Method, mobile terminal device and system for evaluating laser cutting edges

Abstract

The invention relates to a method for evaluating a laser cut edge (16) of a workpiece (12), the method comprising the following steps: A) detecting image data of the laser cut edge (16) and its surroundings; B) segmenting the image data and identifying an interesting section of the image data, wherein the interesting section has image data of the laser cut edge (16); C) performing image quality recognition for the interesting section; D) generating an output for a user based on the image quality recognition. The invention also relates to a mobile terminal device and a system.

Description

Technical Field

[0001] The invention relates to a method for evaluating a laser cut edge, a mobile terminal device and a system for executing the steps of the method. Background Art

[0002] Laser cutting, also known as laser beam cutting, refers to the cutting of solid bodies by material ablation with the aid of continuous or pulsed laser radiation. In this way, metals, plastics, glass and organic materials can be cut. The parameters of the laser radiation, such as the wavelength of the application, the average power, the pulse energy and the pulse duration, must be adapted accordingly. In addition, a cutting nozzle, which is usually arranged concentrically with the laser, supplies a process gas, which protects the focusing optics from steam and splashes and further removes the ablated material from the cut seam. In particular when cutting metals, in particular sheet metal, the cut edges can vary greatly. Depending on the temperature reached in the effective range and the type of process gas supplied, different aggregation states of the seam material can occur. Characteristic features of high cutting quality are that there are few grooves formed on the cutting edge and no burrs formed on the bottom side of the cut. Through continuous feeding and the resulting melting of the material, the material can solidify on the cutting edge. Solidification can be carried out in a wave-like manner, which can be accompanied by the characteristic groove structure or roughness of the cut. The formation of flash or burrs (Bart-oder Gratbildung) is usually attributed to the small driving force of the gas flow. The droplets on the lower edge of the cut can solidify and form flash / burrs that are more or less strongly attached. Parameters that affect the cutting quality include, among others: focal position, feed speed, laser power, intensity distribution or cutting gas pressure. The evaluation of the cutting edge is usually achieved by observation or feel of the worker or by measuring the height profile using expensive measuring sensor devices. Depending on the subsequent use of the workpiece being cut, very different and sometimes very high requirements are placed on the quality of the cutting edge.

[0003] US 2006 / 0049158 A1 describes a method for controlling an automated laser process, wherein the results of the process are recorded with a camera and used for further controlling the process. Summary of the invention

[0004] In contrast, the object of the present invention is to further improve the quality of the evaluation of process results.

[0005] This object is achieved by a method according to claim 1, a mobile terminal according to claim 8 and a system according to claim 15. The dependent claims relate to advantageous developments of the invention.

[0006] The method for evaluating a laser cut edge of a workpiece has the following steps:

[0007] A) detecting image data of the laser cutting edge and its surrounding environment;

[0008] B) segmenting the image data and identifying a section of interest of the image data, wherein the section of interest has image data of the laser cut edge;

[0009] C) performing image quality identification on the segment of interest;

[0010] D) Generating an output for a user based on the image quality identification.

[0011] The workpiece is preferably made of a solid material, in particular a highly reflective material, such as glass or metal, in particular a sheet of metal.

[0012] In the present method, in step A), an image of the laser cut edge and its surroundings is captured by means of a camera. The captured image is stored in a digital form as image data in an image file. A pixel is a single image point of an image. It is possible that the image portion captured by the camera includes areas that are not related to the laser cut edge, but are located on the workpiece at a distance from the laser cut edge or even outside the workpiece. Compared to the image portion directly related to the laser cut edge, this so-called surroundings of the laser cut edge are less relevant for judging the image quality.

[0013] The segmentation of the image data in step B) is carried out by means of electronic image processing and results in the identification of an interesting section of the image, wherein the interesting section of the image has image data of the laser cut edge. In order for the method to work well, the interesting section should have image data of the laser cut edge and only to a small extent image data of the surroundings of the laser cut edge.

