Machine and method for processing and / or moving metal sheets or sheet metals with edge recognition means

Abstract

A machine and method for processing or handling metal sheets or sheet metals, comprising a vision system for imaging at least a portion of the metal sheet or sheet metal, and a processing unit for edge recognition. In some embodiments, this step is performed by a deep learning algorithm. If this recognition step is insufficient, a processor reprocesses the image with a masking means that can be applied to at least a portion of the image to better distinguish edges against the background.

Description

[Technical Field]

[0001] The present invention relates to a machine and method for processing and / or moving sheet metal and / or metal plates, in particular to a machine for processing and / or moving sheet metal and / or metal plates, semi-finished products, scrap, which is provided with an artificial vision system adapted to detect geometric information such as the position and orientation of the pieces to be processed, and which makes it possible to set optimal operating parameters for operating means for processing and / or moving the pieces. The present invention also relates to a method for processing and / or moving sheet metal and / or metal plates based on geometric information about the pieces detected by the artificial vision system. [Background technology]

[0002] Machines for processing and / or moving all or parts of sheet metal and / or metal plates, such as laser cutting machines, punching machines, combined cutting and punching machines, bending machines, transport and manipulation machines (e.g. Cartesian manipulators, transport belts, etc.), are known which are provided with an artificial vision system which is used to determine the position and orientation of the pieces or parts of sheet metal and / or metal plates to be processed and / or moved, so that the control unit of the machine can adapt and configure the operating parameters, i.e. the processing and / or movement program, in order to appropriately control the operating means which perform the processing, e.g. the cutting and / or punching and / or bending means or the means which move or manipulate the pieces.

[0003] The pieces to be processed typically include sheet metal and / or plate metal, and in some cases include scrap or skeletons of sheet metal or plate metal that can be used to make other pieces.

[0004] Known artificial vision systems used for this purpose include a camera capable of taking a photograph or image of one or more pieces to be processed and / or moved.

[0005] The processing and computing unit of the artificial vision system processes the images acquired by the camera using a suitable contour extraction algorithm so as to extract and extrapolate the contours of the pieces in the processed image, i.e. the lines or sets of lines that limit and delimit the pieces.

[0006] In some applications, the extrapolated contour is compared to a previously stored reference figure or drawing of the same piece to obtain geometric information about the alignment of the piece on the work surface relative to the machine's frame of reference.

[0007] The alignment information includes in particular the offset, i.e., distance and rotation, of one or more contour sections, in particular substantially linear, of the piece represented in the processed image relative to a corresponding contour section of the piece represented in the reference graphic. In this way, it is possible to determine the position and orientation of the piece relative to the machine's reference system. The "position" of a piece means, for example, the distance of a reference point (usually an edge or angle) of the piece from the origin of the machine's reference system along two orthogonal axes, and the "orientation" of a piece means the angle formed by the reference section of the piece contour with one of the two axes of the reference system.

[0008] The geometric information thus obtained about the position and orientation of the pieces is sent to the control unit of the machine, which uses the information to adapt and configure the processing parameters, i.e. the processing program, and then suitably controls the operating means or moving and manipulating means that perform the processing, e.g. cutting and / or punching and / or bending. Summary of the Invention [Problem to be solved by the invention]

[0009] A drawback of known machining and / or moving machines and associated machining and / or moving methods is that the contour extraction algorithms are often unable to accurately and completely recognize and identify the contour of the piece, and in particular are usually unable to distinguish all contour sections from the background, i.e., from the work surface, due to low contrast with the background, surface finish, material type, color scheme of the piece, and / or due to lighting conditions and / or position of the piece relative to the camera (viewpoint).

[0010] Furthermore, artificial vision systems are unable to determine the alignment and positioning of scrap or skeleton sheet or plate metal where no reference graphic or drawing is available for comparison.

[0011] Therefore, in both cases, the operator is required to intervene through a control panel and manually select the necessary points on the processed image of the piece that will enable the artificial vision system to identify missing or poorly represented contour sections of the piece.

[0012] However, this manual identification procedure, in addition to requiring the intervention of a skilled operator, can be quite tedious and time-consuming, and in any case can be prone to errors. [Means for solving the problem]

[0013] One object of the present invention is to improve known machines and methods for processing and / or moving sheet metal and / or metal plates.

[0014] Another object is to provide a machine and method that allows for precise, accurate and substantially automatic determination of the position and orientation of a piece placed on the working surface of the machine and then configuring and / or adjusting the operating parameters of the operating means that process and / or move / manipulate the piece.

[0015] A further object is to provide a machine and method that allows for determining at least the position and orientation on the working surface of the machine of pieces, including sheet metal, metal plate, semi-finished products, scrap or skeletons of sheet metal or metal plate, in particular without the need to process and store a reference figure or drawing of the piece.

[0016] Another object is to provide a machine and a method that make it possible to determine geometric information of a piece to be machined, including in particular at least the distances and rotation angles of one or more contour sections of the piece.

[0017] In a first aspect of the present invention, there is provided a machine for processing sheet metal and / or metal plates according to claim 1.

[0018] In a second aspect of the present invention, a method for processing sheet metal and / or metal plates according to claim 8 is provided.

[0019] In a third aspect of the present invention there is provided a machine for processing sheet metal and / or metal plates as claimed in claim 15.

[0020] In a fourth aspect of the present invention, there is provided a method for processing sheet metal and / or metal plates according to claim 22.

