks expresses an assignment of the pixels of the converted digital image to a certain reference colour tone and therefore to the respective raw binary mask. For example, if the first value corresponds to the black colour, each raw binary mask can be displayed on a screen with black pixels in correspondence with pixels of the converted digital image that are assigned to the raw binary mask (and therefore to the respective reference colour tone) and with colourless or white pixels in correspondence with pixels of the converted digital image that are not assigned to the raw binary mask. In other words, the first value of the pixels of the raw binary masks expresses an active pixel from the point of view of the control of the actuators 14.
[0076] Each binary control mask is obtained from a respective raw binary mask by performing, on the raw binary masks, a plurality of transformations such that each pixel of the converted digital image is assigned to only one of the binary control masks.
[0077] Said plurality of transformations comprise one or more morphological transformations selected from a group consisting of:
[0078] - erosion (block 105);
[0079] - closing (block 106); and
[0080] - removal of those groups of pixels of said first value that are adjacent to each other and cover an area smaller than a predefined threshold value NATH, in terms of number of pixels (block 107).
[0081] The morphological transformations are aimed at eliminating isolated pixels or areas of pixels below a certain quantity in order to help the distribution device 10 work within requirements of its in terms of resolution.
[0082] According to the embodiment shown in figure 3, the transformations performed on the raw binary masks include all the aforementioned morphological transformations(blocks 105-107). Furthermore, the aforementioned morphological transformations, when present, are performed on each raw binary mask, preferably immediately after the generation of said raw binary mask. For this reason, the morphological transformations are shown, in a schematic manner, as part of a cycle that is repeated for each binary mask to be generated, wherein the index i varies from 1 to M and indicates the i-th binary mask. For the sake of simplicity, figure 5 also shows the selection of the reference colour tones as a selection of a reference colour tone before generating the respective raw binary mask. The selection of all reference colour tones before the cycle is substantially equivalent to the selection of one reference colour tone at a time within the cycle.
[0083] It should be pointed out that the blurring of the converted digital image (block 102) allows the result of the morphological transformations of erosion and closing (blocks 105 and 106) to be improved, since, as it is known, the erosion and closing transformations are differential operators and work better on a blurred image.
[0084] According to a further embodiment which is not shown herein, none of said morphological transformations are carried out after the generation of the raw binary masks.
[0085] The generation of the raw binary masks and / or the execution of any morphological transformations could determine the assignment of pixels to several raw binary masks and this is not desired. For this reason, the transformations performed on the raw binary masks include a first operation for managing the assignment of the pixels of the converted digital image, hereinafter simply called first pixel assignment management (block 108), which is carried out immediately after the generation of the raw binary masks or, if performed, after the morphological transformations.
[0086] During the first pixel assignment management, for each pixel of the converted digital image that is assigned to at least two raw binary masks, at least one of said at least two raw binary masks is modified to maintain the assignment of that pixel to only one of said at least two raw binary masks, according to colorimetric distances calculated between that pixel and the reference colour tones of said at least two raw binary masks. In other words, those raw binary masks in which the assignment of that pixel must be eliminated are modified.
[0087] With reference to figure 6, wherein NPM is the number of pixels of the converted digital image that are assigned to several raw binary masks, j is an index that varies from 1 to NPM and Pj is precisely the j-th pixel of the converted digital image withmultiple assignment, the first pixel assignment management comprises a cycle of steps to be performed for each pixel of the converted digital image that is assigned to at least two raw binary masks, and in particular:
[0088] - calculating the colorimetric distances DCj between that pixel Pj and the reference colour tones relating to said at least two raw binary masks (block 201);
[0089] - calculating the minimum value Dmin between the said colorimetric distances (block 202);
[0090] - comparing the minimum value Dmin with a second threshold DCTH2 of colorimetric distance (block 203);
[0091] - if the minimum value Dmin is less than or equal to the second threshold DCTH2, then modifying at least one of said at least two raw binary masks to maintain the assignment of the pixel to the raw binary mask to which the minimum value corresponds (block 204); namely, in other words, those raw binary masks relating to those reference colour tones to which the minimum value does not correspond are modified, eliminating the assignment of the pixel Pj;
[0092] - otherwise, identifying the raw binary mask that has the highest number of pixels of the first value within a certain radius from said pixel and modifying at least one of said at least two raw binary masks to maintain the assignment of said pixel to the identified raw binary mask (block 205); namely, in other words, the other raw binary masks are modified to eliminate the assignment of the pixel Pj .