[0014] Preferably, in step B), for each pixel of the image data, it is determined whether the pixel is part of the laser cut edge or part of the surroundings of the laser cut edge. Further preferably, as a result of step B), for each pixel of the image data, it is stored whether it is part of the laser cut edge or part of the surroundings of the laser cut edge.

[0015] In one embodiment of the present invention, step B) is implemented with the aid of a neural network. A neural network, also known as an artificial neural network, is an algorithm that is usually run on a computer. It has an input layer and an output layer and optionally one or more intermediate layers, and can be trained, for example, with the aid of "deep learning". In the input layer, data, such as image data, can be fed to the neural network. In the output layer, the neural network can output suggestions for results, such as suggestions for segmenting image data. These suggestions can be additionally provided with the following values ​​by the neural network: the value outputs a probability with which the algorithm evaluates the segmentation as being successfully performed. Then, the user can evaluate the correctness of the output data. This evaluation of the user can use the neural network to improve one or more of its algorithms. For this purpose, in one or more intermediate layers, the data of the input layer can be associated with other data using factors or calculation functions, and thus new data is generated in the intermediate layer. The other data can be the data of the intermediate layer or another intermediate layer or the output layer. This matching of factors or calculation functions can be called "training" of the neural network. By training the neural network, the neural network can learn from past experience. In the training phase of the neural network, example images of the training database can also be used in which the laser cut edge is imaged in front of a (for example monochrome, in particular green) background. For these example images, labels can be added to the pixels by means of image processing, which pixels indicate which pixel belongs to the laser cut edge and which pixel belongs to the (for example green) background. The set of these labels and the associated pixels is called the "ground truth" of the associated example image. In addition, the green background can be replaced by another background and thus simulate a real shooting scene. In this way, the training database can be enlarged.

[0016] In one embodiment of the invention, a combination of a "Spatial Pyramid Pooling" algorithm and an "Encoder-Decoder Algorithmus" is used to identify segments of interest. Preferably, the two algorithms are combined in such a way that a plurality of semantic information is contained in the encoder module and detailed object boundaries, such as boundaries of laser cut edges, are obtained in a simple and efficient decoder module. The encoder module allows features of arbitrary resolution to be extracted with the aid of an "Atrous Convolution" algorithm. An example of a suitable algorithm is the DeepLabV3+ network, as described, for example, in "Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation" by Chen, LC, Zhu, Y., Papandreou, G., Schroff, F., Adam H. et al. (arXiv: 1802.02611v3 [cs.CV]; August 22, 2018).

[0017] In step C), an image quality detection of the section of interest is performed. The image quality detection is limited to the section of interest, since the section of interest primarily represents the laser cut edge. The image quality detection of the surroundings of the laser cut edge, i.e., areas of the image that do not represent the laser cut edge, is less relevant for the assessment of the quality of the laser cut edge.

[0018] Step C) preferably includes one or more aspects of image quality, such as exposure identification and / or sharpness identification.

[0019] For sharpness recognition, the image data of the image is preferably transformed into the frequency domain (Frequenzbereich). The transformation to the frequency domain can be achieved, for example, by means of Fourier transformation. Then, the sharpness recognition is preferably performed based on the image data in the frequency domain. In one embodiment of the present invention, an algorithm is used for sharpness recognition in the frequency domain, which only works with the image data captured in the section of interest, that is, a contrast image with good or satisfactory sharpness is not required. An example of a suitable algorithm is the NR-IQA method (NR-IQA Methode), as described, for example, in De, K., Masilamani, V.'s "Image Sharpness Measure for Blurred Images in Frequency Domain" (Procedia Engineering No. 64 (2013), pp. 149-158).