[0021] The present invention may be better understood and practiced with reference to the accompanying drawings, which show illustrative and non-limiting embodiments thereof. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a partial view of a machine for processing and / or moving sheet metal and / or metal plates according to the present invention, in relation to the pieces to be processed, in particular semi-finished pieces, placed on the working surface of the machine. [Figure 2] is an image of a piece, specifically a new sheet metal, placed on the work surface of the machine, captured by the camera of the machine's artificial vision system in FIG. 1. [Figure 3] is a processed image obtained by the processing unit of the machine by processing the image of Figure 2 by a deep learning algorithm. [Figure 4] 4 is an enlarged view of the processed image of FIG. 3 showing the contour section of the contour of the piece. [Figure 5] is an image of a further piece to be machined positioned on the work surface of the machine, the image being acquired by a camera of the machine's artificial vision system of FIG. 1. [Figure 6] 1 is a perspective view of a processed image of a piece, in particular a sheet metal, processed by a contour recognition algorithm. [Figure 7] 1 is a perspective view of a processed image of another piece, specifically a sheet metal, processed by a contour recognition algorithm. [Figure 8] FIG. 7 is a plan view of a processed image obtained by linearizing the image of FIG. 6. [Figure 9] 1 shows a mask means including a plurality of elliptical masks applied to a portion of the processed image of FIG. 8 containing the contour section to be recognized and identified, and an enlarged view of one of the masks containing a portion of the contour section. [Figure 10] 8 shows a mask means including a plurality of elliptical masks applied to a portion of a processed image obtained by linearizing the image of FIG. 7 containing the contour section to be recognized and identified, and an enlarged view of one of the masks containing a portion of the contour section. [Figure 11a] 11a to 11c show enlarged views of two elliptical masks each enclosing a portion of a contour section, and an enlarged view of a rectangular mask enclosing a portion of a contour section, which is not the object of the present invention. [Figure 11b] 11a to 11c show enlarged views of two elliptical masks each enclosing a portion of a contour section, and an enlarged view of a rectangular mask enclosing a portion of a contour section, which is not the object of the present invention. [Figure 11c]11a to 11c show enlarged views of two elliptical masks each enclosing a portion of a contour section, and an enlarged view of a rectangular mask enclosing a portion of a contour section, which is not the object of the present invention. [Figure 12a] 12a to 12d show a plurality of masks of the masking means arranged along the detection line and superimposed on the horizontal and vertical contour sections of the outline of the piece in FIG. 8, respectively. [Figure 12b] 12a to 12d show a plurality of masks of the masking means arranged along the detection line and superimposed on the horizontal and vertical contour sections of the outline of the piece in FIG. 8, respectively. [Figure 12c] 12a to 12d show a plurality of masks of the masking means arranged along the detection line and superimposed on the horizontal and vertical contour sections of the outline of the piece in FIG. 8, respectively. [Figure 12d] 12a to 12d show a plurality of masks of the masking means arranged along the detection line and superimposed on the horizontal and vertical contour sections of the outline of the piece in FIG. 8, respectively. DETAILED DESCRIPTION OF THE INVENTION

[0023] With reference to Figure 1, a machine 1 for processing and / or moving sheet metal and / or metal plates according to the present invention is partially and diagrammatically shown, comprising operating means 2 capable of processing and / or moving pieces 50, 51, 52 of sheet metal and / or metal plates, an artificial vision system 10 provided with one or more cameras 11 adapted to acquire one or more images 20 of one or more pieces 50, 51, 52 positioned on a working surface 3 of the machine 1, and a control unit 5 connected to the operating means 2 and to the artificial vision system 10 and arranged to configure and / or adjust operating parameters for controlling the operating means 2, in particular based on geometric information of the pieces.

[0024] The pieces include, for example, sheet metal 50 or metal plate, especially rectangular, or semi-finished products 51 (i.e. pieces that have already been partially processed, as shown in FIG. 1), or scrap sheet metal or metal plate 52 (obtained from previous processing, as shown in FIG. 5).

[0025] In the embodiment shown in Figure 1, the machine 1 is, for example, a machine for processing pieces, for example for laser cutting, and the operating means 2 comprise a laser cutting head movable above a fixed work surface 3 along three orthogonal axes XYZ of the machine's reference system S (the origin of which is a defined reference point, for example on the work surface 3).

[0026] The artificial vision system 10 comprises one or more, for example two, commercially available cameras 11 of a known kind arranged to acquire at least one image 20 of one or more pieces positioned on the work surface 3. A processing unit 12 is provided for processing the acquired image 20.

[0027] The processing unit 12 is connected to the camera 11 and the control unit 5 for transmitting to the control unit 5 geometric information obtained by processing the image 20 with respect to the piece / pieces 50, as will be better explained in the following description.

[0028] The processing unit 12 may be separate and distinct from the control unit 5, e.g., associated with the artificial vision system 10, or may be embedded in the control unit 5, i.e., occupy the same space as the control unit 5, and serve as the processing unit 12 for the images 20 acquired by the camera 11.

[0029] The processing unit 12 is configured to process the images 20 acquired by the camera 11 using a deep learning algorithm to identify and obtain a processed shape image 30 in which the shapes 150 of the pieces 50 with their respective contours 160 are represented. In the processed shape image 30, the shapes 150 of the pieces 50 acquired by the deep learning algorithm are superimposed on the image of the pieces 50 acquired by the camera (Figures 3 and 4).

[0030] "Contour" means a line or a collection of straight and / or curved lines that limit and delimit the area that defines shape 150, i.e., the form processed by the deep learning algorithm that identifies and defines piece 50.

[0031] It should be noted that these contour lines include the actual edges and / or corners of the piece that define the extension of the piece, as well as lines that separate areas of the image that have different intensities and / or colors.