[0093] By way of example, figure 9 shows an input image relating to a natural stone having veins, figures 10 to 12 show three raw binary masks obtained from the input digital image based on three respective reference colour tones, at the end of the morphological transformations, and figures 13 to 15 show the three raw binary masks after the first pixel assignment management. The input digital image and the consequent raw binary masks have a longitudinal development that is in accordance with the advancement direction 5, therefore they will have a transverse development in accordance with the transverse direction 11.
[0094] In the example shown in figures 9 to 15, the first value of the pixels assigned to the raw binary masks corresponds to the colour black. This example clearly shows, even at a first glance, the effect of the first pixel assignment management, i.e. making sure that each pixel of the converted digital image, and therefore of the input digital image, is assigned to only one of the raw binary masks, and therefore to only one of the binary control masks.The transformations described above could not be sufficient to allow the smaller areas of pixels of the binary masks to be reproduced by the distribution device 10, namely the smaller areas of pixels of the masks could not meet the resolution requirements of the distribution device 10.
[0095] For this reason, the transformations performed on the raw binary masks also include operations to adapt the original resolution of the raw binary masks, longitudinally to a first resolution of the ceramic powder distribution phase in the advancement direction 5, and transversely to a second resolution of the ceramic powder distribution phase in the transverse direction 11. For the sake of simplicity, with reference again to figure 5, the two resolution adaptation operations will hereinafter be referred to as longitudinal resolution adaptation (block 110) and transverse resolution adaptation (block 114), respectively.
[0096] By construction, the resolution of each of the raw binary masks is identical to the resolution of the converted digital image. The resolution adaptation entails that the resolution of the converted digital image, and therefore of the raw binary masks, is greater than said first resolution of the ceramic powder distribution phase in the advancement direction 11 and smaller than said second resolution of the ceramic powder distribution phase in the transverse direction.
[0097] Preferably, though not necessarily, the longitudinal resolution adaptation and the transverse resolution adaptation are performed after the first pixel assignment management, as in the embodiment shown in figure 5.
[0098] Preferably, though not necessarily, the longitudinal resolution adaptation precedes the transverse resolution adaptation, as in the embodiment shown in figure 5.
[0099] More in detail, assuming that the rows of pixels of the raw binary masks in the longitudinal direction, i.e. according to the advancement direction 5, are called “columns” and the rows of pixels of the raw binary masks in the transverse direction, i.e. according to the transverse direction 11, are called “rows”, the longitudinal resolution adaptation (block 110) includes, for each raw binary mask:
[0100] - identifying, by scanning the raw binary mask one column at a time, groups of pixels of the first value that are adjacent to one another and have, in terms of number of pixels, a length that is smaller than a minimum length NP1 defining said first resolution; and
[0101] - deleting or bringing the identified groups of pixels to the minimum length NP1, according to a threshold length NPTH1 that is smaller than the minimum length NP1and based on a certain criterion.
[0102] The transverse resolution adaptation is similar to the longitudinal resolution adaptation, namely it takes place by scanning each raw binary mask one row at a time and adopting corresponding pixel length parameters, namely a minimum length NP2 defining said second resolution and a threshold length NPTH2 that is smaller than the minimum length NP2.
[0103] With regard to the adaptation to the resolution requirements of the distribution device 10, the following is noted. The minimum length NP1 depends, among other things, on the speed of change of state of the actuators 14 of the distribution device 10, namely on the speed with which each actuator 14 switches between any two containing chambers 12 for communication with the respective discharge opening 13. The minimum length NP2, on the other hand, depends on the width, measured along the transverse direction 11, of each discharge opening 13.
[0104] Preferably, though not necessarily, according to some embodiments, the morphological transformation of removing those groups of pixels of the first value that are adjacent to each other and cover an area that is smaller than the threshold value NATH is performed on each raw binary mask after the first pixel assignment management, before and / or after the longitudinal resolution adaptation. Figure 5 shows the embodiment in which said morphological transformation is performed before (block 109) and after (block 111) the longitudinal resolution adaptation.