[0020] The method enables the user to record an image of the laser cut edge, the quality of which is sufficient to determine the quality of the cut edge himself. In step D), the user receives feedback on the setting of the best possible parameters of the camera and / or feedback on the optimization of, for example:

[0021] -3D clarity, optimal still contrast, motion blur;

[0022] - Light intensity: Avoid overexposure and underexposure and use HDR (High Dynamic Range) to extend the maximum dynamic range;

[0023] -Noise: optimization, ISO parameterization;

[0024] - Gamma correction: automatically matches parameters to ambient light (e.g. daylight, neon lights, etc.) for realistic color reproduction;

[0025] - Aperture: Matched to the highest possible depth of field

[0026] The quality of the cut edge itself can then be determined in step E) of the method, for example, by observing an image of the laser cut edge and / or mechanically by image processing.

[0027] In one embodiment of the method, before step A), the laser cut edge is cut by means of a laser beam and in particular using a compressed air jet. This laser cutting is preferably carried out by means of a laser cutting machine. Preferably, the result of the evaluation of the quality of the laser cut edge is fed back to the laser cutting machine and can thus respond to future cutting processes.

[0028] In one embodiment of the method, prior to step A), parameters related to the laser cut edge and / or the workpiece are received. This receiving can be achieved, for example, by input from a user. This receiving can also be achieved by receiving data from a laser cutting machine or a higher-level controller. In addition, for example, the identification of the workpiece or the form of the workpiece can be detected by a camera, and suggestions about the parameters can be displayed to the user.

[0029] The image data are recorded with the aid of a camera. A camera is an image recording device, in particular in the visible range. A camera can also be a system consisting of a plurality of recording devices and an image processing component, which generates an overall image from a plurality of images. Preferably, the camera is configured to output the image data in digital form.

[0030] When carrying out the method, a mobile terminal with a camera can be used, which is provided to capture image data of the laser cut edge together with the surroundings.

[0031] The mobile terminal device is configured to transmit image data to a computing unit and to receive image data from the computing unit, wherein the received image data are segmented and a section of interest is identified in the received image data, wherein the section of interest has image data of a laser cut edge; and wherein the received image data includes a result of an image quality identification for the section of interest. The mobile terminal device is thus configured to receive image data that are processed in the computing unit according to steps B) and C) of the method.

[0032] In this case, the computing unit can be arranged, for example, in a laser cutting machine or, for example, in a superordinate control unit of a superordinate controller. This may be the case, for example, when the mobile terminal is designed as a webcam.

[0033] In one embodiment of the invention, the computing unit in which steps B) and C) of the method are carried out can also be arranged in the mobile terminal. This may be the case, for example, if the mobile terminal is designed as a mobile phone or a tablet computer.

[0034] In one embodiment, the mobile terminal device can have an output interface, which is configured to generate an output for a user, wherein the output is based on the result of the image quality detection. If the mobile terminal device is designed as a mobile phone, such an output interface is present in the mobile terminal device, for example. The user can then use the camera of the mobile phone to take an image of the laser cut edge and receive feedback via the same mobile phone as to whether the image quality is sufficient and which parameters (e.g. image part, lighting, etc.) still need to be changed in order to obtain a better image quality the next time the image is taken.

[0035] In one embodiment of the invention, the mobile terminal device has at least one interface for receiving parameters related to the laser cutting edge and / or the workpiece. Preferably, the parameters are received before step A) of the method. The receiving can be realized, for example, via an input interface, by an input by a user. The receiving can also be realized, for example, by receiving data from the laser cutting machine or a higher-level controller via a wireless communication interface.

[0036] The steps of the method can be performed on elements of a system. In this case, the system can include a laser cutting machine and a mobile terminal. In this case, the laser cutting machine can be configured to cut the laser-cut edge into a workpiece by means of a laser beam and in particular additionally by means of a compressed air jet. The mobile terminal can be configured as described above.