[0032] Contour 160 includes the outer contour of the piece (e.g., the lateral straight sections or edges 160a of the two long sides and the straight sections or edges 160b of the two short sides of the sheet metal in Figures 2 to 5), but also includes inner contours that define internal areas that have already been machined into the piece, such as holes, openings, etc. (e.g., the openings made in the semi-finished product 51 in Figure 1 and the scrap 52 in Figure 5).

[0033] Deep learning algorithms are a known type of artificial intelligence algorithm, for example Google's DeepLab v3 algorithm, which, after an appropriate training step, is able to identify and recognize pieces of different shapes and sizes, processing each shape or mask with identifiable contours or edges.

[0034] If the contour 160 of the shape 150 processed by the deep learning algorithm is precisely and completely defined and can be distinguished from the background of the processed shape image 30, i.e., from the work surface 3, overall, i.e., in all of its contour sections 160a, 160b (thus the contour 160 can be completely and precisely recognized and identified), the processing unit 12 can recognize and identify one or more contour sections 160a, 160b of the contour 160 of the shape 150 and extract geometric information of the piece 50 based on the identified contour sections 160a, 160b.

[0035] As is known, a contour or contour section is defined accurately and distinguishably from the background of an image when any contour recognition algorithm of known type, such as the Canny algorithm, succeeds in detecting such a contour or section clearly, correctly and completely, i.e., when it is able to successfully complete the recognition process.

[0036] Conversely, if the contour 160 of the shape 150 processed by the deep learning algorithm is not precisely and completely defined (especially when the deep learning algorithm roughly identifies the region of the image 20 of the piece 50 where the material is depicted, but is unable to have the edge recognition algorithm accurately identify all contour sections), and includes at least one section that is indistinguishable or poorly or barely distinguishable from the background of the processed shape image 30, i.e., from the work surface 3 (and therefore the contour 160 cannot be accurately and completely recognized and identified), the processing unit 12 processes the image 20 previously acquired by the camera 11 using a known contour recognition algorithm, such as the Canny algorithm, to obtain a processed outline image 40 including an outline 260 of the piece 50. By "outline" 260 is meant a line or a set of straight and / or curved lines that delimit and define the outer contour of the piece 50 and / or the inner contour of holes, openings, etc. present in the piece.

[0037] The processing unit 12 is further configured to process the processed outline image 40 using a mask means 15 applied to at least a portion of the processed outline image 40 that contains at least one contour section 260a of the outline 260 that is not sufficiently precise, i.e., indistinguishable or difficult to distinguish from the background of the processed outline image 40, i.e., from the work surface 3.

[0038] The portions of the processed outline image 40 to which the mask means 15 should be applied are identified using the shape 150 in the processed shape image 30, in particular by superimposing the two processed images, the shape image 30 and the outline image 40. In other words, the processed shape image 30 processed by the deep learning algorithm can be used to identify one or more portions of the processed outline image 40 that contain contour sections 260a of the outline 260 that are not precise enough to be distinguished from the background, and to which the mask means 15 should be applied in order to process the processed outline image 40 and recognize and identify the contour sections 260a of the outline 260.

[0039] Therefore, the mask means 15 can advantageously separate the contour section 260a of the outline 260 demarcated by the mask means 15 from the background of the image, i.e., the work surface 3, particularly more precisely and clearly, so that the processing unit 12 can precisely recognize and identify the contour section 260a of the outline 260.

[0040] In particular, the mask means 15 includes one or more masks 16, in particular a plurality of masks 16 aligned and regularly spaced along the detection lines R, T and superimposable on the contour sections 260a of the outline 260, in particular contour sections that are not precisely defined and not completely distinguishable, each mask 16 being defined by a closed flat curve, in particular an elliptical, oval or circular shape, that surrounds a respective part of the processed outline image 40 containing a respective part of the contour section 260a.

[0041] 9, 10, 11a and 11b, which show an elliptical mask 16, this very elliptical shape allows for a better separation of outline contour sections 260a that are substantially straight, imprecise and / or difficult to distinguish due to the outline contour sections 260a being partial and / or due to the presence of other "disturbance" or "noise" sections 261, 262 (FIG. 9), for example belonging to the support elements of the work surface 3, and / or the reflections and / or finish of the surface of the piece 50 (FIG. 10), and / or due to low image contrast. In particular, horizontal (FIG. 11b) or diagonal (FIG. 11a) contour sections 260a can be clearly and unambiguously recognized and identified, even if they are flanked by several "noise" sections.

[0042] Also, Figure 11c shows a rectangular mask which is not the object of the present invention, but which introduces at least two straight line segments which can be replaced with the contour segments actually searched, for example forming the two upper edges of the mask.

[0043] A further advantage of using a mask 16 having an elliptical shape is that it takes less time to process only the portion of the processed outline image 40 defined by the mask 16 compared to the time required to process the entire processed outline image 40.

[0044] The mask 16 of the mask means 15 can be positioned along the horizontal detection line R and can be superimposed on the horizontal contour sections of the lower side 260a and upper side 260b of the outline 260 of the piece 50 (Figures 12a and 12b), and can be positioned along the vertical detection line T and can be superimposed on the vertical contour sections of the left side 260c and right side 260d of the outline 260 of the piece 50 (Figures 12c and 12d).

[0045] The processing unit 12 can then extract geometric information about the piece 50 based on one or more contour sections 260a of the outline 260 recognized and identified in the processed outline image 40 using the mask means 15, with respect to the reference system S of the machine 1.