[0105] The resolution adaptation, both longitudinal and transverse, could determine the assignment of pixels to several raw binary masks and this is not desired.
[0106] For this reason, the transformations performed on the raw binary masks include a second operation for managing the assignment of the pixels of the converted digital image, hereinafter simply called second pixel assignment management, which is carried out after the longitudinal resolution adaptation and / or after the transverse resolution adaptation. Figure 5 shows the embodiment in which the second pixel assignment management is performed after the longitudinal resolution adaptation (block 112) and after the transverse resolution adaptation (block 115).
[0107] During the second pixel assignment management, for each pixel of the converted digital image that is assigned to at least two raw binary masks, at least one of said two raw binary masks is modified to maintain the assignment of that pixel to only one of the raw binary masks, according to the number of pixel of the first value of the two raw binary masks that are adjacent in a row to said pixel of the converted digital image. Inparticular, those raw binary masks that do not have the highest number of pixels of the first value adjacent in a row to said pixel of the converted digital image are modified to eliminate the assignment of said pixel.
[0108] When the second pixel assignment management is performed after the longitudinal resolution adaptation, a row of pixels is considered longitudinally, namely in the direction of the columns of the raw binary masks. On the other hand, when the second pixel assignment management is performed after the transverse resolution adaptation, a row of pixels is considered transversally, namely in the direction of the rows of the raw binary matrices.
[0109] With reference to figure 7, wherein NPM is the number of pixels of the converted digital image that are assigned to several raw binary masks, j is an index that varies from 1 to NPM, Pj is precisely the j-th pixel with multiple assignment, KM is the number of raw binary masks to which the pixel Pj is assigned and i is an index that varies from 1 to KM and indicates the i-th binary mask, the second pixel assignment management comprises an external cycle to be performed for each pixel of the converted digital image that is assigned to at least two raw binary masks and an internal cycle of steps to be performed for each of the raw binary masks to which said pixel is assigned. The internal cycle, performed on the i-th raw binary mask, includes:
[0110] - calculating the number NPi of pixels of the first value that are adjacent to one pixel of the first value corresponding to the pixel Pj in a row (namely, along a column or along a row) (block 301);
[0111] - comparing the number NPi with a minimum number Nmin of pixels (block 302);
[0112] - if the number NPi is greater than or equal to the minimum number Nmin, then modifying at least one of the KM raw binary masks to maintain the assignment of the pixel Pj on the i-th raw binary mask (block 303), namely, in other words, all the other raw binary masks are modified to eliminate the assignment of the pixel Pj;
[0113] - otherwise, identifying the raw binary mask that has the highest number of pixels of the first value within a certain radius from said pixel and modifying at least one of the KM raw binary masks to maintain the assignment of the pixel Pj to the identified raw binary mask (block 304); namely, in other words, all the other raw binary masks are modified to eliminate the assignment of the pixel Pj.
[0114] The raw binary masks are ordered according to an appropriate criterion and the internal cycle is performed following this order. Therefore, the actual assignment of thepixel Pj to one single raw binary mask will take place after the last raw binary mask involving the pixel Pj is processed.
[0115] The resolution adaptation, both longitudinal and transverse, could also leave some pixels of the converted digital image without assignment to any raw binary mask.
[0116] For this reason, the transformations performed on the raw binary masks include a third operation for managing the assignment of the pixels of the converted digital image, hereinafter simply called third pixel assignment management, which is carried out after the longitudinal resolution adaptation and / or after the transverse resolution adaptation. Figure 5 shows the embodiment in which the third pixel assignment management is performed both after the longitudinal resolution adaptation (block 113) and after the transverse resolution adaptation (block 116).
[0117] Preferably, though not necessarily, the third pixel assignment management is performed after the second pixel assignment management, as in the embodiment shown in figure 5.
[0118] During the third pixel assignment management, for each pixel of the converted digital image that is not assigned to any of the raw binary masks, only one of the raw binary masks is modified to assign that pixel to it based on a scan of a quantised mask obtained by overlapping the raw binary masks.