[0037] The described method, the described mobile terminal device and the described system all enable the highest possible image quality in important and relevant image areas by processing image data and providing targeted feedback to the user. The highest possible image quality that can be achieved can then be a good basis for a further step E) for evaluating the quality of the photographed object, for example, a laser cut edge. The evaluation of step E) can then be carried out manually or automatically. The described method also enables, for example, the use of a smartphone to photograph the laser cut edge. The method offers the advantage that the user can be supported in taking high-quality images of the laser cut edge even if the smartphone's autofocus, for example, cannot focus on the laser cut edge, which may only occupy a small part of the image, or even if the user cannot hold the smartphone completely still in his hand. The user is given the possibility of taking repeated photos through feedback.

[0038] In one embodiment, the invention is present as a machine-readable, in particular non-volatile, in particular tangible storage medium, which is encoded with processor-readable instructions for executing the steps of the method for evaluating a laser cut edge.

[0039] Further advantages of the present invention are obtained from the description and the drawings. According to the present invention, the above-mentioned features and the features to be further implemented can also be used individually or in combination in any combination. The embodiments shown and described should not be understood as exhaustive, but have exemplary characteristics for explaining the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A flow chart schematically illustrating the method;

[0041] Figure 2 A system having a mobile terminal device (10) and a laser cutting machine (14) is schematically shown;

[0042] Figure 3 schematically showing possible displays on a touch-sensitive screen (24);

[0043] Figure 4 The products of the segmentation are shown exemplarily;

[0044] Figure 5 A schematic illustration of an embodiment of a system (100) having a laser cutting machine (14) is shown.

[0045] Figure 6 A schematic diagram of one embodiment of a control system (600) is shown. DETAILED DESCRIPTION

[0046] Figure 1 Schematically showing a method for evaluating a workpiece 12 ( Figure 5 ) of the laser cutting edge 16 ( Figure 5 ), the method comprising steps A), B), C) and D). In step A), image data of a laser cut edge 16 and its surroundings are detected with the aid of a camera 18. In step B), the image data are segmented, in particular with the aid of a neural network, and with the aid of the segmentation, sections of interest in the image data are determined. For the method for evaluating the quality of a laser cut edge, the laser cut edge itself is an important area of ​​the image. Therefore, when segmenting the image data, image points, pixels of the image data located on the laser cut edge 16 are assigned to the section of interest. Other image points, pixels are assigned to the surroundings of the laser cut edge 16. Then, in step C), image quality recognition, in particular image clarity recognition, is performed for the section of interest, preferably in the frequency domain. In step D), an output for the user is generated and preferably output on a screen 24 ( Figure 2 ). In this output, the user is informed of the result of the image quality identification and can therefore obtain an indication of whether the quality of the image is sufficient for a possible subsequent assessment of the quality of the laser cut edge 16 in step E). If the image quality is not sufficient for assessing the quality of the laser cut edge 16, the user can be advised to repeat the recording of the image. This is continued until enough images, for example at least one image, of sufficient quality are available.

[0047] Figure 2 A system with a mobile terminal device 10 and a laser cutting machine 14 is schematically shown. The mobile terminal device 10 has a camera 18, optionally a keyboard 20 as an input interface, a communication interface 22 (for example, a communication interface for wireless communication) and a touch-sensitive screen 24. The screen 24 can also be constructed as a screen without a touch-sensitive surface, that is, without an input function. The mobile terminal device is constructed to communicate with the laser cutting machine 14 or other units via the communication interface 22. This communication is preferably carried out wirelessly, but wired communication can also be provided. Through the communication interface 22, the mobile terminal device can, for example, receive parameters related to the laser cutting edge 16 and thus the workpiece 12 from the laser cutting machine 14. These parameters include, for example, data on the type of material of the workpiece 12, on the type of cutting method and / or on the thickness of the workpiece 12.