[0046] The geometric information includes at least the distance and rotation angle of one or more of the contour sections 160a, 160b, 260a of the contour 160 or outline 260 of the shape 150 recognized and identified in the processed image 30, 40 of the piece with respect to the reference system S.

[0047] The geometric information may also include the size of one or more contour sections, the shape and size of the entire piece, and / or the location, orientation, shape and size of any internal regions of the processed (semi-finished) piece, in particular openings, holes, slots, etc.

[0048] The processing unit 12 is configured to send the geometric information of the pieces thus obtained to the control unit 5 of the machine 1 in order to configure and / or adjust the operating parameters of the operating means 2 .

[0049] In particular, the geometric information extracted from the artificial vision system 10 enables the control unit 5 to precisely calculate the position and orientation of each piece 50 placed on the work surface 3 relative to the reference system S of the machine 1. The "position" of a piece means, for example, the distance of the reference point of the piece from the origin of the machine reference system S along two orthogonal axes X and Y, and the "orientation" of a piece means the angle formed by one of the two axes of the reference system and the reference section of the piece's contour.

[0050] The operation of the machine 1 can be described by the various steps that define the inventive method for processing sheet metal and / or metal plates, which are described below.

[0051] The method according to the invention for processing sheet and / or plate metals on a machine 1 provided with operating means 2 for processing and / or moving pieces 50, 51, 52 of sheet and / or plate metal, an artificial vision system 10 for acquiring an image 20 of at least one piece 50, 51, 52, and a control unit 5 connected to the artificial vision system 10 for controlling the operating means 2, comprising: - positioning at least one piece 50, 51, 52 to be processed on the working surface 3 of the machine 1; - acquiring at least one image 20 of the pieces 50, 51, 52 using at least one camera 11 of the artificial vision system 10; - processing the image 20 using a deep learning algorithm to identify and obtain a processed shape image 30 containing the shapes 150 of the pieces 50, 51, 52 with their respective contours 160; - recognizing and identifying at least one contour section 160a, 160b of the contour 160 of the shape 150 in the processed shape image 30, provided that the contour 160 is precisely and completely defined and completely distinguishable from the background of the processed shape image 30, i.e. from the working surface 3; or if the contour 160 of the shape 150 in the processed shape image 30 is not precisely and completely defined and contains at least one contour section that is indistinguishable, i.e., difficult to distinguish or barely distinguishable, from the background of the processed shape image 30; - processing the image 20 using a contour recognition algorithm to obtain a processed outline image 40 containing the outlines 260 of the pieces 50, 51, 52; processing the processed outline image 40 with a mask means 15 applied to at least a portion of the processed outline image 40, identified on the basis of a shape 150 in the processed shape image 30, in particular by superimposing the processed shape image 30 and the processed outline image 40, so as to precisely separate the contour section 260a of the outline 260 delimited by the mask means 15 from the background, in particular containing at least one contour section 260a, 260b, 260c, 260d of the outline 260, indistinguishable from the background of the processed outline image 40, in order to recognize and identify the contour sections 260a, 260b, 260c, 260d of the outline 260; - extracting geometric information of the pieces 50, 51, 52 based at least on the contour 160 or the contour sections 160a, 160b, 260a, 260b, 260c, 260d of the outline 260 recognized and identified in the processed shape image 30 or in the processed outline image 40 using the mask means 15, and with respect to the reference system S of the machine 1; - sending the geometrical information to the control unit 5 to configure and / or adjust the operating parameters of the operating means 2; Includes:

[0052] In particular, the contour 160 and / or the outline 260 of the shape 150 for the piece 50 each include a number of contour sections 160a, 160b, 260a, 260b, 260c, 260d, in particular substantially rectilinear.

[0053] The geometric information includes the distance and rotation angle of one or more contour sections 160 a , 160 b , 260 a , 260 b , 260 c , 260 d of the contour 160 or outline 260 relative to the reference system S of the machine 1 .

[0054] The method further includes calculating the position and orientation of the piece 50 positioned on the work surface 3 relative to the reference system S based on geometric information extracted from an analysis of the contours of the piece recognized and identified in the processed shape image 30 or in the processed outline image 40 using the mask means 15.

[0055] The mask means 15 used comprise in particular at least one mask 16 formed by a closed flat curve, in particular elliptical or oval or circular, surrounding a portion of the processed outline image 40 containing a portion of each of the contour sections 260a, 260b, 260c, 260d to be recognized and identified.

[0056] In particular, the mask means 15 includes a plurality of masks 16, each having the shape of an ellipse, oval or circle, aligned and regularly spaced along the detection lines R, L and superimposable on the defined contour section 260a of the outline 260.

[0057] According to the method of the invention, it is also provided to obtain the image 20 of the piece 50 by superimposing two partial images obtained by two respective cameras 11 of the artificial vision system 10 .

[0058] Pieces to be worked that are positioned on the work surface 3 include sheet metal 50, metal plate, semi-finished pieces 51, scraps of sheet metal or metal plate 52.

[0059] Thus, the machine and method for processing sheet metal and / or metal plates of the present invention by using an artificial vision system 10 provided with a camera 11 and a processing unit 12 (e.g. the same control unit 5 of the machine 1) makes it possible to precisely, accurately and automatically determine the geometric information of the pieces positioned on the working surface 3 of the machine 1, in particular their position and orientation relative to the reference system S of the machine 1, without operator intervention, and based on this geometric information to configure and / or adjust the operating parameters of the operating means 2 for processing and / or moving the pieces 50, 51, 52.