[0119] With reference to figure 8, the third pixel assignment management comprises: - generating, based on an overlapping of the raw binary masks, a quantised mask such that each pixel of the converted digital image that is assigned to a certain raw binary mask corresponds to a pixel of the quantised mask having a second value that is uniquely associated with that certain raw binary mask (block 401);
[0120] - scanning the quantised mask pixel by pixel lengthwise or crosswise (block 402); to this regard, figure 6 schematically illustrates the scanning cycle in which the index j varies from 1 to NPtot and indicates the j -th pixel of the quantised mask and therefore the corresponding pixel Pj of the converted digital image and in which NPtot is the total number of pixels of the quantised mask;
[0121] - identifying any pixels of the quantised mask that do not assume a second value, namely checking whether each pixel corresponds to a pixel Pj of the converted digital image that is not assigned to any raw binary mask (block 403);
[0122] - for each identified pixel of the quantised mask, assigning the corresponding pixel Pj of the converted digital image to the raw binary mask associated with that second value pixel of the quantised mask that precedes, in the scanning direction, theidentified pixel (block 404); and
[0123] - for the other pixels of the quantised mask, i.e. those that do not have a second value (namely, those that correspond to pixels of the converted digital image that are assigned to a raw binary mask), doing nothing (block 405).
[0124] By way of example, if M = 3, namely if there are three raw binary masks, then the pixels of the quantised mask can assume three different second values.
[0125] Preferably, the quantised mask is scanned longitudinally when the third pixel assignment operation is performed after the longitudinal resolution adaptation (block 113) or transversally when the third pixel assignment operation is performed after the transversal resolution adaptation (block 116).
[0126] According to a further embodiment of the invention shown in figure 16, in which corresponding elements are indicated with the same numbers and abbreviations as in figure 5, the method comprises, instead of the processing indicated with 110 to 116 in figure 5, other processing described below.
[0127] After the optional step of removing those groups of pixels of the first value that are adjacent to each other and cover an area smaller than a predefined threshold value NATH (block 109), a step that follows the first pixel assignment management (block 108), the transformations performed on the raw binary masks comprise a further operation for managing the assignment of the pixels of the converted digital image, hereinafter simply called further pixel assignment management (block 117).
[0128] During said further pixel assignment management, for each pixel of the converted digital image that is not assigned to any of the raw binary masks, only one of the raw binary masks is modified to assign that pixel to it based on a scan of a quantised mask obtained by overlapping the raw binary masks.
[0129] In particular, with reference to figure 17, the further pixel assignment management comprises:
[0130] - generating, based on a superimposing of the raw binary masks, a quantised mask such that each pixel of the converted digital image that is assigned to a certain raw binary mask corresponds to a pixel of the quantised mask having a second value that is uniquely associated with that certain raw binary mask (block 501);
[0131] - scanning the quantised mask pixel by pixel lengthwise or crosswise (block 502); to this regard, figure 17 schematically illustrates the scanning cycle in which the index j varies from 1 to NPtot and indicates the j -th pixel of the quantised mask and therefore the corresponding pixel Pj of the converted digital image and in which NPtot is the totalnumber of pixels of the quantised mask;
[0132] - identifying any pixels of the quantised mask that do not assume a second value, namely checking whether each pixel corresponds to a pixel Pj of the converted digital image that is not assigned to any raw binary mask (block 503);
[0133] - for each identified pixel of the quantised mask, calculating the colorimetric distances DCj between the corresponding pixel (Pj) of the converted digital image and the reference colour tones relating to all the raw binary masks (block 504);
[0134] - calculating the minimum value Dmin between the said colorimetric distances (block 505);
[0135] - assigning the corresponding pixel Pj of the converted digital image to the raw binary mask associated with the reference colour tone that determines the minimum value Dmin (block 506); and
[0136] - for the other pixels of the quantised mask, i.e. those that do not have a second value (namely, those that correspond to pixels of the converted digital image that are already assigned to a raw binary mask), doing nothing (block 507).
[0137] With reference again to figure 16, the further pixel assignment operation is followed by an operation for adapting the original resolution of the raw binary masks, hereinafter simply called resolution adaptation (block 118), which therefore replaces the longitudinal resolution adaptation (block 110) the transverse resolution adaptation (block 114) of the embodiment of figure 5.