[0048] In one embodiment of the present invention, the mobile terminal device 10 is also capable of transmitting image data captured by the camera 18 to a computing unit via a communication interface 22 and receiving image data from the computing unit. The received image data is segmented, and a segment of interest is identified in the received image data. The received image data has a result of image quality recognition for the segment of interest. The computing unit is further configured to segment the image data and identify the segment of interest in the image data, wherein the segment of interest has image data of a laser cutting edge; and wherein the computing unit is configured to perform image quality recognition for the segment of interest. The aforementioned computing unit can, for example, be used in Figure 5 The computing unit 122 shown in FIG.

[0049] In another embodiment of the present invention, the computing unit is implemented in the mobile terminal device 10. For this embodiment, the communication between the mobile terminal and the computing unit is performed in the mobile terminal 10.

[0050] Figure 3 A schematic representation of a possible display is shown, in particular a display on a touch-sensitive screen 24, for example on a mobile phone or tablet. In the image (display) 26, the situation before step A) of the method is shown by way of example, in which parameters about the material used, the material thickness used and the process used have been received. This can be achieved, for example, via a communication interface, by an input by a user or by reception. At the same time, it can be seen in the image 26 that the camera is ready to record.

[0051] In image (display) 28, the product of step A) of the method is shown by way of example, i.e., the image recorded by the camera is shown. In image (display) 30, the product of step B) of the method is shown by way of example, i.e., the result of the segmentation is shown in such a way that the user can recognize in which region of the image the method has located the laser cut edge.

[0052] The product of steps C) and D) of the method is shown by way of example in the image (display) 32, i.e. the result of the image quality recognition is output to the user. In the example presented, it is shown in different gray tones, in practice preferably in different colors: how sharp the image is in different image areas, and whether this image sharpness is sufficient or whether it is recommended to take another picture.

[0053] Figure 4An example of the product of segmentation is shown. Images 34.1, 34.2 and 34.3 are exemplary test images of laser cut edges. Images 36.1, 36.2, 36.3 are respectively assigned reference truth values, i.e. "correct" desired segmentations, in which the cut edges are uniquely and clearly identified. Images 38.1, 38.2, 38.3 show the results of the first segmentation algorithm, and images 40.1, 40.2, 40.3 show the results of the second segmentation algorithm. It can be seen in images 38.1, 38.2, 38.3, 40.1, 40.2, 40.3 that each of the two algorithms has identified the section of interest with the laser cut edge - this can be seen in the lighter colored areas in the corresponding images 38.1, 38.2, 38.3, 40.1, 40.2, 40.3.

[0054] Figure 5 A system 100 for machining a workpiece 12 using a laser cutting machine 14 is shown by way of example.

[0055] At least one material parameter 118, which characterizes in particular the workpiece material used and / or its thickness, at least one machine parameter 120, and preferably at least one desired cut edge quality feature are input via the input unit 116. The material parameter characterizes in particular the workpiece material used and / or its thickness, and the machine parameter characterizes in particular the laser cutting machine 14 used. Furthermore, at least one desired method parameter, in particular laser power, depth of focus, feed speed and / or gas flow, can be input via the input unit 116.

[0056] The system 100 can be designed to autonomously determine the material parameters 118 used by measuring, in particular by weight measurement and comparison with stored material property data, and in particular to autonomously determine the workpiece dimensions of the workpiece 12 by measuring the workpiece 12. Furthermore, the system 100 can be designed to autonomously determine the laser cutting machine 14 used. This design reduces the input expenditure in the preparation phase for processing the workpiece by the laser cutting machine 14.

[0057] A computing unit 122 having a method parameter algorithm 124 reads in the input material parameters 118 and machine parameters 120 and in particular the desired cut edge quality characteristics and / or the desired method parameters and stores the information in a data set 126 in a database 128. Based on the input information, the method parameter algorithm 124 determines improved, preferably optimal and / or method parameters required to achieve the desired cut edge quality characteristics.