[0060] In particular, by using deep learning algorithms, the images 20 of each piece 50 acquired by the camera 11 can be processed to obtain a processed shape image 30 containing the shape 150 of the piece 50 from which the contour 160 should be directly extracted, i.e., recognized and identified, and then geometric information of the piece 50 (distances and rotation angles of the various contour sections 160a, 160b of the contour 160 relative to the reference system S of the machine 1) can be extracted to configure and / or adjust the operating parameters of the operating means 2 without the need for human intervention by an operator.

[0061] It should be noted that the time required for the deep learning algorithm to process the image and extract the contours 160 of the pieces is less than the time required for known types of contour recognition algorithms to recognize and identify contours in an image.

[0062] Alternatively, if the contour 160 of the shape 150 of the piece 50 in the processed shape image 30 obtained by the deep learning algorithm is not sufficiently defined and clear to allow its direct recognition and identification, this contour 160 and the associated shape 150 can advantageously be used to identify a portion of the processed outline image 40 to which the mask means 15 should be applied in order to better recognize and identify the contour section 260a of the outline 260 of the piece 50 contained in that portion of the processed outline image 40 (obtained by processing the image 20 using the contour recognition algorithm), again without the need for human intervention by an operator.

[0063] It is also possible to determine the position and orientation of the pieces on the working surface 3 of the machine, since no corresponding drawing or reference figure is required for recognition and identification of the contours of the pieces, including sheet metal 50, metal plate, semi-finished product 51, scrap metal or metal plate 52.

[0064] Deep learning algorithms can also be used to determine the inner contours of the piece that define the interior areas that have already been machined, such as holes, openings, etc.

[0065] A variant of the machine for processing sheet metal and / or metal plates of the invention is provided which differs from the above-described embodiment in that the processing unit 12 is configured to process the image 20 acquired by the at least one camera 11 using a contour recognition algorithm so as to obtain a processed outline image 40 comprising the outline 260 of the piece 50.

[0066] The processing unit 12 is also configured to process the processed outline image 40 using mask means 15, which is applied to at least a portion of the processed outline image 40 containing at least one contour section 260a, in particular a rectilinear section, of the outline 260, that is indistinguishable or difficult to distinguish from the background of the processed outline image 40, i.e. from the work surface 3. The mask means 15 comprises at least one mask 16 defined by a closed flat curve, in particular an elliptical or oval or circular shape, surrounding the portion of the processed outline image 40 containing the contour section 260a, so as to separate and highlight the contour section 260a from the background of the processed outline image 40.

[0067] The processing unit 12 is also configured to recognize and identify contour sections 260a of the outline 260, extract geometric information of the piece 50 based on at least the contour sections 260a of the outline 260 recognized and identified in the processed outline image 40 and with respect to the reference system S of the machine 1, and send the geometric information to the control unit 5 for configuring and / or adjusting operating parameters of the operating means 2 arranged to perform at least one established processing and / or movement of the piece 50.

[0068] The processing unit 12 is also configured to identify portions of the processed outline image 40 for which to provide the mask means 15 based on instructions manually provided by an operator or based on the shapes 150 of the pieces 50 with their respective contours 160 shown in the processed shape image 30 obtained by processing the image 20 using an image recognition algorithm.

[0069] In particular, the portion of the processed outline image 40 to which the mask means 15 should be applied can be indicated by the operator by selecting, on the screen of the control panel of the machine connected to the control unit 5 and the processing unit 12, one or more areas of the screen on which the processed outline image 40 is shown that correspond to the portion of the processed outline image 40 to be identified.

[0070] Alternatively, a deep learning algorithm can be used as an image recognition algorithm to process the image 20 acquired by the camera 11 to identify and obtain a processed shape image 30 containing the shapes 150 of the pieces 50 with their respective contours 160.

[0071] The operation of this variant of the machine 1 of the invention can be explained by the following different steps defining the method of the invention for processing and / or moving metal sheets and plates.

[0072] The method according to the invention for processing and / or moving sheet and / or metal plates on a machine 1 provided with operating means 2 for processing and / or moving pieces 50, 51, 52 of sheet and / or metal plates, an artificial vision system 10 for acquiring an image 20 of at least one piece 50, 51, 52, and a control unit 5 connected to the artificial vision system 10 for controlling the operating means 2, comprising: - positioning at least one piece 50, 51, 52 to be processed on the working surface 3 of the machine 1; - acquiring at least one image 20 of the pieces 50, 51, 52 using at least one camera 11 of the artificial vision system 10; - processing the image 20 using a contour recognition algorithm to obtain a processed outline image 40 containing the outlines 260 of the pieces 50, 51, 52; processing the processed outline image 40 using mask means 15 applied to at least a portion of the processed outline image 40 containing at least one contour section 260a, in particular substantially rectilinear, of the outline 260, indistinguishable from the background of the processed outline image 40, the mask means 15 comprising at least one mask 16 defined by a closed flat curve, in particular elliptical or oval or circular, surrounding the portion of the processed outline image 40 containing the contour section 260a, so as to separate and highlight the contour section 260a from the background of the processed outline image 40, i.e. from the working surface 3; - recognizing and identifying the contour sections 260a of the outline 260; - extracting mechanical information of the pieces 50, 51, 52 based on at least the contour section 260a of the outline 260 recognized and identified in the processed outline image 40 and with respect to the reference system S of the machine 1; sending geometric information to the control unit 5 for configuring and / or adjusting the operating parameters of the operating means 2; Includes:

[0073] The outline 260 includes a plurality of contour sections 260a, 260b, 260c, 260d, which are in particular substantially linear, and processing the processed outline image 40 with the mask means 15 includes applying the mask means 15 to each of the plurality of contour sections 260a, 260b, 260c, 260d.