[0138] With reference to figure 18, the resolution adaptation operations comprises:
[0139] - dividing the area of the converted digital image into a plurality of image portions with the same shape, not overlapping one another and with a size such that the density of image portions per surface unit is defined by said first resolution and second resolution (block 601), wherein NB is the total number of portions and k is an index that varies from 1 to NB and indicates the k-th image portion Bk; and
[0140] - for each image portion (k ranging from 1 to NB), performing the following steps:
[0141] - for each raw binary mask (i ranging from 1 to M), calculating the number NPBki of pixels Pj of the converted digital image that are within the image portion Bk and that are assigned to that raw binary mask (block 602);
[0142] - calculating the maximum value (NPBmax) among the numbers NPBki of pixels that are assigned to the different raw masks (block 603); and
[0143] - assigning the pixels Pj of the converted digital image that are within the imageportion Bk to the raw binary mask that determines the maximum value NPBmax (block 604). The image portions have a same shape that is such as to cover the area of the converted digital image without leaving intermediate empty spaces. According to various embodiments, the image portions are in the form of a parallelogram and arranged in an array, or in the form of a triangle, or in the form of a regular hexagon.
[0144] The transformations of the image processing phase described above in the various embodiments produce subsequent corrections on the raw binary masks so as to define corresponding binary control masks that are increasingly more precise with respect to the technical requirements of the distribution device 10 or the distribution devices 30. In fact, the raw binary masks at the end of the first pixel assignment management (block 108 of figure 5) are already binary control masks that can be used by the distribution device 10 (figures 1 and 2) or by the distribution devices 30 (figures 3 and 4), since each pixel of the converted digital image is assigned to only one of the binary control masks.
[0145] According to the embodiment of figures 1 and 2, a group of pixels of the first value in a given transverse position of a binary control mask, said transverse position corresponding to the position of a discharge opening 13 of the distribution device 10, indicates that the actuator 14 associated with the discharge opening 13 must switch over the containing chamber 12 of the ceramic powder associated, by means of the respective reference colour tone, with the binary control mask.
[0146] According to the embodiment of figures 3 and 4, a group of pixels of the first value in a given transverse position of a binary control mask that is associated with a respective ceramic powder, said transverse position corresponding to the position of a discharge opening 37 of the distribution device 30 relating to that ceramic powder, indicates that the actuator 38 associated with the discharge opening 37 must operate the respective discharge element 34 to cause the fall of the ceramic powder.
[0147] The method described above and the plant 1, operating according to said method, permit the production of ceramic articles that reproduce the veins of natural stones in a realistic way also throughout the entire thickness of the ceramic articles, thanks to the creation of the binary control masks that are associated with respective ceramic powders of different types and that enable a precise control of the distribution device 10 of the feeding unit 7, or of the distribution devices 30 of the feeding unit 29. In addition, the method makes it possible to produce ceramic articles reproducing any type of veins or decoration, for example also wood veins, which extends throughout the entire thicknessof the articles. To this aim, the acquired digital input image will relate to the veins to be reproduced.
[0148] According to a further embodiment, the control unit 28 is configured to control the distribution device 10, or the distribution devices 30, and the printing device 25 so as to synchronize the printing of the graphic decoration with the creation of the layer of ceramic powders 9. In this way, it is possible to create a graphic decoration on the surface of the ceramic articles that is coordinated with the veins in the thickness of the ceramic articles.
Claims
C L A I M S1. Method for producing ceramic articles, in particular ceramic slabs or tiles, presenting veins or decorations extending into the thickness of the ceramic articles, the method comprising:- a feeding phase, during which at least two ceramic powders of different respective types are fed onto an area (8) of a conveyor unit (2) while the conveyor unit (2) advances, according to an advancement direction (5), the ceramic powders received so as to form a layer of ceramic powders (9); and- a compacting phase, during which the layer of ceramic powders (9) is compacted while it is advanced by the conveyor unit (2) so as to obtain a compact layer of ceramic powders (16);the feeding phase comprising:- a ceramic powder distribution phase, during which each of the ceramic powders is distributed in said zone (8) of the conveyor unit (2) in a variable manner along a transverse direction (11), which is transverse to the advancement direction (5), while the conveyor unit (2) advances the ceramic powders received in the advancement direction (5) so as to define said veins in the layer of ceramic powders (9);the method being characterised in that the ceramic powders fed can be selected from a plurality, equal to a first number (Nl), of available ceramic powders, and in that it comprises:- a data acquisition phase (100), during which an input digital image relating to veining and a second number (M) between 2 and the first number (Nl) are acquired;- an image processing phase (101-116; 101-109, 117, 118), during which the input digital image is processed in order to obtain a plurality, equal to the second number (M), of binary control masks, each of which represents a respective type of veining or decoration and is associated with a respective ceramic powder; and- during the ceramic powder distribution phase, controlling the distribution of each ceramic powder along said transverse direction (11) according to the respective binary control mask.