[0058] For this purpose, the method parameter algorithm has a data aggregation routine 127 . The method parameter algorithm is preferably designed in the form of a data aggregation routine 127 .

[0059] The method parameters thus determined are output via a display 130 and / or transmitted to a controller 132 for controlling the laser cutting machine 14. After the improved, preferably optimal method parameters have been published, the user can either approve the method parameter recommendations for the application or perform other adjustments to the method parameters and start the method run. Subsequently, the workpiece 12 is processed by the laser cutting machine 14 and according to the predefined method parameters. The method parameters that are decisive for processing the workpiece 12 by the laser cutting machine 14 and the method parameters suggested by the method parameter algorithm 124 are added to the data set 126 for the workpiece processing.

[0060] In order to enable a clear assignment of the workpiece 12 to the data set 126, the workpiece 12 can be marked manually or automatically within the process, in particular by laser marking, preferably by laser marking of a QR code, in particular during the laser cutting process. Such marking also has the advantage that the workpiece can be automatically assigned in another process flow by simply scanning the workpiece 12. If a corresponding marking of the workpiece 12 is carried out, the corresponding information is added to the data set 126 of the workpiece processing.

[0061] After the workpiece 12 has been processed, the quality of the resulting cutting edge, in particular different cutting edge quality features, is determined. This can be achieved using a method having steps A), B), C) and D). By means of step D), it is displayed to the user whether the quality of the optical recording of the cutting edge 16 is good enough or whether the recording process should be repeated. Image data representing the quality of the cutting edge 16 are added to the data set 26 of the workpiece processing. Preferably, after the workpiece 12 has been processed by the laser cutting machine 14, an objective determination of the cutting edge quality features is performed in step E) of the method. In this case, based on the image data, the quality of the cutting edge 16 is automatically determined by means of image processing methods in step E). Preferably, the measurement results are added to the corresponding data set 126 of the workpiece processing.

[0062] The database 128 is designed to store all data sets 26 for machining a workpiece. The database 128 thus forms the basis for changing, in particular improving, preferably optimizing, the method parameter algorithm 124 .

[0063] Preferably, the already machined workpieces 12 are evaluated with regard to their cut edge quality features and used to improve the method with regard to machining subsequent workpieces 12 .

[0064] The instantaneous method parameters 136 measured during laser cutting, for example by sensors, can also be stored in the database 128 and can be supplemented with the data set 126 of the current workpiece processing. This offers the advantage, in particular, that fluctuations in the method parameters can be determined during laser cutting and taken into account when assessing the cut edge quality. This makes it possible to achieve a particularly high degree of predictability with regard to the cut edge quality and the machine state.

[0065] Based on the data sets 126 stored in the database 128, a change, in particular an improvement, preferably an optimization, can be achieved for at least one, in particular all, data aggregation routines 127 of the method parameter algorithm 124. In this case, data sets 126 of different users of the system 100 can also be used together in order to determine an improved, in particular optimal, relationship between input parameters and output parameters of at least one data aggregation routine 127.

[0066] Figure 6 A schematic illustration of an embodiment of a control system 600 is shown, which is suitable for implementing instructions for implementing one or more aspects of the method in an apparatus. The components described should be understood as examples and do not limit the scope of use or functionality of hardware, software, firmware, embedded logic components, or combinations of multiple such components for implementing a particular embodiment of the invention. Some or all of the components presented can be part of the control system 600.

[0067] In this embodiment, the control system 600 includes at least one processor 601, such as a central processing unit (CPU, DSP) or a programmable logic module (PLD, FPGA). The control system 600 can also include a working memory 603 and a data memory 608, which communicate with each other and with other components through a bus 640. The bus 640 can also connect a display 632, one or more input devices 633, one or more output devices 634, one or more storage devices 635, and various storage media 636 to each other and to one or more devices of the processor 601, the working memory 603, and the data memory 608. All of these elements can be coupled to the bus 640 directly or through one or more interfaces 622, 623, 624, 625, 626 or adapters.