[0074] The mask means 15 includes a plurality of masks 16 arranged in alignment along the detection lines R, T and regularly spaced apart, each defined by a closed flat curve, in particular an elliptical or oval or circular shape, superimposable on the contour sections 260a, 260b, 260c, 260d of the outline 260.

[0075] The geometric information of the piece 50 includes the distance and rotation angle of one or more contour sections 260a, 260b, 260c, 260d of the outline 260 relative to the reference system S.

[0076] The method further includes calculating the position and orientation of the piece 50 positioned on the work surface 3 with respect to the reference system S based on geometric information extracted from an analysis of the contours of the pieces recognized and identified in the processed outline image 40 using the mask means 15.

[0077] The method includes identifying portions of the processed outline image 40 to which the mask means 15 should be applied based on instructions manually provided by an operator or based on the shapes 150 of the pieces 50 represented in a processed shape image 30 having respective contours 160 and obtained by processing the image 20 using an image recognition algorithm.

[0078] Advantageously, the image recognition algorithm is a deep learning algorithm adapted to process the image 20 and identify and obtain a processed shape image 30 containing the shapes 150 of the pieces 50 with their respective contours 160.

[0079] According to the method of the invention, it is also provided to obtain the image 20 of the piece 50 by superimposing two partial images obtained by two respective cameras 11 of the artificial vision system 10 .

[0080] Pieces to be worked positioned on the work surface include sheet metal 50, sheet metal, semi-finished products 51, scrap sheet metal or sheet metal 52.

[0081] Using the variants of the machine and method for processing sheet metal and / or metal plates of the present invention, it is possible to accurately determine the geometric information of the pieces 50, 51, 52 positioned on the working surface 3 of the machine 1, in particular their position and orientation relative to the reference system S of the machine 1, and based on this geometric information it is possible to configure and / or adjust the operating parameters of the operating means 2 for processing and / or moving the pieces.

[0082] In particular, the mask means 15 comprising a plurality of elliptical masks 16 aligned along the detection lines R, T and regularly spaced apart, applied to a portion of the processed outline image 40 obtained by processing the image 20 acquired by the camera 11 using a contour recognition algorithm, allows for precise and rapid recognition and identification of contour sections 260a of the outline 260, even though they are difficult to define and distinguish from the background, as shown in Figures 10 to 12. The portion or portions of the processed outline image 40 to which the masks 16 are applied can be identified based on instructions provided manually by an operator acting on the machine's control panel, or advantageously based on the shape 150 of the piece 50 represented in the processed shape image 30 by processing the same image 20 of the piece 50 using an image recognition algorithm, in particular a deep learning algorithm.

[0083] Based on the contour sections 260a, 260b, 260c, 260d of the outline 260 thus recognized and identified in the processed outline image 40, precise and accurate geometric information of the piece 50 can be extracted, which makes it possible to calculate the position and orientation of the piece 50 on the work surface 3 relative to the reference system S.

[0084] Thus, recognition and identification of the contours of pieces including sheet metal 50, metal plate, semi-finished product 51, scrap or skeleton of sheet metal or metal plate 52 does not require a corresponding drawing or reference figure, so that the position and orientation of the piece on the working surface 3 of the machine 1 can be determined.

Claims

1. A machine (1) for processing and / or transporting metal sheets and / or metal plates, comprising: operating means (2) for processing and / or moving pieces of sheet metal and / or metal plates (50, 51, 52); an artificial vision system (10) provided with at least one camera (11) capable of acquiring at least one image (20) of at least one piece (50, 51, 52) positioned on a work surface (3) of the machine (1); a control unit (5) connected to the artificial vision system (10) for controlling the operating means (2); Equipped with The machine (1) processing said image (20) using a deep learning algorithm to identify and obtain a processed shape image (30) comprising shapes (150) of said pieces (50, 51, 52) with their respective contours (160); Recognizing and identifying at least one contour section (160a, 160b) of said contour (160), if said contour (160) is precisely and completely delineated and sufficiently distinguishable from the background of said processed shape image (30); or If the contour (160) of the shape (150) is not precisely and completely depicted and includes at least a contour section that is indistinguishable from the background of the processed shape image (30), processing said image (20) using a contour recognition algorithm to obtain a processed outline image (40) containing outlines (260) of said pieces (50, 51, 52); processing the processed outline image (40) by means of a masking means (15) applied to at least a portion of the processed outline image (40) containing at least one contour section (260a, 260b, 260c, 260d) of the outline (260), in particular indistinguishable from the background of the processed outline image (40); and identifying the portion of the processed outline image (40) based on the shape (150) in the processed shape image (30), in particular by superimposing the processed shape image (30) with the processed outline image (40), so as to precisely separate the contour sections (260a, 260b, 260c, 260d) of the outline (260) demarcated by the masking means (15) against the background in order to recognize and identify the contour sections (260a, 260b, 260c, 260d) of the outline (260); extracting geometric information of the piece (50, 51, 52) based on the at least one contour section (160a, 160b, 260a, 260b) of the contour (160) or the outline (260) recognized and identified in the processed shape image (30) or the processed outline image (40) and with respect to a reference system (S) of the machine (1); Sending said geometric information to said control unit (5) for configuring and / or adjusting the operating parameters of said operating means (2). A machine (1) comprising a processing unit (12) configured to:

2. 2. The machine (1) according to claim 1, wherein the geometric information comprises at least a distance and a rotation angle of one or more contour sections (160a, 160b, 260a, 260b, 260c, 260d) of the contour (160) or the outline (260) recognized and identified in the processed shape image (30) or the processed outline image (40) relative to the reference system (S).