2. The method according to claim 1, wherein said different types of ceramic powders differ from each other by their respective colours.
3. The method according to claim 1 or 2, wherein during the feeding phase, the ceramic powders are fed onto said zone (8) of the conveyor unit (2) by means of a feeding unit (7; 29) comprising a quantity, equal to the first number (Nl), of containers(12; 32), each of which is able to contain a respective ceramic powder, a plurality of discharge openings (13; 37) arranged in succession along said transverse direction (11), and a plurality of actuators (14; 38), each of which is able to cause the transfer of a ceramic powder from the respective container (12; 32) to a discharge opening (13; 37); the method comprising:- during the ceramic powder distribution phase, selectively controlling the actuators (14; 38) according to the binary control masks in order to vary the distribution of each ceramic powder along said transverse direction (11).
4. The method according to any one of claims 1 to 3, wherein the input digital image is in a starting colour space and the image processing phase (101-116; 101-109, 117, 118) comprises:- converting (101) the input digital image into the Lab colour space to obtain a converted digital image;- selecting (103) a plurality, equal to the second number (M), of reference colour tones from among the colours of the plurality of available ceramic powders;- for each reference colour tone, generating (104) a respective raw binary mask, which has pixels of a first value only in correspondence with those certain pixels of the converted digital image that are within a first threshold (DCTH1) of colorimetric distance from the reference colour tone, said first value expressing an assignment of said certain pixels of the converted digital image to the reference colour tone and therefore to the respective raw binary mask;- obtaining each binary control mask from a respective raw binary mask by performing, on the raw binary masks, a plurality of transformations (105-116; 105-109, 117, 118) such that each pixel of the converted digital image is assigned to only one of the binary control masks.
5. The method according to claim 4, wherein the selection (103) of the reference colour tones comprises:- for each binary control mask to be obtained, selecting at least one respective pixel on the input digital image and acquiring its colour.
6. The method according to claim 4 or 5, wherein the image processing stage (101-116; 101-109, 117, 118) comprises:- blurring (102) the converted digital image;the generation (104) of the raw binary masks being done starting from the converted digital image after blurring.
7. The method according to any one of claims 4 to 6, wherein said plurality of transformations (105-116; 105-109, 117, 118) comprises one or more morphological transformations selected from a consistent group of: erosion (105); closing (106); and removing (107) those groups of pixels of said first value that are adjacent to each other and cover an area smaller than a predefined threshold value (NATH); in particular, the morphological transformations being performed on each raw binary mask, preferably immediately after the generation (104) of said raw binary mask.
8. The method according to any one of claims 4 to 7, wherein said plurality of transformations (105-116; 101-109, 117, 118) comprises a first pixel allocation management (108), during which, for each pixel (Pj) of the converted digital image that is assigned to at least two raw binary masks, at least one of the said at least two raw binary masks is modified to maintain the assignment of this pixel (Pj) to only one of the raw binary masks according to colorimetric distances (DCj) calculated between this pixel (Pj) and the reference colour tones of the said at least two raw binary masks.
9. The method according to any one of claims 4 to 7, wherein said plurality of transformations (105-116; 105-109, 117, 118) comprises a first pixel assignment management (108), which comprises, for each pixel (Pj) of the converted digital image that is assigned to at least two raw binary masks:- calculating (201) the colorimetric distances (DCj) between this pixel (Pj) and the reference colour tones relating to the said at least two raw binary masks;- calculating (202) the minimum value (Dmin) between the said colorimetric distances (DCj);- if the minimum value (Dmin) is less than or equal to a second threshold (DCTH2) of colorimetric distance (203), then modify at least one of the said at least two raw binary masks to maintain (204) the assignment of that pixel (Pj) to the raw binary mask to which said minimum value (Dmin) corresponds;- otherwise, identifying the raw binary mask that has the highest number of pixels of said first value within a certain radius from said pixel (Pj) and modifying at least one of said at least two raw binary masks to maintain (205) the assignment of said pixel (Pj) to the identified raw binary mask.