[0068] The control system 600 can have any suitable physical form, including but not limited to one or more integrated circuits (ICs), printed circuit boards (PCBs), mobile handheld devices, laptops or notebooks, distributed computer systems, computing grids, or servers. The processor 601 or central processing unit (CPU) may include a cache memory unit 602 for local temporary storage of commands, data, or processor addresses. The processor 601 is configured to support the execution of instructions stored on at least one storage medium.

[0069] The working memory 603 and the data memory 608 can each be configured as a machine-readable, in particular non-volatile, in particular tangible storage medium. They can have various components, including but not limited to: random access storage components (Direktzugriffsspeicherkomponent), such as RAM 604, in particular static RAM "SRAM", dynamic RAM "DRAM", etc., read-only components, such as ROM 605, and any combination thereof. ROM 605 can also be used to communicate data and instructions unidirectionally to the one or more processors 601, and RAM 604 can also be used to communicate data and instructions bidirectionally to the one or more processors 601.

[0070] The memories 603, 608 and storage media can be bidirectionally connected to the one or more processors 601, selectively connected through the memory control unit 607. Both memories 608, 603 can be used to store the operating system 609, programs 610, data 611, applications 612, application programs, etc. Usually, but not always, the memories 603, 608 are supported by secondary storage media (e.g., hard disk) that is slower than the main memory (e.g., memory 603). For example, the memories 603, 608 can also include magnetic, optical, transistorized, solid-state storage devices (e.g., flash memory-based systems) or a combination of multiple arbitrary elements of the above elements.

[0071] The bus 640 connects multiple subsystems. The bus 640 can be any one of multiple types of bus structures, such as a memory bus, a memory controller, a peripheral bus, a local bus, and all combinations thereof when multiple bus architectures are used. Information and data can also be displayed through the display 632. Examples of the display 632 include, but are not limited to: a liquid crystal display (LCD), an organic liquid crystal display (OLED), a cathode ray tube (CRT), a plasma display, and any combination thereof. The display 632 can be connected to the processor 601, the memory 603, 608, the input device 633, and other components through the bus 640.

[0072] The bus 640 can connect all the aforementioned components with the network interface 620 to an external network 630. The external network can be, for example, a LAN, a WLAN, etc. It can construct connections with other storage media, servers, printers, and displays. It can have access to telecommunications equipment and the Internet. The bus 640 can connect all the aforementioned components to a graphics controller 621 and a graphics interface 622, which can be connected to at least one input device 633.

[0073] The bus 640 can connect all the aforementioned components with the input interface 623, which can be connected with at least one input device 633. The input device can include, for example, a keyboard, a keyboard, a mouse, a pen, a touch screen, and the like.

[0074] The bus 640 can connect all the aforementioned components to the output interface 624, which can be connected to at least one output device 634. The output device 634 can have a light display, LED display, information display, such as LCD, OLED, etc., or can have an interface to such a device.

[0075] The bus 640 can connect all the aforementioned components to a memory access interface 625, which can be connected to at least one storage device 635. The bus 640 can connect all the aforementioned components to another memory access interface 626, which can be connected to at least one storage medium 636. The storage device 635 or the storage medium 636 can be, for example, a solid-state memory, a magnetic memory or an optical memory, and in particular can have a non-volatile memory. The storage medium can be separated from the control system during operation of the control system without losing data. The storage medium can be tangible, i.e., an object that exists in a material form.

[0076] The bus 640 can be implemented completely or partially via cables or lines (eg LAN, RS232, etc.) or completely or partially without cables via a radio connection or the like (eg WLAN, WIFI, Bluetooth, NFC, etc.).

[0077] The display 632, the input device 633, the output device 634, the storage device 635 and / or the storage medium 636 can be arranged outside the control system 600 or integrated therein, respectively. They can also be connected to the control system 600 by connecting to the Internet or a network.