3. 3. The machine (1) according to claim 1 or 2, wherein the control unit (5) is configured to calculate at least a position and an orientation of the at least one piece (50, 51, 52) on the work surface (3) relative to the reference system (S) based on the geometric information.

4. 4. The machine (1) according to claim 1, wherein the masking means (15) comprises at least one mask (16) formed by a closed plane curve, in particular an elliptical or oval or circular shape, surrounding the portion of the processed outline image (40) containing a portion of each of the contour sections (260a, 260b, 260c, 260d).

5. 5. The machine (1) according to any one of claims 1 to 4, wherein the masking means (15) comprises a plurality of masks (16) each formed by a closed planar curve, in particular an elliptical, oval or circular shape, the masks being aligned and regularly spaced along the detection line (R, T) and superimposable on the contour sections (260a, 260b, 260c, 260d) of the outline (260).

6. 6. Machine (1) according to any one of the preceding claims, wherein the operating means (2) comprise one or more of cutting means, punching means, bending and / or shaping means, moving and / or manipulating means.

7. The machine (1) according to any one of claims 1 to 6, wherein the processing unit (12) is included in the control unit (5).

8. 1. A method for processing and / or moving sheet metal and / or metal plates in a machine (1) provided with operating means (2) for processing and / or moving pieces (50, 51, 52) of sheet metal and / or metal plates, an artificial vision system (10) for acquiring an image (20) of at least one piece (50, 51, 52), and a control unit (5) connected to the artificial vision system (10) for controlling the operating means (2), comprising: positioning at least one piece (50, 51, 52) to be processed on a working surface (3) of the machine (1); acquiring at least one image (20) of said pieces (50, 51, 52) using at least one camera (11) of said artificial vision system (10); processing said image (20) using a deep learning algorithm to identify and obtain a processed shape image (30) containing shapes (150) of said pieces (50, 51, 52) with their respective contours (160); Recognizing and identifying at least one contour section (160a, 160b) of said contour (160), if said contour (160) is precisely and completely delineated and sufficiently distinguishable from the background of said processed shape image (30); or If the contour (160) of the shape (150) is not precisely and completely depicted and includes at least a contour section that is indistinguishable from the background of the processed shape image (30), processing said image (20) using a contour recognition algorithm to obtain a processed outline image (40) containing outlines (260) of said pieces (50, 51, 52); processing the processed outline image (40) by means of a masking means (15) applied to at least a portion of the processed outline image (40), in particular containing at least one contour section (260a, 260b, 260c, 260d) of the outline (260) that is indistinguishable from the background of the processed outline image (40) and identified on the basis of the shape (150) in the processed shape image (30), in particular by superimposing the processed shape image (30) and the processed outline image (40), so as to precisely separate the contour section (260a, 260b, 260c, 260d) of the outline (260) demarcated by the masking means (15) against the background in order to recognize and identify the contour section (260a, 260b, 260c, 260d) of the outline (260); extracting geometric information of the pieces (50, 51, 52) based on at least the contour sections (160a, 160b, 260a, 260b, 260c, 260d) of the contour (150) or the outline (260) recognized and identified in the processed shape image (30) or the processed outline image (40) and with respect to a reference system (S) of the machine (1); sending said geometric information to said control unit (5) to configure and / or adjust the operating parameters of said operating means (2); A method comprising:

9. The method of claim 8, wherein the contour (160) and / or the outline (260) comprises a plurality of contour sections (160a, 160b, 260a, 260b, 260c, 260d).

10. 10. The method according to claim 8 or 9, wherein the geometric information comprises at least distances and rotation angles of one or more contour sections (160a, 160b, 260a, 260b, 260c, 260d) of the contour (160) or the outline (260) relative to the reference system (S) of the machine (1).

11. 11. The method according to any one of claims 8 to 10, comprising calculating at least a position and an orientation of said at least one piece (50, 51, 52) on said work surface (3) relative to said reference system (S) based on said geometric information.

12. 12. The method according to any one of claims 8 to 11, wherein the masking means (15) comprises at least one mask (16) formed by a closed planar curve, in particular an elliptical or oval or circular shape, surrounding the part of the processed outline image (40) containing a part of each of the contour sections (260a).

13. 13. The method according to any one of claims 8 to 12, wherein the mask means (15) comprises a plurality of masks (16), in particular each having the shape of an ellipse, an oval or a circle, the masks being aligned and regularly spaced along the detection line (R, T) and being superimposable on the contour sections (260a, 260b, 260c, 260d) of the outline (260).

14. 14. The method according to any one of claims 8 to 13, wherein the pieces (50, 51, 52) to be processed comprise one of sheet metal (50), metal plate, semi-finished product (51), sheet metal scrap (52), or sheet metal scrap.