10. The method according to any one of claims 4 to 9, wherein said plurality of transformations (105-116; 105-109, 117, 118) comprises a further morphological transformation consisting of the removal (109) of those groups of pixels of said first value that are adjacent to each other, which cover an area smaller than a predefinedthreshold value (NATH), this further morphological transformation being performed on each raw binary mask after said first pixel assignment operation (108).
11. The method according to claim 9 or 10, wherein said plurality of transformations (105-109, 117, 118) comprises a further pixel assignment management (117), during which, for each pixel (Pj) of the converted digital image that is not assigned to any of the raw binary masks, only one of the raw binary masks is modified to assign that pixel (Pj) to it based on a scan of a quantised mask obtained by superimposing the raw binary masks.
12. The method according to claim 9 or 10, wherein said plurality of transformations (105-109, 117, 118) comprises a further pixel assignment management (117), which comprises:- generating (501), on the basis of an overlapping of the raw binary masks, a quantised mask such that each pixel of the converted digital image that is assigned to a certain raw binary mask corresponds to a pixel of the quantised mask having a second value that is uniquely associated with that certain raw binary mask;- scanning (502) the quantised mask pixel by pixel lengthwise or crosswise to identify (503) any pixels that do not assume a second value;- for each identified pixel of the quantised mask, calculate (504) colorimetric distances (DCj) between the corresponding pixel (Pj) of the converted digital image and the reference colour tones relating to all the raw binary masks;- calculating (505) the minimum value (Dmin) among said colorimetric distances; and- assigning (506) the corresponding pixel (Pj) of the converted digital image to the raw binary mask associated with the reference colour tone that determines the minimum value (Dmin).
13. The method according to any one of claims 4 to 9, wherein said converted digital image and said raw binary images have a longitudinal development that is consistent with said advancement direction (5) and said plurality of transformations (105-109, 117, 118) comprises a resolution adaptation, during which the original resolution of each raw binary mask is adapted longitudinally to a first resolution of the ceramic powder distribution phase in the advancement direction (5) and transversely to a second resolution of the ceramic powder distribution phase in said transverse direction (11); preferably the resolution adaptation is performed after said first pixel assignment management (108).
14. The method according to claim 13, wherein the resolution adaptation comprises:- dividing (601) the area of the converted digital image into a plurality of image portions (Bk), of equal shape, non-overlapping and of such a size that the density of portions per surface unit is defined by said first resolution and second resolution; and - for each of said image portions (Bk), performing the following steps:- for each raw binary mask, calculating (602) a number of pixels (NPBki) of the converted digital image that are inside the image portion (Bk) and that are assigned to this raw binary mask;- calculating (603) the maximum value (NPBmax) among the numbers of pixels (NPBki) of the converted digital image that are assigned to the different raw masks; and - assigning (604) the pixels (Pj) of the converted digital image that are within the image portion (Bk) to the raw binary mask that determines the maximum value (NPBmax).
15. The method according to any of claims 4 to 9, wherein said converted digital image and said raw binary images have a longitudinal development that is consistent with said advancement direction (5) and said plurality of transformations (105-116) comprises a longitudinal resolution adaptation (110), during which the original resolution of each raw binary mask is longitudinally adapted to a first resolution of the ceramic powder distribution phase in the advancement direction (5), and a transverse resolution adaptation (114), during which the original resolution of each raw binary mask is adapted transversely to a second resolution of the ceramic powder distribution phase in said transverse direction (11); preferably the longitudinal resolution adaptation (110) and the transverse resolution adaptation (114) are performed after said first pixel assignment management (108).