[0078] The controller described in the present invention can be implemented in such a control system 600. The method steps described in the present invention can be at least partially performed on such a control system 600.

[0079] The computing unit described in the present invention can be implemented in such a processor 601 .

[0080] The input unit 116 described in the present invention can be implemented in such an input device 633 .

[0081] The display 24 described in the present invention can be implemented in such a display 632 .

Claims

1. A method for evaluating a laser cut edge (16) of a workpiece (12), the method comprising the following steps: A) detecting image data of the laser cutting edge (16) and its surroundings; B) segmenting the image data and identifying interesting segments of the image data, wherein: The section of interest has image data of the laser cut edge (16); C) performing image quality identification on the segment of interest; D) generating an output for a user based on the image quality identification, in which the user is informed of the result of the image quality identification, The method is characterized in that in step B), for each pixel of the image data, it is determined and stored whether the pixel represents a part of the laser cutting edge (16), wherein when segmenting the image data, the image points or pixels of the image data located on the laser cutting edge (16) are assigned to the section of interest. Wherein, if the image quality is sufficient, the method comprises the following steps: E) Evaluating the quality of the laser cut edge (16).

2. The method according to claim 1, characterized in that Step B) is carried out with the aid of a neural network.

3. The method according to any one of the preceding claims, characterized in that The image quality recognition includes definition recognition.

4. The method according to claim 3, characterized in that The sharpness identification includes transforming the image data of the section of interest into a frequency domain, and performing the sharpness identification in the frequency domain.

5. The method according to any one of the preceding claims, characterized in that Prior to step A), the laser-cut edge (16) is cut by means of a laser beam.

6. The method according to any one of the preceding claims, characterized in that Prior to step A), parameters related to the laser cut edge (16) are received, wherein the parameters include data on the type of material of the workpiece (12), on the type of cutting method and / or on the thickness of the workpiece (12).

7. The method according to claim 5, wherein: The laser-cut edge (16) is additionally cut by means of a compressed air jet.

8. A mobile terminal device (10), comprising: A camera (18) configured to capture image data of the laser cutting edge (16) together with the surroundings; in, The mobile terminal device (10) is configured to: The image data are transmitted to a computing unit and image data are received from the computing unit, wherein the received image data are segmented and a section of interest is identified in the received image data, wherein the section of interest has image data of the laser cut edge (16); wherein the received image data has a result of an image quality identification for the section of interest, The mobile terminal device has the computing unit. wherein the computing unit is configured to segment the image data and identify interesting sections of the image data, wherein the computing unit is configured to perform an image quality detection for the interesting sections, The mobile terminal device (10) has an output interface (24), which is configured to generate an output for a user, wherein the output is based on a result of the image quality identification, and wherein the user is informed of the result of the image quality identification. The invention is characterized in that the computing unit is configured to determine and store, for each pixel of the image data, whether the pixel represents a part of the laser cutting edge (16), wherein when segmenting the image data, image points or pixels of the image data located on the laser cutting edge (16) are assigned to the section of interest. Therein, if the image quality is sufficient, the quality of the laser cut edge (16) is evaluated.

9. The mobile terminal device according to claim 8, characterized in that: The mobile terminal device (10) is a mobile phone, a tablet computer or a web camera.

10. The mobile terminal device according to claim 8 or 9, characterized in that At least one interface (20, 22, 24) for receiving parameters related to the laser cutting edge (16).

11. The mobile terminal device according to claim 10, characterized in that: The at least one interface is designed as an input interface (20, 22) for a user and / or as a communication interface for communicating with a laser cutting machine (14).

12. A system for processing a workpiece (12) using a laser cutting machine (14), comprising: A laser cutting machine (14) configured to cut a laser cutting edge (16) into a workpiece (12) by means of a laser beam; as well as The mobile terminal device (10) according to any one of claims 8 to 11.

Images