15. A machine (1) for processing and / or transporting metal sheets and / or metal plates, comprising: operating means (2) for processing and / or moving pieces of sheet metal and / or metal plates (50, 51, 52); an artificial vision system (10) provided with at least one camera (11) capable of acquiring at least one image (20) of at least one piece (50, 51, 52) positioned on a work surface (3) of the machine (1); a control unit (5) connected to the artificial vision system (10) for controlling the operating means (2); Equipped with The machine (1) processing said image (20) using a contour recognition algorithm to obtain a processed outline image (40) comprising outlines (250) of said pieces (50, 51, 52); processing the processed outline image (40) by means of a masking means (15) applied to at least a portion of the processed outline image (40) containing at least one contour section (260a, 260b, 260c, 260d) of the outline (260), in particular indistinguishable from the background of the processed outline image (40), the masking means (15) comprising at least one mask (16) formed by a closed plane curve, in particular an elliptical, oval or circular shape, surrounding the portion of the processed outline image (40) so as to separate and highlight the contour section (260a, 260b, 260c, 260d) from the background, Recognizing and identifying the contour sections (260a, 260b, 260c, 260d) of the outline (260), extracting geometric information of the pieces (50, 51, 52) based on at least the contour sections (260a, 260b, 260c, 260d) of the outline (260) recognized and identified in the processed outline image (40) and with respect to a reference system (S) of the machine (1); Sending said geometric information to said control unit (5) for configuring and / or adjusting the operating parameters of said operating means (2). A machine (1) comprising a processing unit (12) configured to:

16. 16. The machine (1) according to claim 15, wherein the geometric information comprises at least distances and rotation angles of one or more contour sections (260a, 260b, 260c, 260d) of the outline (260) recognized and identified in the processed outline image (40) relative to the reference system (S).

17. 17. The machine (1) according to claim 15 or 16, wherein the control unit (5) is configured to calculate at least a position and an orientation of the at least one piece (50, 51, 52) on the work surface (3) relative to the reference system (S) based on the geometric information.

18. 18. The machine (1) according to any one of claims 15 to 17, wherein the masking means (15) comprises a plurality of masks (16) each formed by a closed planar curve, in particular an elliptical, oval or circular shape, the masks being aligned and regularly spaced along the detection line (R, T) and superimposable on the contour section (260a) of the outline (260).

19. 19. The machine (1) according to any one of claims 15 to 18, wherein the processing unit (12) is configured to identify the at least part of the processed outline image (40) to which the mask means (15) should be applied based on instructions manually provided by an operator or based on shapes (150) of the pieces (50, 51, 52) with their respective contours (160) shown in a processed shape image (30) obtained by processing the image (20) using an image recognition algorithm, in particular a deep learning algorithm.

20. 20. Machine (1) according to any one of claims 15 to 19, wherein said operating means (2) comprise one or more of cutting means, punching means, bending and / or shaping means, moving and / or manipulating means.

21. The machine (1) according to any one of claims 15 to 20, wherein the processing unit (12) is included in the control unit (5).

22. 1. A method for processing and / or moving sheet metal and / or metal plates in a machine (1) provided with operating means (2) for processing and / or moving pieces (50, 51, 52) of sheet metal and / or metal plates, an artificial vision system (10) for acquiring an image (20) of at least one piece (50, 51, 52), and a control unit (5) connected to the artificial vision system (10) for controlling the operating means (2), comprising: positioning at least one piece (50, 51, 52) to be processed on the working surface (3) of the machine (1); acquiring at least one image (20) of said pieces (50, 51, 52) using at least one camera (11) of said artificial vision system (10); processing said image (20) using a contour recognition algorithm to obtain a processed outline image (40) containing the outlines (260) of said pieces (50, 51, 52); processing the processed outline image (40) by means of a masking means (15) applied to at least a portion of the processed outline image (40) containing at least one contour section (260a, 260b, 260c, 260d) of the outline (260), in particular indistinguishable from the background of the processed outline image (40), the masking means (15) comprising at least one mask (16) formed by a closed planar curve, in particular an elliptical, oval or circular shape, surrounding said portion of the processed outline image (40) so as to separate and highlight said contour section (260a, 260b, 260c, 260d) from the background of the processed outline image (40); Recognizing and identifying the contour sections (260a) of the outline (260); - extracting geometric information of said pieces (50, 51, 52) based on at least said contour sections (260a, 260b, 260c, 260d) of said outline (260) recognized and identified in said processed outline image (40) and with respect to a reference system (S) of said machine (1); sending said geometric information to said control unit (5) for configuring and / or adjusting the operating parameters of said operating means (2); A method comprising:

23. 23. The method of claim 22, wherein the outline (260) comprises a plurality of contour sections (260a, 260b, 260c, 260d), and wherein the step of processing the processed outline image (40) with the mask means (15) comprises applying the mask means (15) to each of the plurality of contour sections (260a, 260b, 260c, 260d).

24. 24. The method according to claim 22 or 23, wherein the masking means (15) comprises a plurality of masks (16) each formed by a closed planar curve, in particular an elliptical or oval or circular shape, the masks being aligned and regularly spaced along the detection line (R, T) and superimposable on the contour section (260a) of the outline (260).

25. 25. The method according to any one of claims 22 to 24, wherein the geometric information comprises at least a distance and a rotation angle of one or more contour sections (260a, 260b) of the outline (260) relative to the reference system (S).

26. 26. The method according to any one of claims 22 to 25, comprising calculating at least a position and an orientation of said at least one piece (50) on said work surface (3) relative to said reference system (S) based on said geometric information.

27. 27. The method of any one of claims 22 to 26, comprising identifying the at least part of the processed outline image (40) to which the mask means (15) should be applied based on instructions manually provided by an operator or based on the shape (150) of the piece (50) with its respective contour (160) shown in a processed shape image (30) obtained by processing the image (20) using an image recognition algorithm.

28. 28. The method of claim 27, wherein the image recognition algorithm is a deep learning algorithm configured to process the image (20) and identify and obtain the processed shape image (30) containing shapes (150) of the pieces (50) with their respective contours (160).

29. 29. The method of any one of claims 22 to 28, wherein the pieces (50, 51, 52) to be processed comprise one of sheet metal (50), metal plate, semi-finished product (51), sheet metal scrap (52), or sheet metal scrap.

Images