16. The method according to claim 15, wherein each of said longitudinal resolution adaptation (110) and transverse resolution adaptation (114) comprises, for each raw binary mask:- for each longitudinal or transverse row of pixels, identifying groups of pixels of said first value that are adjacent to each other and that have, in terms of number of pixels, a length less than a first minimum length (NP1) defining said first resolution or, respectively, a second minimum length (NP2) defining said second resolution; and - deleting or bringing the identified groups of pixels to the first minimum length (NP1) or, respectively, to the second minimum length (NP2), according to a firstthreshold length (NPTH1) less than the first minimum length (NP1) or, respectively, a second threshold length (NPTH2) less than the second minimum length (NP2).
17. The method according to claim 15 or 16, wherein said plurality of transformations (105-116) comprises a further morphological transformation consisting in the removal (109, 111) of those groups of pixels of said first value that are adjacent to each other and cover an area smaller than a predefined threshold value (NATH), this further morphological transformation being performed on each raw binary mask before and / or after said longitudinal resolution adaptation (110).
18. The method according to any one of claims 15 to 17, wherein said plurality of transformations (105-116) comprises a second pixel assignment management (112, 115), which is performed after said longitudinal resolution adaptation (110) and / or after said transversal resolution adaptation (114) and during which, for each pixel (Pj) of the converted digital image that is assigned to at least two raw binary masks, at least one of said two raw binary masks is modified to maintain the assignment of said pixel to only one of the raw binary masks according to the number of pixels (NPi) of said first value of said at least two raw binary masks that are adjacent in a row to said pixel (Pj) of the converted digital image; preferably said row being in the longitudinal or transversal direction when the second pixel assignment operation (112, 115) is performed after the longitudinal resolution adaptation (110) or, respectively, after the transversal resolution adaptation (11 ).
19. The method according to any one of claims 15 to 18, wherein said plurality of transformations (105-116) comprises a third pixel assignment management (113, 116), which is performed after said longitudinal resolution adaptation (110) and / or after said transversal resolution adaptation (114) and during which for each pixel (Pj) of the converted digital image that is not assigned to any of the raw binary masks, only one of the raw binary masks is modified to assign that pixel (Pj) to it on the basis of a scan of a quantised mask obtained by superimposing the raw binary masks.
20. The method according to any one of claims 15 to 18, wherein said plurality of transformations (105-116) comprises a third pixel assignment management (113, 116), which is performed after said longitudinal resolution adaptation (110) and after said transverse resolution adaptation (114) and comprises:- generating (401), on the basis of an overlapping of the raw binary masks, a quantised mask such that each pixel of the converted digital image that is assigned to a certain raw binary mask corresponds to a pixel of the quantised mask having a secondvalue that is uniquely associated with that certain raw binary mask;- scanning (402) the quantised mask pixel by pixel lengthwise or crosswise to identify (403) any pixels that do not have second values; and- for each identified pixel of the quantised mask, assigning (404) the corresponding pixel (Pj) of the digital image converted to the raw binary mask associated with that second value pixel of the quantised mask that precedes, in the scanning direction, the identified pixel;preferably, the quantised mask is scanned longitudinally when the third pixel assignment operation (113) is performed after the longitudinal resolution adaptation (110) or transversally when the third pixel assignment operation (116) is performed after the transversal resolution adaptation (114).
21. A plant for making ceramic articles, in particular ceramic slabs or tiles, presenting veins or decorations extending into the thickness of the ceramic articles, the plant (1) comprising: a conveyor unit (2) for advancing ceramic powders; a feeding unit (7) for feeding different types of ceramic powders onto an area (8) of the conveyor unit (2) while the latter advances, in a advancement direction (5), the ceramic powders received so as to form a layer of ceramic powders (9); a compacting unit (15) to compact the layer of ceramic powders (9) as it is fed by the conveyor unit (2) in order to obtain a compact layer of ceramic powders (16); a human -machine interface (27) to acquire data and commands; and a control unit (28) configured to dialogue with the man-machine interface (27) and control the feeding unit (7) and the compacting unit (15); the feeding unit (7) comprising at least one distribution device (10) to distribute the ceramic powders in said zone (8) of the conveyor unit (2) in a variable manner along a transverse direction (11), which is transverse to the advancement direction (5), while the conveyor unit (2) advances the received ceramic powders in the advancement direction (5) so as to define said veins in the layer of ceramic powders (9); the system (1) being characterised in that the control unit (28) is configured to implement the method according to any of claims 1 to 20.